summaryrefslogtreecommitdiff
path: root/info/ltx3pub/l3d007.tex
blob: 8816b1cd44c5fbb0271cd752a63179b017725de3 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
%%% ====================================================================
%%%  @LaTeX3-report { LaTeX3-L3-007
%%%  filename        = "l3d007.tex",
%%%  archived        = "ctan:/tex-archive/info/ltx3pub/",
%%%  author          = "Justin Ziegler",
%%%  doc-group       = "Project core team",
%%%  title           = "Technical report on Math Font Encoding",
%%%  version         = "2.00",
%%%  date            = "June 1, 1994",
%%%  time            = "18:52:36 GMT",
%%%  status          = "Public, official",
%%%  abstract        = "This is a report of the \LaTeX3
%%%                     Project work on math font encoding.
%%%                     ",
%%%  keywords        = "Math fonts encoding",
%%%  project-address = "LaTeX3 Project            \\
%%%                     c/o Dr. Chris Rowley      \\
%%%                     The Open University       \\
%%%                     Parsifal College          \\
%%%                     Finchley Road             \\
%%%                     London NW3 7BG, England, UK",
%%%  project-tel     = "+44 20 7794 0575",
%%%  project-FAX     = "+44 20 7556 6196",
%%%  project-email   = "latex-team@latex-project.org",
%%%  copyright       = "Copyright (C) 1994 LaTeX3 Project.
%%%                     All rights reserved.
%%%
%%%                     Permission is granted to make and distribute
%%%                     verbatim copies of this publication or of
%%%                     coherent parts from this publication provided
%%%                     this copyright notice and this permission
%%%                     notice are preserved on all copies.
%%%
%%%                     Permission is granted to copy and distribute
%%%                     translations of this publication or of
%%%                     individual items from this publication into
%%%                     another language provided that the translation
%%%                     is approved by the original copyright holders.
%%%
%%%                     No other permissions to copy or distribute this
%%%                     publication in any form are granted and in
%%%                     particular no permission to copy parts of it
%%%                     in such a way as to materially change its
%%%                     meaning.",
%%%  generalinfo     = "To subscribe to the LaTeX3 discussion list:
%%%
%%%                      Send mail to listserv@urz.uni-heidelberg.de
%%%                      with the following line as the body of the
%%%                      message (substituting your own name):
%%%
%%%                        subscribe LaTeX-L First-name Surname
%%%
%%%                     To find out about volunteer work:
%%%
%%%                      look at the document vol-task.tex which can
%%%                      be obtained electronically, see below.
%%%
%%%                     To retrieve project publications electronically:
%%%
%%%                      Project publications are available for
%%%                      retrieval by anonymous ftp from ctan hosts:
%%%                          cam.ctan.org
%%%                          dante.ctan.org
%%%                          tug.ctan.org
%%%                      in the directory /tex-archive/info/ltx3pub.
%%%
%%%                      The file ltx3pub.bib in that directory gives
%%%                      full bibliographical information including
%%%                      abstracts in BibTeX format.  A brief history
%%%                      of the project and a description of its aims
%%%                      is contained in l3d001.tex.
%%%
%%%                     If you only have access to email, and not ftp
%%%                      You may use the ftpmail service.
%%%                      Send a message just containg the word
%%%                          help
%%%                      to ftpmail@dante.ctan.org
%%%                      for more information about this service.
%%%
%%%                     For offers of financial contributions or
%%%                      contributions of computing equipment or
%%%                      software, contact the project at the above
%%%                      address, or the TeX Users Group.
%%%
%%%                     For offers of technical assistance, contact the
%%%                      project at the above address.
%%%
%%%                     For technical enquiries and suggestions, send
%%%                      e-mail to the latex-l list or contact the
%%%                      project at the above address.",
%%%  checksum        = "23322 5582 30455 207129",
%%%  docstring       = "The checksum field above contains a CRC-16
%%%                     checksum as the first value, followed by the
%%%                     equivalent of the standard UNIX wc (word
%%%                     count) utility output of lines, words, and
%%%                     characters.  This is produced by Robert
%%%                     Solovay's checksum utility.",
%%%  }
%%% ====================================================================

\NeedsTeXFormat{LaTeX2e}

\begin{filecontents}{l3d007.sty}
%%% Action macros for this document
%%%

%%% table stuff from testfont whole thing need wrapup!!!!!!

\tracinglostchars=0

\newcount\mcnt \newcount\ncnt \newcount\pcnt \newdimen\dim
%JWZ the next two counters can be changed for 128 glyphs or 256 glyphs.
\newcount\maxz \maxz=128
\newcount\maxiz \maxiz=127
\newif\ifskipping
\def\setbaselineskip{\setbox0=\hbox{\ncnt=0
\loop\char\ncnt \ifnum \ncnt<\maxiz \advance\ncnt 1 \repeat}
\baselineskip=6pt \advance\baselineskip\ht0 \advance\baselineskip\dp0 }
\def\hours{\ncnt=\time \divide\ncnt 60
  \mcnt=-\ncnt \multiply\mcnt 60 \advance\mcnt \time
  \twodigits\ncnt\twodigits\mcnt}
\def\twodigits#1{\ifnum #1<10 0\fi \number#1}
\def\startfont#1{\font\testfont=#1
%  \leftline{\scriptsize Test of #1\unskip\ on \today\ at \hours}
%  \medskip
  \testfont \setbaselineskip
  \ifdim\fontdimen6\testfont<10pt \rightskip=0pt plus 20pt
  \else\rightskip=0pt plus 2em \fi
  \spaceskip=\fontdimen2\testfont % space between words (\raggedright)
  \xspaceskip=\fontdimen2\testfont \advance\xspaceskip
  by\fontdimen7\testfont}

\def\oct#1{\hbox{\rm\'{}\kern-.2em\it#1\/\kern.05em}} % octal constant
\def\hex#1{\hbox{\rm\H{}\tt#1}} % hexadecimal constant
\def\setdigs#1"#2{\gdef\h{#2}% \h=hex prefix; \0\1=corresponding octal
 \mcnt=\ncnt \divide\mcnt by 64 \xdef\0{\the\mcnt}%
 \multiply\mcnt by-64 \advance\mcnt by\ncnt \divide\mcnt by 8
 \xdef\1{\the\mcnt}}
\def\testrow{\setbox0=\hbox{\penalty 1\def\\{\char"\h}%
 \\0\\1\\2\\3\\4\\5\\6\\7\\8\\9\\A\\B\\C\\D\\E\\F%
 \global\pcnt=\lastpenalty}} % \pcnt=1 if none of the characters exist
\def\oddline{\cr
  \noalign{\nointerlineskip}
  \multispan{19}\hrulefill&
  \setbox0=\hbox{\lower 2.3pt\hbox{\hex{\h x}}}\smash{\box0}\cr
  \noalign{\nointerlineskip}}
\def\evenline{\loop\skippingfalse
 \ifnum\ncnt<\maxz \mcnt=\ncnt \divide\mcnt 16 \chardef\next=\mcnt
 \expandafter\setdigs\meaning\next \testrow
 \ifnum\pcnt=1 \skippingtrue \fi\fi
 \ifskipping \global\advance\ncnt 16 \repeat
 \ifnum\ncnt=\maxz \let\next=\endchart\else\let\next=\morechart\fi
 \next}
\def\morechart{\cr\noalign{\hrule\penalty5000}
 \chartline \oddline \mcnt=\1 \advance\mcnt 1 \xdef\1{\the\mcnt}
 \chartline \evenline}
\def\chartline{&\oct{\0\1x}&&\:&&\:&&\:&&\:&&\:&&\:&&\:&&\:&&}
\def\chartstrut{\lower4.5pt\vbox to14pt{}}
\def\dofonttable#1{\startfont{#1}$$\global\ncnt=0
  \halign to\hsize\bgroup
    \chartstrut##\tabskip0pt plus10pt&
    &\hfil##\hfil&\vrule##\cr
    \lower6.5pt\null
    &&&\oct0&&\oct1&&\oct2&&\oct3&&\oct4&&\oct5&&\oct6&&\oct7&\evenline}
\def\endchart{\cr\noalign{\hrule}
  \raise11.5pt\null&&&\hex 8&&\hex 9&&\hex A&&\hex B&
  &\hex C&&\hex D&&\hex E&&\hex F&\cr\egroup$$\par}
\def\:{\setbox0=\hbox{\char\ncnt}%
  \ifdim\ht0>7.5pt\reposition
  \else\ifdim\dp0>2.5pt\reposition\fi\fi
  \box0\global\advance\ncnt 1 }
\def\reposition{\setbox0=\vbox{\kern2pt\box0}\dim=\dp0
  \advance\dim 2pt \dp0=\dim}
\def\centerlargechars{
  \def\reposition{\setbox0=\hbox{$\vcenter{\kern2pt\box0\kern2pt}$}}}

%%% use of code example

%\begin{figure}
%\dofonttable{msbm7}
%\caption{Output of the \texprog{nfssfont.tex} program for the font
%         \texttt{msbm7}}
%\label{fig:testfontmsbm}
%\end{figure}


\newcommand\textem[1]{\emph{#1}}

\newcommand{\note}[1]{%
  \par \vskip .25cm \textbf{Note:\ }#1 \par \vskip .25cm}

\chardef\bslash=`\\
%    Command name
\newcommand{\cn}[1]{$\mathchar"026E${\tt#1}}
%    File name
\newcommand{\fn}[1]{{\tt#1}}

\def\jzitem#1{\item{\bf #1\\}}

\def\thisisnew{\marginpar {\textbf
                 {\textsl {\begin{raggedright}This part is
                                 new.\end{raggedright}}}}}

\def\changedabit{\marginpar {\textbf
        {\textsl {\begin{raggedright}Changed this a little
                bit.
                        \end {raggedright}}}}}

\def\seeappendix{\marginpar {\textbf
         {\textsl {\begin{raggedright}Here is an
                 appendix reference.
                 \end {raggedright}}}}}

% don't warn about bold typewriter
\DeclareFontShape{OT1}{cmtt}{bx}{n}{ <-> ssub * cmtt/m/n }{}
% \DeclareFontShape {T1}{cmtt}{bx}{n}{ <-> ssub * cmtt/m/n }{}
\end{filecontents}



\begin{filecontents}{l3d007a.tex}
\section{Introduction}
  This document aims to put on paper what could be the backbone or the
  skeleton of a new math encoding for \TeX. This is not the complete
  description of an encoding, but a sort of grid, or global picture of
  what things could look like. This document refers to many glyph groups
  defined in another document called ``Towards a list of math glyphs''.
  Same author.





\section{A few definitions}

\begin{description}

\item[An ``encoding table''.] This conveys the
traditional meaning of an encoding. That is to say a set of 256
glyphs in a given order. The expression ``encoding table'' is usually
abbreviated to ``encoding''.

\item[A ``slot''.] It is the usual word used for referring
to a position in an encoding, that can contain a glyph. It is usually
an integer between 0 and 255. A slot is
certainly not a family, nor anything to do with it.

\item[A ``math kernel''.] This terminology is used to
specify the fonts that are necessary for the math facility to work
as it is described in most \TeX\ documentation\footnote{documentation
on LA\TeX, AMS\TeX, etc, also fits in here.}.
In DEK's implementation the math kernel consists of the
families from 0 to 3. On top of the kernel, many other fonts, with
whatever encoding is available, could be optionally loaded and used.

\item[A ``math encoding''.] It is considered here as a whole; not
just one 256-glyph encoding table, but a set of encoding tables.
This concept will be referd to as ``M-encoding''.

\item[The ``core symbols''.] They are made of two
groups. The group of symbols that must live with the default alphabet
for kerning reasons, and the group of symbols that must live with the
default alphabet for design reasons.

\item[The ``default alphabet''.] It is the alphabet that is
used  when a user types \texttt{\$abc\$}. In the present encoding that
produces $abc$.

\item[``Glyph compatibility''.] Two encodings are glyph compatible
when they contain the same glyphs but not in the same positions. As well
as containing the same glyphs, it must also be possible for the
corresponding fonts to contain the same sidebearings and the same
kerning and ligaturing information.

\end{description}





\section{Global policy}
  \subsection{Text in math mode}
   It is generally agreed that for best quality documents, if one wants
   to put text in a math formulae, one should switch back into text
   mode, using something like the \cn{text} macro in the AMS
   package. None of the encodings can be expected to support bad
   usage. In a math encoded font, the letters are not kerned in the same
   way as in a text encoded font, and there are no letter ligatures,
   because they are not needed.

  \subsection{A global rule for boldface}
   It has been decided not to mix ligth face and bold face symbols in
   the same encoding, but to generate a separate boldface version of all
   lightface math encoded fonts when necessary.  This does not make it
   impossible to mix the two: either one can use the
   \texttt{\string\boldsymbol} approach, or one can load an extra bold
   face font in a given family, and have it directly and permanently
   accessible.

  \subsection{Sans serif and typewriter fonts}
  Extra fonts could be designed in sans serif, or in typewriter using
  some of the proposed new encodings. Another solution is to load the
  Cork encoded sans serif fonts (or typewriter fonts) in free
  families. In either case, the new math encoding will not have any
  slots containing specifically sans serif or typewriter glyphs.

  \subsection{Concerning the Euler shapes}
  Euler shapes could be a good example implementation of the new math
  encoding. Thus no Euler glyphs will be included in the new
  encoding. However, it may be useful to use the Euler Fraktur for a
  first implementation example, if the new encoding includes a Fraktur
  or old  German alphabet.


\section{Concerning Cyrillic letters}
  These would be available, but not as part of the math encoding. They
  would be loaded as an extra family, with whatever encoding exists,
  together with suitable \cn{mathchardef}s.

  \subsection{Compatibility with other typesetting systems}
  \subsubsection{Grouping all \TeX\ specific glyphs in one font
  encoding}
  The present \texttt{cmex} font contains glyphs that cannot be used
  by other typesetting systems, because they are set in a strange way.

  The present \texttt{cmsy} font contains one glyph that is set in a
  strange way --- the radical sign, and thus makes that whole font
  unusable for the outer world. It would be a good idea to make sure
  that this does not happen again.

  The \LaTeX 3 project is hoping to set a new standard, that will not
  only be used by \TeX, but by all systems that typeset mathematical
  formulae. If every thing goes according to plan, in the next few years
  many math fonts will exist, for many different systems, and they will
  all use the same encoding.
  The fact that they all use the same encoding means that it will be
  very easy to exchange fonts from one system to another. So one day a
  \TeX\ user will be able to take a math font used by Microsoft Word,
  and convert it easily in order to use it with \TeX.

  If \TeX\ specific glyphs are grouped in one font, there will only be
  one problematic font. As it happens, all \TeX\ specific glyphs are
  more or less geometric, so they could be used with more than one
  math font.

  On the other hand, if \TeX\ specific glyphs are spread around in
  many fonts, then many ``imported'' fonts will not be usable by \TeX\
  without major messing about.

  Concerning this problem, the real question is: is the \LaTeX 3 project
  setting a real standard for the next few years, or just
  making another \TeX\ math font encoding?
  If the answer is: ``the \LaTeX 3 project is setting a real standard
  for the next few years'', then \TeX\ specific glyphs must be grouped
  in one font. If that is really not possible,
  then one can maybe consider putting them in two fonts.

  If a font designer designs a math font for adobe, the work necessary
  for adapting his font to the \TeX\ world should be reduced as much as
  possible.
  Otherwise nobody will provide any fonts for \TeX.

  The \TeX\ specific glyphs that are concerned here could be visually
  compatible with many math fonts.

  Which are the \TeX\ specific glyphs? So far:
  \begin{itemize}
  \item The delimiters.
  \item The large and small `bigops'.
  \item The radicals.
  \end{itemize}


 \subsubsection{The space issue}
  To enable easier font exchange between the \TeX\ world and the rest of
  the world, the new math encoding will have a space in position 32
  (decimal) of every encoding table, if it is possible.


  \subsection{General document compatibility}
  It is not worth beeing totaly compatible. A lot of glyph positions
  will change, thus direct \cn{mathchardef}s will not always work.
        Documented names from AMSLA\TeX\ (this includes names from
        LA\TeX, \TeX, and AMS\TeX, and LAMS\TeX) ought to be kept.


  \subsection{Grouping all Plain and \LaTeX\ glyphs in 4 fonts}
  The main reason for this is compatibility. Let us consider a user
  that has typed a document with the present math encoding and has
  saturated the families for this document. If the new math encoding
  does not guarantee Plain and \LaTeX\ glyph compatibility with a
  minimum of 4 fonts, then that document cannot run with the new math
  encoding: not enough families. Thus one should make the first four
  font encodings of the new math encoding glyph compatible with
  the group made by: cmr, cmmi, cmex, cmsy.

  \subsection{Grouping all AMS\TeX\ and AMS\LaTeX\ glyphs in less than
    6 fonts} The main reason for this is compatibility. Let us consider
  a user that has typed a document with the existant AMS\TeX\ or
  \LaTeX\ package, and has saturated the families for this document.
  If the new math encoding does not guarantee AMS\LaTeX\ and AMS\TeX\
  glyph compatibility with less than 6 fonts, then that document cannot
  run with the new math encoding: not enough families. The first 6
  font encodings must be one way glyph compatible with the fonts
  provided in the AMS package.

  \subsection{Replacing \texttt {cmex}}
  The math font group has decided that the encoding due to replace the
  \texttt {cmex} encoding will be designed in such a way that the
  corresponding font can be loaded in three sizes or in one size.

  Loading such a font in three sizes produces better typesetting. But
  the resulting page and line-breaks will not be the same as
  when the font was loaded in one size only. Some people will not like
  that change, in their old documents. But for new documents loading
  the extensibles font in three sizes will be better.

  \subsection{Accents in maths}
  There seems to be an agreement that math accents should not change
  with the font or style of the letter. But it is not a problem to keep
  the accents that are already in plain \TeX\ (and the \fn{cm} fonts) in
  the same position as in the T1 encoding. That will allow them to be
  variable family; thus any T1 encoding could be loaded, and its
  accents used. Since there will be millions of documents using
  bold hats this possibility must be preserved even if by default
  all math accents are non-variable family.


\section{The base: a Cork encoded text font}

Main use: things like \texttt{\string\log}. This would generally be a
Latin font.

If it is a Latin upright font, it would probably also be used by
physicists (and chemists) for operators, and more generally whenever
upright letters are needed.

\textbf{Separating this set from the rest} enables the user to decide
how `log' and `sin', etc. should be typeset. Thus the multiletter
operators can be compatible with the text font, or with the rest of
the math glyphs, or even set in yet another font.

Math mode should not be used for setting text phrases in mathematical
material.  For example:
 \begin{verbatim}
$$x=y \quad {\rm is\ a\ direct\ consequence\ of} \quad y=x$$
\end{verbatim}
would be better input as:
\begin{verbatim}
$$x=y \quad\hbox{\rm is a direct consequence of}\quad y=x$$
\end{verbatim}
or, better still:
\begin{verbatim}
$$x=y \quad\text{is a direct consequence of}\quad y=x$$
\end{verbatim}
where \verb|\text| is a macro which sets its argument in horizontal
mode.  This avoids unpleasant surprises such as:
\begin{verbatim}
$$X \quad{\rm is\ a\ sub-object\ of}\quad Y$$
\end{verbatim}
However, for upward compatibility with existing plain \TeX\
documents, it will still be possible to set text phrases in math mode,
as long as they only contain $\langle\mbox{character}\rangle$s of type
Variable.








\section{The ``text symbol'' encoding: the TS encoding}
Here one would put the old style numerals, and most of what
is comming out of the present math encoding. Other symbols could
be added. The Text Symbol encoding is definetly \textem{not part of
  the math kernel.} But since it will contain symbols that
previously were accessed via the math fonts, its
encoding must be supplied. This font will not be loaded in a family. It
will just be loaded as a normal text font\footnote{If some users really
feel the need to load it in a math family, they can.}.


\section{The core: the MC encoding (263)}
Counting: 1,10,1 , 54,5 , 124,14 , 12,24,9 ,9= 263 glyphs

The accents are no longer here. They had no real reason to be here.
Most of them are geometrics anyway. But they do have reasons to be
elsewhere. One of the main consequences of taking these accents out
is that the core can be made more coherent, and more complete.  The MC
encoding would contain:
  \begin{itemize}
        \itemsep -0.1cm

  \item The skewchar in position 0: 1
  \item The core digits: 10
  \item The space character in position 32: 1

  \item The core Latin alphabet, which is the default alphabet, in
    uppercase, and lowercase, together with the dotless i and j: 54

  \item The Latin friends: 5

  \item All the Greek material: 124

  \item The Greek friends, next to the Greek: 14

  \item The core symbols for kerning reasons (punctuation and
    delimiters): 12

  \item The core symbols for design reasons: 24

    More for kerning reasons:
  \item The basic geometric delimiters: 9

    Some new glyphs:
  \item New basic delimiters: 9

  \end{itemize}

Sacrifices can be made in the greek material, and in the core symbols
for design reasons.


\section{The MX encoding: 243}

Count up: 1 1   78 8   10   24 6  16   7 7   26   47 = 230

The usual extensible characters, together with some new ones, would live
here. Here any characters that have strange \TeX\ features, like big
descenders are included, thus grouping glyphs that are not compatible
with the outer world.

For various reasons discussed in ``Replacing \texttt {cmex}?'' (Same
author), the math font group has decided that the MX encoding will be
designed in a way such that when it is loaded in one size (like in the
present \TeX), every thing works OK, and the user can still have
access to the new symbols. However, the MX encoding will produce better
quality typesetting when loaded in three sizes.

Detailed contents of MX:
  \begin{itemize}
        \itemsep -0.1cm

  \item Maybe a skewchar: 1

    The space is questionable here, because MX will not be usable by
    other typesetting systems, see comment:
  \item Maybe a space: 1

  \item Big and extensible \TeX\ delimiters from cmex: 78

    Any characters that have strange \TeX\ features like big descenders:
  \item The radicals: 8

    Technically the following can come out, but then they must go in
    MSP.
    To make this possible one could take the bar accent out of MSP.
  \item Horizontal curly braces: 10

  \item All existant big and small ``bigops'' except the integrals: 24

  \item The existant Plain vertical extensible arrows: 6


    One has to limit the number of wide accents, otherwise there is not
    enough place.
  \item The wide tildes, and the wide hats: 16

  New glyphs:

    It is a little bothering that the following will be separated from
    their small versions, but there is nothing much that can be done
                about it:
  \item The big ``big integral'' family: 7
  \item The small ``big integral'' family: 7

  \item The new big and small `bigops': 26
  \item New multisized and extensible delimiters: 47


  \end{itemize}

Note: all integral glyphs must be kerned with themselves, so that two
integrals following each other can be kerned easily with a little
care. Same for various other glyphs.


\section{The math symbol `privilege' font ``MSP'': 250}
 Count up: 1,1, 54 , 18,7,3 , 23 , 8,4,2,4,4 , 20,7, 10 , 8,8,12 ,  14 ,
 6,4,5,4,5 , 8,2  , 16 = 250
  \begin{itemize}
        \itemsep -0.1cm

  \item A skewchar in position 0: 1
  \item A spacechar in position 32: 1

  \item The script/calligraphic Latin letter set:54

    According to AMS statistics, the script/cal are
    used more often than the Blackboard bold.

    Having the script/calligraphic here also achieves backward
    compatibility with the existing \TeX, without loading MS1 and MS2.

  \item The basic accents: 18   %  all basic accents should stay here
  \item The double accents: 7   %  could go elsewhere
  \item The underaccents: 3     %  could go elsewhere

    The following must stay here:
  \item The ``Basic symbols'' group: 23

    The next 5 are needed for compatibility with plain:
  \item The ``Greater than plain'' group: 8
  \item The ``Subset plain'' group: 4
  \item The ``In / ni plain'' group: 2
  \item The ``Sqsubset plain \& ams'' group: 4
  \item The ``Succ without sim plain'' group: 4

  \item The ``Small binops plain'' group: 20
  \item The small ints: 7
    These should probably live with the other `succ' members for design
    reasons:
  \item The ``Succ without sim ams'' group: 10

    The next three make a homogenous group, and must live with sim.
    Sim itself must live here because of compatibility with Plain:
  \item The ``Greater than with sim'' group: 8
  \item The ``Succ with sim'' group: 8
  \item The ``Sim'' group: 12

    The arrows, for compatibility, (improved a little though):
  \item ``Plain horizontal arrows'': 14 or 10
  \item ``Plain vertical arrows'' : 6 \\
    Does not include the extensible arrows.
    The latter are in MX as before.
  \item ``Plain oblique arrows'': 4\\
    Also called ``Plain other arrows''.
  \item ``Latex arrows'': 5
  \item Plain miscellaneous geometric symbols: 4

    Extras ---~these are new glyphs~--- to improve a little what \TeX\
    can already do:
  \item Extra arrows for use with plain: 5

    The ``lasy'' triangles are included in the AMS fonts and thus are
                included in the following group:
  \item AMS left-right open triangles: 8

    Should live with the ``Plain oblique arrows'':
  \item ``AMS obliques'':2

    Some new glyphs: Some of this could come out.
  \item Wide accents bar: 8

\end{itemize}







\section{The MS1, MS2, Math-Symbol encodings}
Each of these encodings will contain a set of Latin letters, like for
instance fraktur or blackboard bold, in uppercase or lowercase or
both. In some cases a place should be reserved for a set of matching
numbers too (i.e. Open). The rest would be filled up with symbols.
An MS$_i$ encoding is needed for:

\begin{itemize}
\item An extra script/calligraphic, (see below comment on script and
  calligraphic) the default caligraphic is in the MSP encoding.
\item Open + (Arrows or relations) + other geometrics.
\item Old german,
\end{itemize}


\textbf{Note:} Barbara Beeton writes ``Regarding script vs.
calligraphic, I do understand the difference; however, at AMS I believe
we only very rarely get a request to use both styles in the same paper.

For that there are two possibilities:
\begin{enumerate}
        \item designing one encoding table where the positions A-Z (and
        probably  a-z and 0-9 even if they are not all filled) are
        supposed to contain a  ``calligraphy/script'' set of characters.
        Then there would be  instances of that encoding that would
        contain script chars and others  that would contain calligraphic
        chars.  Suppose our standard would say that this encoding is to
        be used as family 4.  A designer would then choose  one such
        font with this encoding for family 4 (thereby effectively
        deciding  what \texttt{\string\cal} and a lot of other symbols
        look like (the ones whose \cn {mathchardef} points into family
        4)). For those who in addition would like to use another
        script/call math alphabet: they can then just allocate one of
        the free families. Access to this would then be trivial.

        \item Having two different encodings; one for cal, and one for
        script. The remaining symbols in both encodings would be
        different too. Thus both encodings would need to be part of the
        standard suite of math encoding tables.

        Which solution is preferable depends a bit on the number of
        symbols that ought to go in the standard.''
\end{enumerate}

Also J\"org Knappen writes: ``I strongly support having two different
encodings, one for cal and one for script. If users have the choice
between cal and script, they prefer script (at least in
Mainz\footnote{Maybe Americans prefer it the other way
  round.}). However, the old calligraphic alphabet still
needs to be supported for compatibility reasons.''



\section{The MS1 encoding: 232}
Count up: 1 1 54 10  32  36 30 12 10 21 10 15= 232
  \begin{enumerate}
        \itemsep -0.1cm

    \item A skewchar in position 0: 1
    \item A spacechar in position 32: 1
    \item The BBB alphabet uppercase and lowercase: 54
    \item The BBB digits: 10
    \item The last WIDE ACCENTS: arc, back-to-front vector, and
    double-sided  vector, normal vector: 32

      For AMS inclusion:
    \item The ``AMS horizontal arrows'' group: 22
    \item The AMS other arrows group: 12
    \item The ``Greater than AMS'' group: 30
    \item The ``Subset AMS'' group: 12
    \item ``AMS Equals friends'': 10
    \item ``AMS Miscellaneous geometric symbols'': 21
    \item ``AMS Vdash group'': 10
    \item ``AMS boxes and friends: 15

      For fun if there is place to spare:
    \item Alan's arrow construction set: ?

  \end{enumerate}








\section{Other requested typefaces}
  \begin{itemize}

  \item In general, users may want MC fonts in arbitrary styles (bold
    sans serif MC for instance) in order to get the Greek letters in
    their favourite styles.

  \item A ``text-like'' italic or slanted for computer science
    identifier names and the like.  This would be Cork encoded and
    optionally loaded.

  \item A ``bold upright'' for use as variables -- e.g. vectors in
    physics notation rather than the arrow over an italic letter. This
    would be Cork encoded, and optionally loaded or accessed via the
    \texttt{\string\boldsymbol} concept in which case no family would be
    required.

  \item Bold italic for use as variables: either optionally loaded as
    a second font with MC or cork encoding (using only variable family
    symbols) or accessed via something like
    \texttt{\string\boldsymbol}.

  \item Bold Old german (occasional) suggested
    \texttt{\string\boldsymbol} approach.

  \item Bold script (occasional) suggested \texttt{\string\boldsymbol}
    approach.

  \item Sans serif lightface (occasional): optionally loaded cork
    encoded font.

  \item Sans serif boldface (occasional): optionally loaded cork
    encoded font.

  \item Bold symbols: either \texttt{\string\boldsymbol} or optionally
    loaded in remaining slots.

  \item Ultra bold symbols: either \texttt{\string\boldsymbol} or
    optionally loaded in remaining slots.

  \item An MC-encoded bold font containing upright bold Latin glyphs,
    plus bold upright and bold slanted Greek.  This would contain all
    of the most commonly requested bold glyphs in one font (rather
    than many more).


  \item A cyrillic alphabet. Loaded as an extra family, or in text.

  \end{itemize}





\section{Summarising the families used by the proposed YAASP M-encoding}
\begin{enumerate}
\item Family $0$: A Cork encoded upright text font.

\item Family $2$: An MC encoded font containing the default Latin and
  Greek italic+upright, and core symbols...

\item Family $1$: An MSP encoded font containing cal/script and
  accents...

\item Family $3$: An MX encoded font including all extensible glyphs,
  and double sized operators...

\item Family $y$: An MS$1$ encoded symbol font for Open, and arrows
  or relations.

\item Family $z$: An MS$2$ encoded symbol font for Old German.
\end{enumerate}

\textbf{Comments:}
\begin{itemize}
        \item This leaves 10 families free for anything else, (like bold
        or sans...) and makes many symbols available.

        \item The first four encoding tables put together give total
        \TeX,  La\TeX\ compatibility.

        \item The first six give total \TeX, \LaTeX, AMS\TeX,
        AMS\LaTeX, LAMS\TeX\ compatibility.

        \item The six put together: do wonders, using no more font
        families than  the present AMS\TeX.

\end{itemize}

\section{Discussion}

\subsection{Advantages}

For MC: A big advantage here, is kerning. In this encoding kerning
is possible between the Latin default alphabet, and both italic and
upright Greek alphabets. This is necessary for compatibility, and for
tidyness. On top of this both letter sets (in actual fact there are
three) can be kerned with the core symbols that are in the MC
encoding. This last point is the most important, and gives new and
better automatic math spacing. (For compatibility reasons,
the Greek italic must be kerned with the period, the comma, and the
slash.)

The bold version of the MC encoding gives the user access to a lot of
bold letter sets in one go. The global family consumption is therefore
largely reduced: 1 bold font instead of 2 or 3.

Taking the accents away from the letters, means that the accents do
not change when the text face changes, i.e. bold letters and medium
letters get the same accents.

One can get more than compatibility with plain \TeX\ only using 4
families (the same number as standard \TeX\ currently uses).

One can get more than compatibility with AMS\TeX\ using 6 families.
This is less or equal than the number of families used by AMS\TeX.

The calligraphic alphabet is more used than the open, so putting it
with the accents is a step towards grouping most used glyphs together.

This proposal gives a little room in the MC for free spaces, and good
core material.

With the MSP encoding concept, the MSi encodings can really be
considered as (optional) extensions. Thus somebody who knows he
does not need the arrow kit and the Blackboard bold letter set does not
have to load them. Same for Fraktur.

All the \TeX\ specific glyphs are grouped in MX. Thus all the other
fonts could be used by other typesetting systems.

Using the Cork encoded font in family 0 for things like \cn{log} and
\cn{sin}, means that the Greek users can replace it by a Greek font.
(Apparently Greek mathematicians set these function names
using a Greek alphabet).



\subsection{Disadvantages}

If there is not enough space for all the required symbols,
one can make an MS3 encoding that would contain the other
version of script/cal, together with extra symbols.


\subsection{Comments}

In this proposal the core is really made of two fonts, and the kernel is
made of four.

Comments from Alan about the space slot in MX:
\begin{quote}
  ``MX will be used by \TeX, and the dvi drivers may be outputting to a
  device that does not accept anything but a space in position 32.  So
  if you don't include a space here, then the MX-encoded fonts have to
  be split into two device fonts by the drivers.''
\end{quote}

Comment from Alan about the Cork encoded font:
\begin{quote}
  ``I think it would be good to specify that this is family 0, for
  compatiblity with current \TeX\ documents containing explicit
 \cn{fam 0}
  (naughty them!) and in order to have filled up slots 0 to 3 rather
  than leaving a gap in family 0.''
\end{quote}
\end{filecontents}

\begin{filecontents}{l3d007b.tex}
\section{Introduction}
  The goal of this document is to try and list the future contents of
  the new math encoding. The glyphs have been divided into groups and
  subgroups.

  \textem {This is a draft and a workbench document.  Some of it is out
  of date, and usually the comments are not for the public.  There are
  lots of spelling mistakes, I know! The reader should be indulgent.}



%something added by bb for euler:
%begin{enumerate}
%\item Cursive (normal) capital letters. not often used.
%\item Cursive (normal) lower-case letters.
%\item Script upper-case.
%\item (Some) script lower-case.
%\item Fraktur (old german) upper-case.
%\item Fraktur (old german) lower-case.
%\item Digits.
%\end{enumerate}
%% bb --
%[bb: I will be happy to make a proposal for this.]


\section{Extra font dimensions}
\begin{itemize}
\itemsep -0.15cm
\item The design size,
\item The default script size,
\item The default scriptscript size,
\item Suggested value for mathsurround (in MC)
\item Math\_axis (in every font)
\item Thin mu skip,
\item Med mu skip,
\item Thick mu skip,
\item Recommended rule weight
\item Baselineskip: leadingheight,
\item Baselineskip: leadingdepth
\item Suggested by JMR: the big and bigg params.
\end{itemize}






\section{Kerning}
Better kerning should be made possible in the Latin math
italic, if it is possible. Normal kerning information is put in the
\fn{.tfm}
file. But in math mode, for things to get kerned as specified in the
\texttt{.tfm} file the left atom must be of ordinary type. If the user
interface redefines everything that must be kerned as being ordinary,
old documents will start looking different, and this is not wanted. To
avoid this, the user interface could define a macro
\texttt{\string\mathkerning\{...\}} that would use the kerning specified
in the \texttt{.tfm} file, without globally making all characters
ordinary.

The following glyphs should be kerned:
\begin{itemize}
\itemsep -0.15cm
\item The spacing of [ and ( and ) and ] followed by letters should
  be adjusted.

\item  The spacing of letters followed by [ ] ( ) should be
  adjusted. This is an  important one.

\item Keep the kerning with\quad . \quad , \quad / \quad for most
  letters !
%\item Keep kerning between = and upright $\Gamma$ ! (I must have
  %dreamt this one) Maybe add kerning for other punctuation characters.
\item At least keep the kerning between d and Y,Z,j,f. Maybe add some
  others: dx, dy, d$\alpha$, d$\theta$, d$\phi$ ...

\item New kerning of the integral with itself. This would only
  be available via a \texttt{\string\mathkerning} macro (see
  previous comment).
\item Kerning of the period with itself,
\item Kerning of the centered period with itself.
\item Jan M.R. is sure that kerning is needed between
  Latin and Greek. More precise information is needed.
\end{itemize}













\section{The following should be taken out of the present math encoding}
\begin{itemize}
\itemsep -0.15cm
\item The old digits: 10.
\item The 2 paragraph signs: \P, \S.
\item The Yen sign: Y.
\item The double dagger sign \ddag.
\item The four card families: $\clubsuit, \heartsuit, \diamondsuit,
  \spadesuit$.
\item The musical signs: $\flat, \natural, \sharp$ ?????????
\item The maltese cross. (AMS)
\item The $\mathchar"017F$ seems not to be needed in maths.
\item The circled R must come out of the math symbols. (AMS)
\item The raised asterisk,
\item The two triangles in cmmi: cmmi'56 '57  ???????
\end{itemize}

All these will be put in to the ``Text symbols'' encoding, that would
come in many faces, and be text dependent.
Other glyphs could be put in the ``Text symbols'' font:
\begin{itemize}
\itemsep -0.15cm
\item More numerals,
\item The perthousand sign.
\item Maybe this is a good place for the `fraction' characters from
  adobe.
\item $<$florin$>$, $<$ellipsis$>$ etc.
\item The superior and inferior digits, and put in kerning so that
  $<$onesuperior$>$ $<$fraction$>$ $<$twoinferior$>$ produces a 1/2.
\item The single dagger finds a place here although it is in
  maths as well. This makes them two different symbols, and enables both
  to  have more specific shapes.
\item A real copyright symbol, TM (trademark) and SM (service mark).
\item An interrabang (a combination of ? and !) new. (bb)
\end{itemize}


Alan Jeffrey has worked on the `text symbol font'. Actually it is now
called the companion text font. He has written more on this topic.
``alanje@cogs.susx.ac.uk''






\section{The Greek glyphs: 124}
The following shapes must be included:
\begin{itemize}
\itemsep -0.15cm
\item All the Upper-case in upright. 24
\item All the Upper-case in italic. 24
\item All the Lower-case in upright. 24
\item All the Lower-case in italic face. 24. So far: $24\times 4=96$
\item All the variable shapes in upright. 10
\item All the variable shapes in italic. 10
\item All the special numeric letters in upright. 3
\item All the special numeric letters in italic. If lack of space
  prefer the italic shapes to the upright ones. 3.
\item Some control glyphs: 2 (probably useless)
\end{itemize}


\subsection{Variable shapes: 10}
This list comes basically from: J\"org Knappen.  They are all listed
here including the ones that are already in the \fn{cm} fonts:
\begin{enumerate}
\itemsep -0.15cm
\item Lower-case Phi,
\item Lower-case Pi,
\item Lower-case Kappa, (AMS)
\item Lower-case beta (new),
\item Lower-case Rho,
\item Lower-case Epsilon,
\item Lower-case Sigma,
\item Lower-case Theta.
\item Upper case chi (new),
\item Upper case for upsilon.
\end{enumerate}

\subsection{Extra letters for numerals: 3}
Source: J\"org Knappen.  They are all listed here including the ones
that are already in the \fn{cm} and \fn{ams} fonts. Must be given in
lower-case.
Upper-case Greek numerals exist, although extremely rare. For the
  sake of completeness one could fill them in. But they are surely not
  the hottest characters needed. (J\"org)
\begin{enumerate}
\itemsep -0.15cm
\item Qoppa (new),
\item Sampi (Sanpi?) (new) (J\"org Knappen),
\item Digamma (AMS).
\end{enumerate}


\subsection{Control glyphs: 2}
\begin{enumerate}
\itemsep -0.15cm
\item An italic control glyph, i.e. the following Greek letter is not
taken from the upright, but from the italic Greek,
\item A variable shape control glyph, i.e. the following Greek letter
is not taken from the normal set of letters, but form the variant shape
set. This will not work for all letters. Thus may not be a good idea.
\end{enumerate}

\note {From Alan about the control slots for Greek, ``Er, I'm not very
  sure about those, since they'll affect kerning.  I'd prefer to have
  the choice between italic / upright made by the document designer.
  And I'm not sure why anyone would want to get at an upper case $\xi$
  by a macro \texttt{\string\uppercasegreek}\{\texttt{\string\xi}\}!''
  --- ``True they will affect the kerning. But one could use them
  differently from what you suggested.  Although I'm not sure it is
  interesting, the ligature mechanism does not have to be visible for
  the user, i.e. he can still type \texttt{\string\Gamma}, which is
  expanded to \texttt{\string\up \string\gamma}.''}







\section{Extra Greek-like material: 14}
This group of glyphs should not really be separated from the rest of
the Greek material.
\begin{enumerate}
\itemsep -0.15cm
  \item An upright partial sign,
  \item An italic partial sign,
  \item An upright partial sign with a slanted bar, AMS \cn{eth}
  \item A \cn{thorn} WASY'151 but this one is not very good. There is a
          better one in dcmr'136. There is one is wslipa10'102 (J\"org)
  \item A barred upright lambda, ?????? (probably not J\"org)
  \item A barred italic lambda, this is preferred. (J\"org)
  \item An upright mho sign (upside down Omega),
  \item The back to front epsilon: AMS "7F \cn{backepsilon},
  \item Arabic letter dal: looks something like a back to front $c$.
  \item Hebrew letter msbm'151,
  \item Hebrew letter msbm'152,
  \item Hebrew letter msbm'153,
  \item The $\aleph$ or \cn{aleph} in position CMSY'100,
  \item The Nabla, $\nabla$ in CMSY'162
\end{enumerate}

The barred signs may be obtained by ligatures, or could be
constructed with kerning. In any case some slots for ligatures must be
left free if possible.








\section{The Latin letters: One set$=54$ glyphs}

We shall assume here that all lower case alphabets contain a dotless `i'
and a dotless `j', so that they can take accents other than a dot.
\begin{itemize}
\itemsep -0.15cm
\item The usual \fn{cmmi} italic shapes. Upper-case and lower-case.
\item The calligraphic shapes. Upper-case and lower-case.
  The lower-case shapes are presently maybe not available.
\item The script shapes. Upper-case and lower-case.
  The lower-case shapes are presently maybe not available.
\item The black board bold shapes. Upper-case and lower-case.
\item The Fraktur style. Upper-case, and lower-case.
\end{itemize}

  \subsection{The calligraphic and/or script styles}
  BB: ``How are ``calligraphic'' and ``script'' different here?  I've
  never seen what Knuth calls calligraphic and what most
  mathematicians call script (the ``curly'' style) used in the same
  context, so they are presumably not distinct from one another in
  actual usage.''

  The two should be included if there is enough space.
  Otherwise one is enought.

  \subsection{A hyphen char ?} These Latin letters are not meant for
  typesetting words. It is assumed that all multi-letter words should be
  typeset using the text fonts, not the math fonts. Thus
  the hyphen character is not needed in the math encoding.


  \subsection{Computer science and identifiers}
  It looks as though the new math encoding will not contain anything
  specially designed for computer science. Computer scientists will have
  to use \texttt {cmti*} in an extra family for long identifiers.

  \subsection{Chemists and chemical formulae} Considering the fact that
  chemists do use a lot of mathematical expressions, they need the total
  math mode as it is. On top of that they need a special mode for
  writing chemical equations. One of the particularities of this
  chemical mode would be the different placing of sub- and
  superscript. A possible implementation is something like
  \cn{EnterChemicalMode} and \cn{ExitChemicalMode}, which would
  in actual fact load a new set of fonts (or only the font in family 2),
  in order to have a different font dimensions in family 2.


\section{Latin-like material: 5}
This group should live next to the Latin letter set.
  \begin{enumerate}
        \itemsep -0.15cm
  \item An upright d. This is needed for standard mathematical
  typesetting.

  \item A horizontally barred italic h, for physicists.

  \item A slanted barred italic h, for physicists.

  \item An italic upper-case Vee with a bar, the bar is meant to be
    horizontal. \texttt{jvpurcel@vela.acs.oakland.edu}

  \item An upright upper-case Vee with a bar, the bar is meant to be
    horizontal, and extends through both sides of the Vee almost like
    a strikeout. \texttt{jvpurcel@vela.acs.oakland.edu}

  \end{enumerate}



\section{The different ways needed to write numbers}
  \begin{itemize}
        \itemsep -0.15cm
  \item The normal set of numbers in \fn{cmmi}: upright lining.

  \item The blackboard bold numbers. (Used in physics and a field of
  maths. See  Alan J. for more details.) [Note:  presently no
  satisfactory bbb numbers  seem to exist.]

  \end{itemize}



\section{Empty slots?}
Some free slots could be included, so that people can put their
ligatures in when they are trying to convert fonts coming from other
worlds.

Alan J. can give good explanations for this.


\section{Arrows}
Arrow construction should be possible. But to make sure it does not fail
when used in different sizes, every single glyph used for this purpose,
will be \textem{specifically} designed for this use.  \textem{All of
them will be in the same font table}. This does not mean that a given
construction block can't be used for different types of arrows.  This
sort of thing has to be thought of, and forecasted. These construction
blocks must not be used for any other purpose --- like for instance the
equal or minus sign.

All arrows from \fn{cm}, and from \fn{msam/msbm}, should be taken if
necessary. Maybe some others too.

  \subsection{The ``Plain horizontal arrows'' group: 14 (Alan:10)}
  The first 6 are in cmmi'050 to '055:
  \begin{enumerate}
                \itemsep -0.15cm
    \item leftharpoonup
    \item leftharpoondown
    \item rightharpoondown
    \item rightharpoonup
    \item lhook
    \item rhook

      From \fn{cmsy}:
    \item leftarrow '40
    \item Leftarrow '50
    \item leftrightarrow '44
    \item Leftrightarrow '54
    \item rightarrow '41
    \item Rightarrow '51
    \item CMSY'67 this is the \cn{mapstochar}
    \item CMSY'66 the negation sign/slash: 1
   \end{enumerate}

  \subsection{Extra arrows for use with plain arrows: 5 (Alan 5)}
    \begin{enumerate}
                \itemsep -0.15cm

    \item It would be reasonable to add a \cn{mapsfromchar} in order to
    build things like: $<\!\!-\!|$: 1

    \item It would be reasonable to add a \cn{Mapstochar} that could go
    with the double arrows to build things like $|\!=\!>$ : 1

    \item It would be reasonable to add a \cn{Mapsfromchar}
      that could go with the double arrows to build things like
      $<\!=\!\!|$ : 1

    \item A - for extending arrows: 1
    \item A = for extending arrows: 1

  \end{enumerate}


  \subsection{The ``Plain vertical arrows'' group: 6 (Alan 6)}
  \begin{itemize}
        \itemsep -0.15cm
  \item updownarrow cmsy'154
  \item Updownarrow cmsy'155
  \item uparrow cmsy'042
  \item downarrow cmsy'043
  \item Uparrow cmsy'052
  \item Downarrow cmsy'053
  \end{itemize}

  \subsection{Plain vertical extensible arrows: 6 (Alan 6)}
  \begin{enumerate}
        \itemsep -0.15cm
  \item Top sing arrow: cmex'170
  \item Bottom single arrow cmex'171
  \item Top double arrow cmex'176
  \item Bottom double arrow cmex'177
  \item Middle double arrow cmex '167
  \item Middle single arrow cmex '077
  \end{enumerate}

  \subsection{Plain extra vertical arrows: 0}
  Nothing added here.


  \subsection{The plain other arrows: 8 (Alan)}
  First the oblique arrows:
  \begin{enumerate}
        \itemsep -0.15cm
  \item CMSY'45
  \item CMSY'46
  \item CMSY'55
  \item CMSY'56

    What else: ?
  \end{enumerate}

  \subsection{The ``Ams obliques'' group: 2}
  \begin{enumerate}
        \itemsep -0.15cm
  \item msbm'36
  \item msbm'37
  \end{enumerate}



  \subsection{The ``Latex arrows'' group: 5}
  The four characters in position LASY'50 to '53 from the lasy font
  (These appear in the wasy font as well) must be put with the arrows.
  They are arrow heads. The squig \cn{arrow} in position '73
  of lasy should also be included: 5


  \subsection{The ``Ams other arrows'' group: 4}
  \begin{itemize}
                \itemsep -0.15cm
    \item Circle arrows MSAM: '10 to '11 :2
    \item Horizontal arrows MSAM:'113 '114 :2
  \end{itemize}

  \subsection{AMS horizontal arrows: 22 (Alan 23)}
  This includes all the horizontal arrows and the negated ones,
  that are listed page 280 of ``The joy of tex''.
  \begin{enumerate}
        \itemsep -0.15cm
  \item leftarrowtail
  \item leftleftarrows
  \item leftrightarrows
  \item leftrightsquigarrow
  \item lefttrightharpoons
  \item Lleftarrow
  \item looparrowleft
  \item looparrowright
  \item nleftarrow
  \item nLeftarrow
  \item nLeftrightarrow
  \item nleftrightarrow
  \item nrightarrow
  \item nRightarrow
  \item rightarrowtail
  \item rightleftarrows
  \item rightleftharpoons
  \item rightrightarrows
  \item rightsquigarrow
  \item Rrightarrow
  \item twoheadleftarrow
  \item twoheadrightarrow
  \end{enumerate}

  \subsection{Ams vertical arrows (Alan: 6) (here: 8)}
  \begin{enumerate}
        \itemsep -0.15cm
  \item MSAM:'24 upuparrows
  \item MSAM:'25 downdownarrows
  \item MSAM:'26 upharpoonright
  \item MSAM:'27 downharpoonright
  \item MSAM:'30 upharpoonleft
  \item MSAM:'31 downharpoonleft
  \item MSAM:'36 Lsh
  \item MSAM:'37 Rsh
  \end{enumerate}


  \subsection{Some control glyphs for access to arrows }
  These do not appear in the \fn{.dvi} file, they simply enable the
  construction of some arrows and slahsed arrows using the ligature
  mechanism.

% end of arrows






\section{All sorts of accents}

  \subsection{Basic size accents: 18}
  \begin{itemize}
        \itemsep -0.15cm
  \item All those that are created by macros in the Ams package: the 3
    dotted accent, and the 4 dotted accent.\quad 2

  \item The ones in \TeX: e\char'22\ e\char'23\  e\char'24\ e\char'25
    e\char'26\ e\char'27\ e\char'136\ e\char'137\ e\char'177\ e\char'176
    e$\mathchar"017E$. They all come from \fn{cmr} except for
    the last two from \fn{cmmi}. \quad 11

  \item Extra: a back-to-front vector arrow, \quad 1
  \item Extra: a double sided type vector arrow, \quad 1
  \item Extra: a square bracket used as an accent, \quad 1
  \item Extra: The previous one turned upside down, \quad 1
  \item Extra: an arc is requested by AMS, \quad 1
  \end{itemize}

  \note {The e\char'175\ in \fn{cmr} is not needed in maths, it is just
  a Hungarian accent.}

  \note {The $\mathchar"017F$ seems not to be needed in maths. It
    could be put in the text companion font.}



  \subsection{Double accents: 7}
  \begin{enumerate}
                \itemsep -0.15cm
    \item A bar and a dot on top,
    \item A dot and a bar on top,
    \item 2 dots with a bar on top,
    \item A bar with 2 dots on top,
    \item A hat and a tilde on top,
    \item A hat and bar on top,
    \item A double bar,
  \end{enumerate}

  \note {For the double accents, Spivak and Ralf Rey could do some
    archive research at the AMS. Similar research could be done at the
    APS, and the CUP.}


  \subsection{Variable size accents: $7*8=56$}
  Variable size has meant 5 different sizes until today.
  That number could be raised to 8.
  \note {If the accents are in a font
    loaded in three different sizes, the choice mechanism of
    \cn{mathaccent} will only look in the current style size (unlike
    the delimiter choice mechanism). Thus although one could hope to
    multiply the number of available sizes by three, in actual fact in
    a given style the number of automatically available sizes would not
    be multiplied.  All the same this \textem{would} give better
    results in each style,
    but it would also create \textem{compatibility problems}
    i.e.~formulae heights and widths may change.  Even if not done in an
    automatic way, the user would still have a larger range of accents
    to choose from.
    Compatibility problems could be avoided by redefining
    \cn{mathaccent} to a \cn{mathchoice}. Thus the accents could always
    come from text style, and the accented material could come from the
    current style. But this does not work either. In doing so one would
    no longer be able to take the base accents from the current style.
    Although one could make two macros. See paper ``Repacing
    \texttt{cmex}?'', same author.}

  \begin{enumerate}
                \itemsep -0.15cm
    \item e$\mathchar"017E$ the vector. \quad 8
    \item e\char'176\ the tilde. \quad 8
    \item e\char'136\ the hat. \quad 8
    \item e\char'26\ the bar. \quad 8
    \item Some people request a variable size arc. \quad 8
    \item The back to front vector arrow, \quad 8
    \item The double sided vector arrow, \quad 8
  \end{enumerate}


  \subsection{Under accents: 3 so far}
  Requests exist for the following:
  \begin{enumerate}
                \itemsep -0.15cm
    \item A tilde,
    \item A breve (\char'25)
    \item A bar
  \end{enumerate}

  Like for the double accents, research could be done at the AMS...
  ???????


\section{Core symbols}

The symbols that have some reason to live with the default math
material. There are mainly two reasons for them to be there: one is
kerning, and the other is design similarity.

  \subsection{For kerning reasons: 12}
  \begin{enumerate}
        \itemsep -0.15cm
  \item The period . CMMI
  \item The coma , CMMI
  \item The semi colon ; CMR
  \item The colon : CMR
  \item The exclamation mark ! CMR
  \item The (
  \item and the ) respectively opening-class and closing-class, CMR
  \item The [
  \item and the ] respectively opening-class and closing-class, CMR
  \item The $\{$,
  \item And the $\}$ (design similarity reasons also) in positions '146
  and  '147 of CMSY,
  \item The `/' as a delimiter\footnote {This is not accessible via a
      single key. The key `/' produces the sign $/$ taken from
      \fn{cmmi}.}, and  as a fraction sign, CMR

  \end{enumerate}

  \subsection{Basic geometric delimiters: 9}
  Should go in the core, for kerning reasons, like the other ( ) and [
  ].  If they don't fit in the core, they must go with the basics.  The
  ones listed here are all in CMSY, around '142, and '150.
    \begin{enumerate}
                \itemsep -0.15cm
    \item $\rangle$
    \item $\langle$
    \item $|$
    \item $\|$
    \item $\rceil$
    \item $\lceil$
    \item $\rfloor$
    \item $\lfloor$
    \item The $\mathchar"026E$ in position '156
    \end{enumerate}

     Test: $|f|,\|f\|,\lfloor f\rfloor,$$\lceil f\rceil,\langle
     f\rangle$ $f\lceil,f\lfloor,$$f\langle,f\backslash$ strange that
     no kerning seems to be needed here, where as it is necessary for
     the bracket.

  \subsection{New basic size delimiters: 9}
  Basic size means the same size as the parentheses and brackets in
  \fn{cmr}. The following is a preference order list of desired new
        delimiters:
  \begin{enumerate}
                \itemsep -0.15cm
    \item A $|||$ for use as $|||f|||$ a norme,

      Semantic brackets: $[\![$ and $]\!]$
    \item The opening semantic bracket,
    \item The closing semantic bracket,

    \item Opening multi set brackets $\{|$
    \item Closing multi set bracket $|\}$

      Unicode contains another style of brackets, they call them
      tortoise shell brackets. They look like:
      \begin{verbatim}
         /
        /
       /
       |
       |
       |
       \
        \
         \
      \end{verbatim}
      These are like parentheses, but with straight lines. No curves.
    \item The opening tortoise shell bracket,
    \item The closing turtoise shell bracket,

      Triangle brackets, something like:
      \begin{verbatim}
          /|
         / |
        /  |
        \  |
         \ |
          \|
      \end{verbatim}
    \item The opening triangle bracket,
    \item The closing triangle bracket.

%    \item The opening turtoise shell in Bbb.
%    \item The closing turtoise shell in Bbb.

  \end{enumerate}



  \subsection{For design similarity reasons: 24}
        All this group must live with the default alphabet for design
        similarity reasons.
  \begin{enumerate}
                \itemsep -0.15cm
    \item The question mark ? must live with the ! CMR
    \item The percent sign \% must live with the ! and ? CMR
    \item The at sign @ must live with the \%  CMR
    \item The \$ sign must live with the @ \% ? ! CMR
    \item The \& must live with \$, \% .. CMR
    \item The \# in CMR

    \item The inverted \&: must be found.

    \item The $\ell$ as a rounded `l'. CMMI
    \item The centered dot $\cdot$ for use as a multiplication sign,
    must live with the period. CMSY
    \item The asterisk $\mathchar"0203$ for use as a multiplication
      sign, in position '003 in CMSY.
    \item The $\propto$ sign must live with @, \%, $\ell$. In position
      '057 of CMSY.
    \item The $'$ or prime in position '060 of CMSY, one cannot
      separate the prime from the the prime ligature slots. (2
                        ligatures) Kerning of letters with the prime is
                        not possible, because
      the latter is set in superscript. ????????

    \item The backprime from MSAM'070 should live with the prime. ??????

    \item The $\infty$ sign in position '061,

    \item The $\emptyset$ in position '073 of CMSY,

    \item The \cn{check} mark in MSAM'130, ?????
    \item The \cn{between} double parentheses in MSAM'107
      should go with the normal parentheses.
    \item The $\Re$, could come out if necessary,
    \item The $\Im$ in position '074, and '075 of CMSY, could also
      come out if necessary.
    \item The $\dagger$ or dagger in CMSY'171, ?????
    \item The smile, ???????
    \item The frown form CMMI ????????
    \item Could maybe include the circled S from MSAM'163. ?????
    \item The Weierstrass symbol: $\wp$ only in one style, (could come
    out)
    \item The powerset symbol from
    \texttt{Martin.Ward@durham.ac.uk}. (could  come out if necessary)
      It looks something like this:
    \begin{verbatim}
          XXXXXXXXXXX
      XXXXXXXX    XXXX
    XXXXX           XXXX
    XXXXX           XXXX
  XXXXXXXX          XXXX
  XXX  XXX          XX
  XX   XXXXXXX  XXXXXX
  XX    XXXXXXXXXXXX
  XX    XXXX
  XX    XXXX
  XX      XXXX
  XX      XXXX
  XXXX    XXXX
  XXXX    XXXX
  XXXX  XXXX
    XXXXXX
    \end{verbatim}
  \end{enumerate}












\section{Symbols from \fn{lasy} that must be kept: }
The first four \fn{lasy} symbols are in \fn{msam}.

The ones in positions '50 through '53 are arrow heads, and are counted
as such in the corresponding group.

Character '60 is in the \fn{msam}.

'61 is not in the \fn{msam}, and should be kept.

Char'62: same as \fn{msam}'03 ?

Char'63: same as \fn{msam}'06?

Lasy'72: same as \fn{msbm}'163 or \fn{msam}'166 ?

Lasy'73: same as \fn{msam}'40 ?

A list of what should be kept from \fn{lasy}:
\begin{itemize}
        \itemsep -0.15cm
        \item Character '61: \quad 1
        \item ???
\end{itemize} \textbf{This makes a total of 1.}



\section{The ``Subset'' groups}
\note {None of these have anything to do with the \cn{sim} glyph.}

  \subsection{The ``subset plain'' group: 4}
  \begin{itemize}
                \itemsep -0.15cm
    \item The $\subseteq$ in position CMSY'022
    \item The $\supseteq$ in position CMSY'023
    \item The $\subset$ in position CMSY'032,
    \item The $\supset$ in position CMSY'033,
  \end{itemize}

  \subsection{The ``subset Ams'' group: 12}
  \begin{itemize}
                \itemsep -0.15cm
    \item From MSBM'040 to MSBM'43 : 12
  \end{itemize}

  \subsection{The ``In/ni plain'' group: 2}
  \begin{enumerate}
                \itemsep -0.15cm
    \item The $\in$ sign in position CMSY'062,
    \item The $\ni$ sign in position CMSY'063,
  \end{enumerate}

  \subsection{The ``sqsubset plain \& Ams'' group: 4}
  These do not have a place any where else:
  \begin{enumerate}
                \itemsep -0.15cm
    \item The $\sqsubseteq$, cmsy'166,
    \item The $\sqsupseteq$, cmsy'167,
    \item The sqsubset from MSAM'100,
    \item The sqsupset from MSAM'101,
  \end{enumerate}




\section{The ``Greater than'' group}

  \subsection{The ``Greater than Plain'' group: 8}
  \begin{itemize}
                \itemsep -0.15cm
    \item The $\leq$ in position CMSY'024,
    \item The $\geq$ in position CMSY'025,
    \item The $\ll$ in position CMSY'34,
    \item The $\gg$ in position CMSY'35
    \item $<$ less than CMMI'074,
    \item $>$ Greater than: CMMI'076,
    \item The alternative leq: msam'66
    \item The alternative geq: msam'76
  \end{itemize}

  \subsection{The ``Greater than AMS'' group: 30}
  \begin{itemize}
                \itemsep -0.15cm
    \item From MSBM'000 to '005: 6
    \item From MSBM'010 to '015: 6
    \item From MSBM'024 to '025: 4
    \item From MSBM'154 to '155: 2
    \item From MSAM'060 to '061: 2
    \item From MSAM'065 and '067: 2
    \item From MSAM'075 and '077: 2
    \item From MSAM'121 to '124: 4
    \item From MSAM'156 to '157: 2
  \end{itemize}

  \subsection{The ``greater than with sim'' group: 8}
  \begin{enumerate}
                \itemsep -0.15cm
    \item MSBM'022,
    \item MSBM'023,
    \item MSBM'032,
    \item MSBM'033.
    \item MSAM'046,
    \item MSAM'047,
    \item MSAM'056,
    \item MSAM'057
  \end{enumerate}

  The `shapee' \cn{sim}, and the geometric \cn{sim} are considered to be
  the same glyph, i.e. the difference that is sometimes visible is
  considered  to be a bug.


\section{The ``Succ'' groups}

  \subsection{The ``Succ without sim plain'' group: 4}
    \begin{enumerate}
                        \itemsep -0.15cm
      \item CMSY'026,
      \item CMSY'027,
      \item CMSY'036,
      \item CMSY'037
    \end{enumerate}

  \subsection{The ``Succ without sim AMS'' group: 10}
    \begin{itemize}
                        \itemsep -0.15cm
      \item MSBM'006, '007: 2
      \item MSBM'016, '017: 2
      \item MSBM'026, '027: 2
      \item MSAM'062 - '064: 3
      \item MSAM'074: 1
    \end{itemize}

  \subsection{The ``Succ with sim Ams'' group: 8}
  \begin{itemize}
                \itemsep -0.15cm
    \item MSBM'020, '021: 2
    \item MSBM'030, '031: 2
    \item MSBM'166, '167: 2
    \item MSAM'45,
    \item MSAM'55,
  \end{itemize}

  The `shapee' \cn{sim}, and the geometric \cn{sim} are considered to be
  the same glyph, i.e. the difference that is sometimes visible is
  considered to be a bug.





\section{The ``Sim'' group: 12}
  \begin{enumerate}
\itemsep -0.15cm
    \item sim CMSY'030
    \item approx CMSY '31
    \item simeq CMSY'047
    \item wr CMSY'157
    \item The bold MSBM'034
    \item MSBM'035
    \item MSBM'150
    \item The bold MSBM'163
    \item The bold MSBM'164
    \item MSBM'165
    \item MSAM'166, backsim
    \item MSAM'167, backsimeq
  \end{enumerate}


\section{Binops}
  \subsection{The ``Small binops plain'' group: 20}
  \begin{enumerate}
                \itemsep -0.15cm
    \item cap  CMSY
    \item cup CMSY
    \item uplus CMSY
    \item sqcap CMSY
    \item sqcup CMSY
    \item big circle CMSY
    \item big triangle up CMSY
    \item big triangle down CMSY
    \item vee CMSY
    \item wedge CMSY
    \item oplus CMSY
    \item ominus CMSY
    \item otimes CMSY
    \item oslash CMSY
    \item odot CMSY
    \item amalg CMSY
    \item bullet CMSY
    \item circ CMSY
    \item diamond CMSY
    \item star (5 branches): CMMI'77
  \end{enumerate}

  \subsection{Small plain left right triangles: 2}
  These should be replaced by the ones in the AMS fonts for math usage.

  They are also used as bullets, and should go in the text symbol
  encoding for such a usage.
  \begin{enumerate}
                \itemsep -0.15cm
    \item triangle left: CMMI'56
    \item triangle right CMMI'57
  \end{enumerate}

  \subsection{AMS left right open triangles: 8}
  These are also in LASY.
  \begin{itemize}
        \itemsep -0.15cm
  \item vartriangle left
  \item vartriangle right
  \item triangle left eq
  \item triangle right eq

    The previous four are in \fn{msam}
  \item Same 4 negated in \fn{msbm}: 4
  \end{itemize}

\section{Basic Symbols: 24}
A group of symbols used for typesetting basic mathematics.
These are mainly geometrics. Some have been added for similarity
reasons:
  \begin{enumerate}
        \itemsep -0.15cm
  \item = The equals sign, CMR'075
  \item - The minus sign, CMSY'00
  \item + The plus sign, CMR'053

  \item The $\times$ multiplication sign CMSY'002,
  \item The \cn{divide} sign $\div$ CMSY'004
  \item The \cn{divideontimes} from msbm'076 should live with divide
    and times.
  \item The rtimes from msbm'157 should live with the times.
  \item The ltimes from msbm'156 should live with the times.

  \item The $\pm$ sign in position CMSY'006,
  \item The $\mp$ sign in position CMSY'007,

  \item The $\equiv$ in position CMSY'021, Difficult to separate from
    other similar relations.

  \item The $\forall$ sign in position '070,
  \item The $\exists$ sign in position '071,
  \item The \cn{nexists} sign from msbm'100
  \item The $\neg$ sign CMSY'072,
  \item The \cn{varpropto} from AMS "5F. ????? or should this be left
    as a geometric?????
  \item The varemptyset from MSBM'77, ????

  \item Could go here: the upside down F: Finv from msbm'140 ???????
  \item And the back to front G: Game from Msbm'141 ???????

  \item Unary minus like en dash, could be CMR'173 but I personally
    think it should be shorter.

  \item The \cn{varnothing} from MSBM'77,
  \item smallsetminus from msbm'162

  \item The $\perp$ perp or bot sign in position '077,
  \item top sign CMSY'076
\end{enumerate}




\section{Radical}
  \subsection{The \TeX\ radicals: 10}
  Currently available in \fn{cmex} are:
  \begin{itemize}
                \itemsep -0.15cm
                \item Five radical signs: 5
                \item The vertical bit needed to construct the big
                radical '165 : 1
                \item The top bit of the constructed radical. '166 : 1
  \end{itemize}

  \fn{cmsy} includes the basic size of the radical. It
  has always been loaded in three sizes, and must remain so.
        If it is taken out of \fn{cmsy}, and put in a \fn{cmex}
        replacement,
        then this point must be taken into consideration.

        If the new \fn{cmex} is loaded in one size, it must contain
        three different sizes of the radical in order to stay compatible
        with plain: 3


  \subsection{New radicals: 2}
  Request made by: \texttt { HITT\% USOUTHAL.BITNET@SHSU.edu }.

  One can overload the little vertical extensible module of the
  radical, in order to produce a left quantum operator. For the right
  quantum operator, the glyphs could be available, but the radical
  macro can't be used.
        A specific macro could be designed and it would need two glyphs:
  the top right hand corner, and the repeatable
  vertical bit: 2






\section{The integrals family: 18}

  \subsection{Big `bigops' size: 7}
  \begin{enumerate}
                \itemsep -0.15cm
    \item The single integral.
    \item The double integral. Could be done with kerning if there is
    not enough space.
    \item The triple integral. Could be done with kerning if there is
    not enough space.
    \item The single O integral.
    \item The double O integral.
%    \item The triple O integral. ??????? J\"org says no.
    \item The sigma integral. For physics: J\"org.
    \item The slash integral. For physics: J\"org.
  \end{enumerate}

  \subsection{Small `bigops' size: 7}
  The same as in big `bigops' size.

  \subsection{Small size: 7 }
  This refers to the size of the \cn{smallint} in CMSY.
  \begin{enumerate}
        \itemsep -0.15cm
  \item The single normal integral.
  \item The single O integral.
  \item Double O integral. ???? (J\"org thinks yes)
  \item Double normal integral ???? (J\"org thinks yes) Could be done
    with kerning if there is not enough space.
  \item Triple normal integral ???? (J\"org thinks yes) Could be done
    with kerning if there is not enough space.
  \item The sigma integral.
  \item The slash integral.
  \end{enumerate}

  Mail from HSS:
  \begin{quote} %\fontsize{7}{7}\selectfont
    \cn{doubleoint} is used by Becker in "electromagnetic fields and
    interactions" (Dover). I also saw \cn{tripleoint} used in
    electromagnetic theory books although both are somewhat archaic.

    Concerning the small version (in \fn{cmsy}), I suggested this solely
    for reasons of completeness. The need for it is less now that
    \fn{cmex} will be loaded in three sizes. But the small version of
    \cn{int}
    \& \cn{oint} look a bit large when used in inline formulas. The
    \cn{smallint} \& \cn{smalloint} etc. may be a choice for some
    authors in the latter case.

    YH also pointed out, that the upright versions of integrals are very
    common in textbooks. since the integral sign is one of the most
    common symbols used in maths, it may not be a bad idea to include
    upright versions of *all* integral signs in \fn{cmex} (with
    corresponding small versions in \fn{cmsy}) again for reasons of
    completeness.

  \end{quote}

\section{AMS Vdash group: 10}
\begin{enumerate}
\itemsep -0.15cm
\item MSBM'054 nparallel
\item MSBM'055 nmid
\item MSBM'056 nshortmid
\item MSBM'057 nshortparallel
\item MSBM'061 nVdash
\item MSBM'062 nvDash
\item MSBM'063 nVDash
\item MSBM'160 shortmid
\item MSBM'161 shortparallel
\item MSAM'015 Vdash
\item MSAM'016 Vvdash
\item MSAM'017 vDash
\end{enumerate}

\section{Plain and \fn{lasy} miscellaneous symbols: 6}
\begin{enumerate}
\itemsep -0.15cm
\item CMSY'20
\item CMSY'140
\item CMSY'141

  Should live with the two previous:
\item MSBM'060 What about the back to front version of this ?????
\item LASY'061 the bow tie,
\item LASY'62 the small box. It is smaller than the one in the AMS
  fonts. ????
\end{enumerate}


\section{AMS equals friends: 10}
  \begin{enumerate}
\itemsep -0.15cm
  \item msam'155
  \item msam'154
  \item msam'120
  \item msam'73
  \item msam'72
  \item msam'54
  \item msam'53
  \item msam'52
  \item msam'51
  \item msam'44
  \end{enumerate}


\section{AMS miscellaneous geometric symbols: 21}
\begin{enumerate}
\itemsep -0.15cm
\item msam'174
\item msam'173
\item msam'171
\item msam'170
\item msam'165
\item msam'164
\item msam'161
\item msam'160
\item msam'151
\item msam'150
\item msam'147
\item msam'146
\item msam'141
\item msam'140
\item msam'136
\item msam'135
\item msam'134
\item msam'133
\item msam'132
\item msam'131
\item msam'050
\item msam'005

\end{enumerate}



\section{AMS boxes and friends: 15}
\begin{enumerate}
\itemsep -0.15cm
\item MSAM'000
\item MSAM'001
\item MSAM'002
\item MSAM'003
\item MSAM'004
\item MSAM'006
\item MSAM'007
\item MSAM'014
\item MSAM'106
\item MSAM'110
\item MSAM'111
\item MSAM'112
\item MSAM'115
\item MSAM'116
\item MSAM'117
\end{enumerate}



\section{The horizontal curly braces: 10}
Their design should be the same as the vertical braces. Add two
horizontal extension modules for them, since if they are drawn with
rules, digitization errors may cause them not to line up with the
horizontal brace glyphs. What's more, this would enable the designer
to choose there boldness.

Plus two extra middle bits. So that the designer is not restricted by
the number of slots. Knuths design could use that little amount of
glyphs, but maybe other will need more.

Count: for the downwards brace: 2 end bits, 1 middle bit, 2 extensible
modules. That makes a total of 5 per curly brace. One up, and one
down: makes 10.



\section{Big and extensible \TeX\ delimiters from \fn{cmex}: 78}
This group includes delimiters that are in \fn{cmex}. And an extra
little extensible module for the \{ and \}.
  \begin{itemize}
        \itemsep -0.15cm

  \item Four different sizes of ( and ): 8

  \item Extensible versions of ( and ): 6 \\
    Left and right extensible modules are '102 and '103. Top and bottom
    are '060, '061, '100, '101.

  \item Four different sizes of [ ]: 8

  \item Extensible version of [ and ]: 6 \\
    The extensible modules, one for the right bracket, and one for the
    left bracket are: '066, '067. The top and bottom pieces are: '062
    to '065.

  \item Four different sizes of \{ and \}\,: 8

  \item Extensible module for \{ and \}\,: 7 \\
    The extensible module ('76) used for the curly braces is very small,
      because it is added twice: once above the middle piece, and once
      below the middle piece. Its height is half that of the parentheses
      extensible  module. Other pieces are: '070 - '075.


  \item An extra extensible module for the \{ and \}\,: 1\\
    There is only one extensible module for both the left and the
    right curly brace in \fn{cm}. This is because the left-right spread
    of a curly brace is symmetrical in \fn{cm}, unlike the parentheses
    for example.
    This may not be the case for other designs.

  \item Four different sizes of $\langle$ and $rangle$: 8

  \item Four different sizes of $\backslash$ and $/$: 8

  \item Four different sizes of $\lfloor$ and $\rfloor$: 8 \\
    The extensible version is build with the same pieces as the
    extensible brackets.

  \item Four different sizes of $\lceil$ and $\rceil$: 8 \\
    The extensible version is build with the same pieces as the
    extensible brackets.

  \item Extensible vert and parallel: 2 \\
    Extensible versions of the vertical bar and the double vertical
    bar. They are their own extensible modules: '014 and '015.

  \end{itemize}




\section{Bigops}
  \subsection{Old bigops from \TeX: 28}
  There are two glyphs for each bigop.
  \begin{enumerate}
                \itemsep -0.15cm
    \item The sqcup
    \item The circled integral
    \item The circled dot
    \item The circled plus
    \item The circled times
    \item The sums
    \item The prods
    \item The normal integrals
    \item The bigcups
    \item The bigcaps
    \item The U plus
    \item The wedges
    \item The vees
    \item The coprods
  \end{enumerate}

  \subsection{New double sized `bigops': 26 }
  All these should come in two sizes, in the same font, like the present
  \cn{bigcup}. One for display style, and one for text style. That
  makes two glyphs for each one.

  \begin{enumerate}
                \itemsep -0.15cm

    \item A double sized sqcap $\sqcap$ \cn{bigsqcap} (can be found in
                \fn{cspex})

    \item Two sized $\bigcirc$ with $\vee$ inside.  $\bigcirc
      \!\!\!\!\!\vee$ proposed name: \cn{ovee}, and \cn{bigovee}.
      Can be found in \fn{cspex} and \fn{stmary}.

    \item Two sized $\bigcirc$ with $\wedge$ inside.  $\bigcirc
      \!\!\!\!\!\wedge$ proposed name \cn{owedge}, and
      \cn{bigowedge}. Comment from Alan:
      \begin{quote}
        ``As far as I'm aware nobody has *ever* used these glyphs in a
        paper.  I put them in St Mary's Road because I needed them at
        the time, but I shortly abandoned writing the paper they were
        going to be used in.  Please don't include them!  (If we are
        going to, we need to include <ovee> and <owedge> as well as
        <bigovee> and <bigowedge> which are the ones you described.)''
      \end{quote}

    \item Dijkstra choice: $[\!]$ CSPEX

    \item A wide Dijkstra choice. CSPEX . Comment from Alan:
      \begin{quote}
        If this is the glyph I think it is, it's not quite a wide
        Dijkstra choice in shape (although mathematically it's the
        same thing as Dijkstra choice).  The two glyphs are:

        $<$dijkstrachoice$>$ looks remarkably like [ and ] glued
        together.

        $<$oblong$>$ looks like $<$sqcap$>$ but with the square
        completed.

        $<$oblong$>$ is used in CSP in conjunction with $<$sqcap$>$,
        so it's quite important that they look the same.  In
        particular, they need to be of the same width because if they're
        not, formulae sometimes don't line up properly...
      \end{quote}

    \item Parallel \cn{bigparallel} just a double sized version of
      parallel.

    \item Interleaving $|\!|\!|$ : \cn{biginterleaving}
      \begin{quote}
        `Interleaving' and `parallel' are used in (at least) three
        different ways:
        \begin{itemize}
                                %\itemsep -0.15cm
        \item As delimiters $||$foo$||$ and $|||$foo$|||$.  These
          should come in basic-sized and extensible versions.
        \item As binary operators p $||$ q and p $|||$ q.  These can
          be the same glyphs as for the basic-sized delimiters.
        \item As `big' operators $||_i \;p_i$ and $|||_i \;p_i$
          similar to \cn{bigcup}.  These should come in text style and
          display style versions.
      \end{itemize}
      The big operators are not the same glyphs as the extensible
      delimiters.
    \end{quote}

  \item \cn{bigcupdot}: A `U' with a dot in it. Something like:
    $\bigcup\!\!\!\!\cdot$

  \item \cn{bigcapdot}: an upside down `U' with a dot in it. Something
    like: $\bigcap\!\!\!\!\cdot$

  \item An inverted \& called \cn{dnasrepma}

  \item Large operator symbol based on, an asterisk sign.

  \item Large operator symbol based on a times sign.

  \item Large operator symbol based on, a hash sign.

  \item Large operator symbol based on, an ampersand sign.

  \end{enumerate}



\section{Non classified existing symbols}
Here is a list of some symbols that do not have a place elsewhere:
\begin{itemize}
        \itemsep -0.15cm
  \item The different shapes of \# should find a place, although one is
    already in the core group.
\end{itemize}








\section{A list of new glyphs}

The following symbols should be added in the math fonts. Some have
already been designed by various people, so it should be possible to
find them...


  \subsection{Basic size operators: 2}
  Basic size means the same size as the operators in \fn{cmsy}.
  \begin{enumerate}
                \itemsep -0.15cm
    \item Something like \cn{cupdot} and
    \item Something like \cn{capdot} Frank M. can justify these.
  \end{enumerate}


  \subsection{New multi-sized, and extensible delimiters: 47}
  Count: 8, 6,  1,  8, 8,   8,   8, makes 47 .

  A multi-sized delimiter means: 4 sizes for each side : 8 glyphs.
  Plus and extensible version: top, bottom, extension module for both
  sides: 6 glyphs. Sometimes also a middle: 8 glyphs. Total: 16 or 14.
  \begin{enumerate}
                \itemsep -0.15cm

    \item Four sizes of the semantic brackets $[\![$ and $]\!]$: 8

    \item An extensible version of the semantic brackets: 6 \\
      (Top - Middle - Bottom) * 2 makes 6.

    \item An extensible version of $|||$ for use as $|||f|||$ (a
      norme). Just the extension module: 1

    \item 4 sizes of multiset brackets $\{|$ and $|\}$: 8

    \item An extensible version of multi-set brackets: 8 \\
      (Top - Bottom - Middle - extensible module) *2 makes 8.

    Unicode contains another style of brackets, they are called
      tortoise shell brackets. They look like
        \begin{verbatim}
          /
         /
        /
        |
        |
        |
        |
        |
        \
         \
          \
        \end{verbatim}
    \item Four sizes of tortoise shell brackets: 8 \\
      No extensible version. could add them in.


    \item Four sizes of triangle brackets: 8 \\
      They look like:
      \begin{verbatim}
          /|
         / |
        /  |
        \  |
         \ |
          \|
      \end{verbatim}

  \end{enumerate}



\subsection{Geometrics: 21}
\begin{enumerate}
        \itemsep -0.15cm

  \item The \fn{ams} smaller or equal and greater or equal must not be
        forgotten.

  \item Linear `is implied by' if o-- and o--o [bb: There are also
  versions of these with filled-in circles.]

  \item From JMR: something like: $\raise .1cm \hbox{$|$}\!\_$
    maybe the same upside down.

  \item From JMR: Something like: $\_\!\raise .1cm \hbox{$|$}$
    maybe the same upside down.

  \item \texttt{Here are some other symbol I once needed: \cn{opm},
      \cn{omp}.  Why?  There is \cn{oplus} and \cn{ominus} and there
      is +, -, \cn{pm} and \cn{mp}, but there's no \cn{opm} and
      \cn{omp}.}

    \texttt {I once needed them in a context, where \cn{oplus} and
      \cn{ominus} were used as subscripts to indicate symmetric and
      antisymmetric wave functions that were normalized. It is easy to
      specify formulae that include both cases using \cn{pm} and
      \cn{mp}, but suddenly there was no \cn{opm} and \cn{omp}. I
      constructed symbols using the circle from the copyright sign,
      but that was not actually the perfect size.}

\end{enumerate}

\subsection{New arrows}
\begin{itemize}
\itemsep -0.15cm
  \item Alan J. wrote: \cn{arrownot} and \cn{Arrownot}, so that for
       example \cn{arrownot}\cn{mapsto} is visually compatible with
      \cn{nrightarrow}.  Describing the same thing he also wrote: Add
      the `building blocks' for the AMS negated relations, for example a
       \cn{arrownot} to build \cn{nlongrightarrow} and
         \cn{nrightarrowfill}.

  \item The building blocks to make \cn{mapsfrom} $<\!\!-\!|$
    \cn{Mapsto} $|\!=\!>$ and \cn{Mapsfrom} $<\!=\!\!|$

  \item Lfloor, Rfloor, Lceil, Rceil like $\lfloor\!\lfloor$

  \item Arrows with triangles on the end.
$$<\!\!\!|\!\!-\!\!-,\ -\!\!-\!\!|\!\!\!>,\ <\!\!\!|\!\!-\!\!-|\!\!\!>$$

  \item Equals like symbol: $<\!--\!>$ with $==$ underneath.
\end{itemize}


\subsection{Non geometrics: 19}

\begin{enumerate}
\itemsep -0.15cm
%  \item The double circled integral, or the surface integral for
%    physicists.
%  \item A single integral with on top a $\Sigma$:
%    $\displaystyle\int\!\!\!\!\!\!\textstyle\Sigma$ (J\"org)
%  \item A single integral with a slanted dash:
%    $\displaystyle\int\!\!\!\!\!\textstyle-$
%  \item A triple, circled integral.

  \item Possibly something like $\bar{}\!\!($ and $\bar{}\!\!)$ if the
  bar was touching the parentheses.

  \item Banana brackets: look (sort of) like $(\!|$ and $|\!)$.  Or
    they look like bananas if you believe Jeremy... Alan: The St. Mary
    Road font includes samples of them, in a line-drawing style. Since
    I have not seen them in real use, I cannot comment, if this style
    or rather the look of \& in \fn{cmr} is appropriate.

  \item lightning (\fn{wasy}) --- probably does not belongs in maths.
    Alan: It is actually used though!  It means `interrupt' in process
    theory, c.f.  Communicating Sequential Processes, Hoare, Prentice
    Hall 1985.  I don't know how widely used outside process theory it
    is though...



%  \item Upright partial derivation sign. In the Greek-like stuff.
%  \item Arabic letter dal: looks something like a back to front $c$.
%    This should live with the Hebrew letters.
%  \item Remember that message forwarded by J\"org. The V-bar, and the
%    parenthesis-bar. jvpurcel@vela.acs.oakland.edu. The V-bar is in
%    with the Latin like stuff.

  \item Katakana character that looks like a spiral. (bb)

%  \item Here is a proposal from J\"org: Mylona's and Whitney's
%    ligatures: two extra symbs. These would come in italic, and
%    upright like the rest of the Greek stuff. J\"org himself does not
%    like this.  [bb: But we're not talking about text! Why put these
%    in a math font?]  It now looks as though this is not going to be
%    included in the new math encoding.

  \item A lower-case sigma with a long tail that goes a little bit below
  the baseline.

  \item The two versions of the \# hash sign must not be forgotten. It
                seems that one is geometric, and one is not: the slanted
                hash sign and the upright hash sign.


  \item An \cn{inviota} is sometimes requested on the net. I'll send
    you a reference file for it. (J\"org)

  \item More ?

\end{enumerate}
\end{filecontents}



\begin{filecontents}{l3d007c.tex}
%a macro to put a frame round a box
\newcommand{\frameit}[1]{\vbox{\hrule\hbox{\vrule
      \hskip .1cm\vbox{\vskip .1cm\hbox{#1}\vskip .1cm}\hskip .1cm
      \vrule}\hrule}}


\begin{quote} \small
  The placing of mathematical accents in \TeX{} is done with
  the following control sequence: {\tt\string
    \mathaccent"xyzz\{box\}}. {\tt"xyzz} is the code that
  identifies the accent, and `{\tt box}' is the entity to come
  under the accent. The box can be any type of box known by \TeX{}: a
  single letter, a vbox, an hbox, etc... The accent code used is a
  usual math code (refer to any book on \TeX\ for more details).  The
  accent is basicaly centered over the given box, but there are ways
  of influencing the way the centering is done. The vertical placement
  is as we shall see a little more tricky.
\end{quote}






\section{The accent choosing}

Let us call $x$ the box that is due to be accentuated, and $u$ its
width.
We shall first suppose that $x$ is a single character.\\
\vskip .05cm
\centerline{$x:\qquad \rightarrow$\lower .2cm%
\hbox{\frameit{\huge x}}$\leftarrow$u}
\vskip .2cm

If the accent is part of a list of successors\footnote{See document on
  charlist for more details about successors.}, \TeX{} chooses the one
whose total width is {\em as close as possible} to $u$, but still
{\em smaller or equal} to $u$.

\vskip .25cm \textbf{Note.} A list of successors can only take place
in one and one font only, and Knuth reminds us of this fact in rule 12:
``[...] has a successor in its font [...]'' This makes me wonder...
The hat and tilde come in their smallest size in cmr position '136 and
'176. The other sizes are in cmex position '142 - '147. The
following:\\
\verb|   $\mathaccent"005E{e}$, ...  ,$\mathaccent"005E{eeee}$|\\
produces: $\mathaccent"005E{e}$ and
$\mathaccent "005E{ee}$ and $\mathaccent "005E{eee}$ and
$\mathaccent "005E{eeee}$. So as expected there is no automatic sizing,
seeing as the hat in cmr is not part of a charlist. Same test with
tilde: \\
\verb|   $\mathaccent "007E{e}$, ...  ,$\mathaccent"007E{eeee}$|\\
 produces: $\mathaccent"007E{e}$ and $\mathaccent
"007E{ee}$ and $\mathaccent "007E{eee}$ and $\mathaccent "007E{eeee}$.
In order to get a hat that changes size automatically one must
call the smallest hat in \texttt{cmex} directly , like \cn{widehat}
does. \cn{widehat} is defined as: \cn{mathaccent} \texttt{"0362}.
Family three indicates cmex, and 62 is the hex position of the
smallest hat in \texttt{cmex}. Thus \cn{widehat}\verb|{abda}|
produces: $\widehat{abda}$.





\section{The horizontal placing}

The accent followed by its italic correction is put into a box $y$
whose width we shall call $w$.\\
\vskip .05cm
\centerline{$y:\qquad \rightarrow$\lower .2cm\hbox{\frameit{ acc
{\tt \string\/}}}$\leftarrow\ w$}
\vskip .2cm

Let us call $s$ the kern between the character in box $x$ and the {\em
skewchar\/}\footnote{See below for more details about the skewchar.}
---~taken in that order. The box $y$ is first centered on top of the
box $x$ and then shifted rightwards by the amount $s$.

\textsl {If the box $x$ is {\em not a single character}, the amount
  $s$ is taken equal to zero, therefore the skewchar is ignored and
  the box $y$ is normally centered\footnote{But the box $y$
  contains the italic correction, which influences the centering.}
  over the box $x$.
}
%
%
%
\subsection{The `skewchar'}
The \textem {skewchar} is a specific character that D.E.~Knuth decided
to use ---~in the way described above~--- for placing accents. Each font
should have its own skewchar.

For most characters, the {\tt .tfm} file specifies a particular
kerning of each letter with the skewchar. This is true for the
computer modern fonts, but other font designers may not have used this
feature.

Why choose one skewchar rather than another? This is because the
character $\mathchar"017F$ chosen by Knuth does not have any other
kerning that could have been disturbed by the skewchar kerning. This
choice may not always be good for all fonts, because it depends on
what the character in position '127 is. Thus a font designer might
choose another skewchar and put the necessary kernings in the {\tt
  .tfm} file.  Accent glyphs can be used as skewchars, because they
are usually not subject to kerns from other glyphs.

As long as one uses the {\tt cm} fonts, it does not make much sense to
change the skewchar, unless one wants to get strange effects, or
unless one intends to change the {\tt .tfm file}. One should remember
that an assignment to skewchar is not undone at the end of the group,
but is a part of the global font information. A local change
therefore involves saving the original value, and then restoring it.

\subsection{The italic correction}
The presence of the italic correction here is a little mysterious.
Its effect is to shift the accent to the left compared to the position
it would have without it. One should note that it is still added when
the $x$ box is more than one character.
%My personal theory is that
%whereas the skewchar-kerning enables a general positioning of all
%accents on one character, the italic correction enables an adjustment
%of a particular accent over all characters.


\subsection{Conclusions}
It is useless changing the skewchar unless one also modifies the
{\tt .tfm} file(s), which contain(s) the info for the skewchar kerning,
and for the italic correction.

Accents from any font can be positioned on characters from any other
font if there is a skewchar in the character font.

%
%
%
%
%
%
\section{The vertical placing}
This is a little more tricky.  Here as well one can start
by supposing that the character to be put under the accent is single.
All previous notations are kept. $\chi$ is the x-height
of the accent's font i.e. {\tt\string \fontdimen5}. Let us call $h_x$
the height of the $x$ box:\\
%
\vskip .1cm

\centerline{\vbox{\hbox{$\ \downarrow$} \frameit{\huge x} \kern
    -.2cm\hbox{$\ \uparrow$}
\hbox{$h_x$}}}
\noindent
and $h'_x$ the height of a box $x'$ containing the character to be put
under the accent, together with an empty superscript, and an empty
subscript.\\
\centerline{%
$x':\qquad $ \lower 1.5cm \vbox{\hbox{$\ \downarrow$}%
\frameit{$\mathop{\frameit{\huge x}}\limits^{\frameit{\scriptsize%
super}}_{\frameit{\scriptsize sub}}$}%
\kern-.2cm
\hbox{$\ \uparrow$}%
\hbox{$h'_x$}}%
}%end of centerline
%

\noindent
Now set $\delta$ with the minimum of $\chi$ and $h_x$, and increase it
by $h'_x-h_x$.

The end result is a \cn{vbox} $z$ containing box $y$ (the accent
correctly positioned lateraly), followed by a (vertical) kern of
$-\delta$, and then box $x'$ .  A normal accent char has the folowing
aspect:
\frameit{\char127}. This explains why the kern of $-\delta$ is needed.
Without it the gap under the accent would be to big. One can now
understand why an `O' used as an accent over an `e' produces the
following: $\mathaccent"714F{e}$.

If the height $h_z$ of $z$ is smaller than $h_x$ then a kern
is added on top of $z$ in order to make the end height that of $x$.
Finally, the width of $z$ is set to the width of $x$.
%

\textsl{If the character to be accentuated is not single, $\delta$ is
  not increased by $h'_x-h_x$. In other words, one can forget about
  the subscript and superscript.
}%end of textsl

\subsection{Conclusions}
The x-height is used for the vertical placing of accents. This means
that one cannot mix in the same font table glyphs designed with
different x-heights.
\end{filecontents}



\begin{filecontents}{l3d007d.tex}
\begin{quote}\small
The re-encoding of the math fonts cannot be thought of without a
closer look at \TeX's mechanisms for dealing with extensible
characters. This includes all characters that come in different sizes,
all characters that are constructed, and the operators that usually
come in two sizes. The ultimate questions are: how
to implement these characters? What sort of kerning can be
done with them?
\end{quote}












\section{Let us start with the easiest: The operators}

A large operator like $\sum$ will be vertically centered with respect
to the axis when it is typeset. Thus, large operators can be used
with different sizes of text. This vertical adjustment is not made for
symbols of other classes. [The \TeX book p.155]

This is a sum in tex: $\sum_{i=0}^{i=n} i$ and this is a prod in text:
$\prod_{j=0}^{i=m} j$.
Same test in display maths:$$ \sum_{i=0}^{i=n}
vi\quad \prod_{j=0}^{j=m} j$$

The sum is defined as follows:
\begin{verbatim}\mathchardef\sum = "1350 \end{verbatim}
If \TeX\ is in display style, it looks to see if the character in
position "50 of family 3 (The extensible \fn{cmex} family) has a
successor\footnote{See below for explanations on successors.}.  If it
does then the successor is taken. When not in display style, \TeX\
just takes character "50 from family 3.

\subsection{Conclusions for operators}
Both occurrencies of a double sized operator must be in the same font
table, because they are linked by the successor mechanism.

%JWZ commented this out: don't understand.
%It would be a good idea to try and make them all similar sizes so that
%they don't require many different sizes. This is due to font
%restrictions.
%\subsection{Going further}
%One wonders if the choosing is recursive, but presumably it isn't,
%because of the ``if we are in display mode'', which does not do any
%size testing.




\section{How characters can be linked}

\subsection{\texttt{charlists}}
They enable several characters of the same font to be linked together.
By simply typing \texttt{charlist oct"000": oct"020": oct"022":
  oct"040": oct"060"} in the metafont source, one links all the
occurrences (in the \fn{cmex} font) of the left parentheses in
increasing size-order.

They are used for:
\begin{itemize}
\item Linking variable-size delimiters,
\item Linking variable-width accents,
\item Making pairs of operators that come in textsize and
  display size.
\end{itemize}

\subsection{The \texttt{extensible} lists p318 metafont book.}

An extensible glyph is identified with one of its pieces. One has
to decide which piece is going to be used for this identification.
For instance, in \texttt{extensible oct"060": oct"060", 0, oct"100",
  oct"102";} the first \texttt{oct"060"} is the identifier of the whole
  glyph.
The next three characters are the top, middle, and bottom pieces. The
last character position is that of the piece to be repeated as many
times as necessary between the top
and middle, and between the bottom and middle pieces. All three pieces
are optional. When they are not needed, they are replaced with `0'.
But if a zero is put in the repeater position, then character `0' will
effectively be used as a repeater.

\subsection{Restrictions compiled from p318 of the Metafont book}
\begin{itemize}
\item An \texttt{extensible} identifier can only appear at the end of a
  \texttt{charlist}.
\item A kerning/ligtable label can only appear at the end of a charlist.
\item One cannot use an extensible identifier as a kerning label.
\item One cannot use an extensible identifier as a ligature label.
%JWZ commented out the next two lines.
%\item this restriction explains why one specifies the kerning and the
%  ligatures in the same table.
\end{itemize}

\subsection{Conclusions}
Kerning with parentheses is going to be very tricky...

A delimiter is made of two sets of characters that can be in two
separate font tables. The glyphs in these two sets (the delimiters),
can be kerned with the characters that accompany them in their font. So
one must place them correctly.

\note {There are no parentheses in the \fn{cmmi} fonts, so this
  possibility has not been used.}

There is a list of all \fn{cm} extensibles, and charlists in ``Computer
modern typefaces'' p66.


\section{The vertical constructables, or ``those that come in pieces''
  --- \textit{Delimiters}}

\subsection{A few notes}
First information about delimiters p.171 of the \TeX book. They are not
all of the same type. Some are (bigl,bigr) Open/close atoms, and
others are (bigm) rel and (big) ord atoms. On the other hand, a
\texttt{\string\left} \texttt{\string\right} grouping is inner.

When a delimiter gets larger, its height and depth both grow by the
same amount.

In the \fn{cmex} font, most of the vertical constructable glyphs are
below the baseline, in the \texttt{.tfm} file.

\subsection{A first description of the choice mechanism}
A delimiter is defined by a small ``character'' and a ``large
character''. \textbf{These characters can be in different families,
and therefore in different font tables.} We shall call the small
character $a$ in family $f_a$, and the large character $b$ in family
$f_b$.

The search first starts by considering the three different sizes of
char a in its family\footnote{i.e. the scriptscriptsize then the
  scriptsize and finaly the textsize. This is done in an optimized
  way: if the current size is bigger than scripscriptsize, no point
  looking in the scriptscriptsize font, same goes for the other two
  sizes.}. When testing a character in a given font table, its
successors are tested before going to the next bigger font. In other
words, for each member of a family, starting with the smallest, and
going to the biggest, \TeX{} first looks at char $a$ and then at its
successors\footnote{See later explanations.} in the same
font table. If nothing suitable is actually found within the family
$f_a$ a similar search is done in the family $f_b$ based on character
$b$.

The search stops when the character being tested has a sufficient
height plus depth, or when it is extensible\footnote{Obviously in this
  case the appropriate delim has been found, seeing as it can be made
  to any given size bigger than the non-extensible characters.}.

If either of the couples $(a,f_a)$, and $(b,f_b)$ are set to
$(0,0)$ then the corresponding part of the search is bypassed.

If none of the characters actually found are suitable, the biggest is
taken, i.e. the one with the greatest height plus depth.


\subsection{A second description of the choice mechanism by Victor E.}
\TeX\ first tries the small variant, and if that is not satisfactory
(or if the left part of the delimiter code is 000) it tries the large
variant. If trying the large variant does not meet with succes \TeX
takes the largest delimiter encountered in this search. If no
delimiter at all is found, (which can happen if the right hand part is
also 000), an empty box of width \texttt{\string\nulldelimiterspace}
is taken.

Investigating a variant means in sequence:
\begin{itemize}
\item If the current style is scriptscript style, the scriptscript font
  of the family is tried.
\item If the current style is script style or smaller the script font of
  the family is tried.
\item Otherwise the text font of the family is tried.
\end{itemize}
Note: The plain format puts the \fn{cmex10} font in all three styles of
family three.

Looking for a delimiter at a certain position in a certain font means:
\begin{itemize}
\item If the character is large enough, accept it.
\item If the character is extensible, accept it.
\item Otherwise, if the character has a successor, (the same but
  bigger), try the successor.
\end{itemize}


\subsection{The final width of the delimiter ?}
If the character chosen ends up to be an \textbf{extensible} one,
\textem{the resulting width is that of the repeatable piece.}
Otherwise (in the case of a normal character) the width is that of the
chosen character \textem{plus its italic correction.}


\subsection{Conclusions for delims}
\begin{itemize}
\item The different sizes of a delimiter can be spread in two font
  tables if it is necessary.
\item A given delimiter does not absolutly have to have two sizes.
\item One can adjust the width of the repeatable piece for
  extensibles.
\item One can adjust the italic correction of the normal ``single
  glyph'' delimiters, in order to get it further away from things like
  $f$, $j$, $g$ and $p$. In the present case of `(' (in text
  size) it comes from the text font \fn{cmr*} so obviously has not got
  any italic correction. The vertical bar has not got any either
  (checked in the \fn{.pl} files). This is quite a global solution and
  the italic correction will be added in all cases: if it is small it
  may not bother anybody and should have the right effect
  in most cases.
\end{itemize}





\section{References}
The \TeX book ``Construction of math symbols'':
151 mathchoice: no good;
152: about delimiters and size choosing;
178: using phantom and vphantom - no good;
358: how large operators are assigned in plain, and
   some horizontal constructables;
359: all the 24 delims that can change size and the big and bigg macros;
360-361: nothing.

The \TeX book ``math symbols'': 127-128: nothing, 289: nothing, 290:
interesting things about delims.

The \TeX book ``Mathcode'':
134: tiny little bit at the bottom about mathcodes,
154: a list of the 8 classes and (3) about variable family and (bottom)
   about mathcode,
155: the mathcode "8000 + mathchardef + mathinner,
156: delcodes and delimiter,
157: radical,
214: nothing,
271: nothing, 289: nothing,
319,326: answer to exercises,
344: where all the mathcodes are set so that `1' comes from fam 0 and
`a' from fam 1,
345: where the delcodes for plain tex are set.

The \TeX book ``mathop'': 155 cf mathcode, 178 cf construction of math
symbs, 291 a bit about mathaccent, radical 324-325  361

The \TeX book ``dilimiters'':
156: delcode;
157: radical;
214: nothing; 289: nothing; 271: nothing;
290: in the middle ``A delim...''; 345: see at the top;
359: Plain tex definitions of some delimiters;
146: A list of plain tex delims;
147: the bigs;
148-149: details for use of left right;
150: still more extensible chars (bottom not on growing delims);
171: info on the type of atoms made by big bigr bigl bigm;
437: openings and closings;
442: The search of the appropriate delimiter: good.

About radicals rule 11 page 443 appendix G. \TeX book

\TeX book: About operators rule 13 page 443, successors. Interesting
about italic corrections. The way limits are typeset rule 13a.

About parameter usage p447 \TeX book.
About math spacing p170 \TeX book.

\TeX\ for the impatient: p.126 nothing, p.194-196 a list of operators
and a few explanations.

\TeX\ by topic p.194: Large operators and their limits, the choosing
method: good.
\end{filecontents}



\begin{filecontents}{l3d007e.tex}
\begin{quote}\small
% jz -- a little change in the first sentence
  While working on the new math encoding, the writer realised that
  the fact the \texttt{cmex} font is only loaded in one size, and not in
  three like the other math fonts, was going to create a few problems.
  This paper deals with the following topics:
  \begin{itemize}
    \item What is in \texttt{cmex}?
    \item Which special mechanisms does \TeX\ use to access glyphs from
      \texttt{cmex}?
    \item What could be added to \texttt{cmex}?
    \item What could be taken out of \texttt{cmex}?
  \end{itemize}
  The aim of this paper is to help the MFG\footnote{Math Font Group.}
  design the MX encoding as a replacement and improvement of the
  \texttt{cmex} encoding.  \note{Most of what is written in this paper
    is pure theory, and has not been applied or tested.}

  Acknowledgements: thanks to Alan Jeffrey and Barbara Beeton for
  their constructive comments, and help.
\end{quote}

% bb -- suggest adding an introductory section 0 giving a bit of
%       background on knuth and early history; i would make suggestions
%       and vouch for the accuracy of the statements, or offer to be
%       quoted as the source if something is conjecture.

\section{What is in \texttt{cmex}?}
  \subsection{Delimiters}
  \begin{itemize}
  \item Four different sizes of ( ) and extensible versions. Left and
  right extensible modules are '102 and '103.
  \item Four different sizes of [ ] and extensible versions. The
  extensible  modules, one for the right bracket, and one for the left
    bracket are: '066, '067.
  \item Same for \{ and \}\,; the extensible module is: '076.
    \note{The extensible module here is very small, because it is
      added twice: once above the middle piece, and once below the
      middle piece. Its height is half that of the parentheses'
      extensible module. Interesting to see that there is only one
      extensible module for both the left and the right curly brace.
      This is because the left-right extension of a curly
      brace is symmetrical, unlike the parentheses for example.}
  \item Four different sizes of $\langle$ and $rangle$. No extensible
  version.
  \item Same for $\backslash$ and $/$. No extensible version.
  \item Four different sizes of $\lfloor$ and $\rfloor$ and an
    extensible version. Extensible modules: '066, and '067.
  \item Same for $\lceil$ and $\rceil$. Same extensible modules as the
    previous one.
  \item Glyphs in positions '014 and '015 are the extensible versions
    of the vertical bar and the double vertical bar. They are their
    own extensible modules.
  \end{itemize}

  \subsection{Large operators}
  Large operators come in pairs:
  \begin{itemize}
    \item The sqcup
% bb -- although the sqcap isn't here, might want to note that it
%       isn't but could be added
    \item The circled integral
    \item The circled dot
    \item The circled plus
    \item The circled times
    \item The sums
    \item The prods
    \item The normal integrals
    \item The bigcups
    \item The bigcaps
    \item The U plus
    \item The wedges
    \item The vees
    \item The coprods
  \end{itemize}

  \subsection{Wide accents}
  \begin{itemize}
    \item 3 sizes for the hat
    \item 3 sizes for the tilde
  \end{itemize}

  \subsection{Radicals}
  \begin{itemize}
    \item Five radical signs
    \item The vertical bit needed to construct the big radical: '165
    \item The top bit of the constructed radical: '166
  \end{itemize}

  \subsection{Arrows}
  \begin{itemize}
  \item The three pieces for the construction of the vertical double
    arrow: '167 '176 '177
  \item The three pieces of the vertical single arrow: '077, '170,
    '171
  \end{itemize}

  \subsection{Horizontal curly braces}
  \begin{itemize}
  \item The four pieces for the construction of horizontal curly
    braces: '172 -- '175
  \end{itemize}







\section{\TeX's behavior with \texttt{cmex} glyphs}
  \subsection{Large operators}
  \begin{itemize}
  \item A large operator is vertically centered with respect to the
    math axis. This means that, whatever the surrounding glyph size,
    things will not look too bad.
  \item With the following definition of a large operator:
    \cn{mathchardef} \cn{sum} \texttt{= "1xyy}, if \TeX\ is in \textem
    {display style}, it looks to see if the character in position
    \texttt{"yy} of family \texttt{x} has a successor. If it does then
    the successor is taken. When not in \textem {display style}, \TeX\
    simply takes character \texttt{"yy} from family \texttt{x}. Thus in
    text style, in script style and in scriptscript style the same
    glyph is used.
  \end{itemize}

  \subsection{Vertical delimiters, and friends}
  Radicals are delimiters, and vertical arrows also, so let us only
  speak about delimiters. Here is a quote from Victor Eijkhout's book:
  \begin{quote}
    A delimiter has two codes: a small variant, and a large variant.
    \TeX\ first tries the small variant, and if that is not
    satisfactory (or if the left part of the delimiter code is 000) it
    tries the large variant. If trying the large variant does not meet
    with success \TeX\ takes the largest delimiter encountered in this
    search. If no delimiter at all is found, (which can happen if the
    right hand part is also 000), an empty box of width
    \texttt{\string\nulldelimiterspace} is taken.

    Investigating a variant means in sequence:
    \begin{itemize}
    \item If the current style is scriptscript style, the scriptscript
      font of the family is tried.
    \item If the current style is script style or smaller the script
      font of the family is tried.
    \item Otherwise the text font of the family is tried.
    \end{itemize}

    Looking for a delimiter at a certain position in a certain font
    means:
    \begin{itemize}
    \item If the character is large enough, accept it.
    \item If the character is extensible accept it.
    \item Otherwise, if the character has a successor (the same but
      bigger), try the successor.
    \end{itemize}
  \end{quote}
  Using the three size mechanism probably did not seem necessary to
  Knuth.  Generally large delimiters are used in display style, and not
  in script or scriptscript style. However, they can also be used in
  the small styles.

  \subsection{Wide accents}
  For the choice of accents, \TeX\ only considers one font, but looks
  to see if the current accent has a successor. Unlike the delimiter
  choice mechanism, the accent choice mechanism does not go through all
  three sizes. \TeX\ chooses the accent in such a way that the accent
  width is as close as possible but smaller than the width of the box
  to cover.

  \subsection{First conclusion}
  For operators, a distinction in size is made between text style and
  display style, whereas with the usual automatic size choosing
  mechanism, glyphs in text style and in display style are taken from
  the text size font, and are therefore the same size.



\section{Consequences of loading \texttt{cmex} in 3 different sizes}

  \subsection{Consequences for operators}
  For operators, let us consider two completely separate policies. In
  the one  case \texttt{cmex} is unchanged, and loaded in three
  sizes. In the other case, an imaginary font derived from \fn{cmex}
  called \texttt{cmex}$'$ is loaded in three sizes.  In \fn{cmex}$'$ the
  operators no longer have two sizes, thus glyphs like \cn{bigcup} do
  not have a successor in their font.

        \textem {All the following supposes that no new macros have been
  written. What would \TeX's automatic behavior would be?}

  \begin{description}
  \item[If \texttt{cmex}$'$ is loaded in three sizes.] In
    such a situation \TeX\ has a large version of \cn{bigcup} (and
    other operators) in text size, a small version in script size, and
    yet a smaller version in scriptscript size.

    The operators can be centered: no problem.

% jz -- change:
    In script and scriptscript style the resulting `big operators' would
    be smaller than if they were produced with today's standard \TeX,
    and today's standard \texttt{cmex}.  What is more, one would be
    smaller than the other, which is also not the case with today's
    standard \TeX, and today's standard \texttt{cmex}.
% ----

    In display style one would get big operators from the text size
    font: this is acceptable.  \textem {But one would also get a big
      operator in text style,} and that does not conform with today's
    standard \TeX\ behavior.


% jz -- I added encoding:
  \item[If the existing \texttt{cmex} encoding is loaded in three
  sizes.]
    The operators can be centered: no problem.

    In script style, one would get the smallest version of a large
    operator. But coming from a small size font, that will produce
    something very small. In scriptscript style, same behavior as in
    script style, but the result would be even smaller.
% jz -- Added this:
    Thus in script, and in scriptscript style, the large `big
    operators' would never automatically be used.
    Hence the \texttt{cmex}$'$ encoding.
% -- maybe should have said that earlier :-)

    In text style, \TeX\ would produce the small version taken from
    the text size font. In display style \TeX\ would produce the big
    version of operators taken from the text size font.

    So in text style and in display style, there would be no change
    compared to what today's standard \TeX\ produces.
% jz -- changed the phrasing here:
    But script and scriptscript style would produce different
    results.
% --

  \end{description}

  In both cases, things could be improved if macros were written to
  override the present behavior of \cn{bigsqcup}. One could think of
  things like \cn{mathchoice}, but $\ldots$



  \subsection{For vertical delimiters, radicals, vertical arrows}
  Let us start by supposing \TeX\ is in scriptscript style, and it has
  to typeset a large delimiter. One should consider two cases:
  \begin{description}
  \item[The delimiter has an extensible variant.]  In this case the
    search will start in scriptscript size, and continue until \TeX\
    finds the extensible variant of scriptscript size. Then the search
    will stop, and the extensible will be used.  This extensible will
    come from scriptscript size, and therefore probably not look the
    same as it would in today's setup, where all extensibles come from
    text size.
  \item[The delimiter does not have an extensible variant.]  As
    previously, the search starts in scriptscript size. If nothing
    big enough is found in scriptscript size, the search continues in
    script size. If still nothing is found, the search then continues
    in text size. If necessary the biggest delimiter from text size
    will be used. If the search stops in text size, there is no
    difference with what \TeX\ produces today. But if the search stops
    before reaching text size, the chosen delimiter will be different
    from the one \TeX\ would use in the present configuration. Its
    strokes would be finer, and better adapted for use in script style.
  \end{description}

  If one supposes that \TeX\ is in script style, the previous two
  cases also apply, except that every occurrence of `scriptscript' must
  be replaced with `script'. If one supposes that \TeX\ is in text
  style, the result of loading three different sizes of \texttt{cmex}
  would be the same as it is in \TeX's current configuration.

  \subsection{For horizontal curly braces}
  If they are automatically taken from script size, or from script
  script size when necessary, the spacing changes a little,
  because the dimensions in the \texttt {.tfm} files would be
  different. A consequence of this could be different line and page
  breaks.

  However, it would be nice if curly braces did come out
  of the correctly sized fonts. Then their boldness would match the
  surrounding text. But apparently from a macro programming
  point of view things could be difficult, even if the glyphs are
  available and loaded.

  \subsection{For wide accents}
  See first paragraph of previous section.

% jz --
%  I guess that in actual fact, the wide accents were put in
%  \texttt{cmex} because they could not fit any where else.
% jz -- I don't want to start any discussion on that topic so i'm
% taking it out

  If accents were taken from the current size, things could only look
  better.
  The accent width would be closer to that of the material under the
  accent, and the accent's boldness would be better adjusted.

  \note {Unlike the delimiter choice mechanism, the accent choice
    mechanism is restricted to one font, and one size.  It will thus
    not look in text size when it is in script size for instance. So
    in script style, accents will always come out of the script size
    font, and in scriptscript style, accents will always come out of
    the scriptscript size font, etc.$\ldots$}

  \subsection{Conclusion}
% jz -- Changed the conclusion
  Nearly everything in \texttt{cmex} could have lived in a normal three
  sized math font, and maybe that would have been better. The only
  problems would have come from the specific ``big operator'' behavior
  required by Knuth.

  Also one must not forget that Knuth did not want to leave any empty
  slots.
% jz -- added this:

  The reduced amount of memory that was available on the machines with
  which \TeX\ was first used could have been another reason for loading
  \texttt{cmex} in one size only.






% jz -- This section has been changed a bit, and extended.

\section{What could be added to \texttt{cmex}?}
Let us now consider possible evolutions of \texttt{cmex}.
In spite of the terminology ``adding to \texttt{cmex}'', the font
resulting from these evolutions would have a different name.
  \subsection{If the \texttt{cmex} encoded font is loaded in three
  sizes}
  In this case big operators would not produce the usual results, and
  the rest would be slightly different, as stated above.
    \begin{itemize}

    \item One could add wide accents, but one would get slightly
      different (better) results. Thus wide accents would match the
      script and scriptscript styles. Macros could be made available
      as a style option to keep the old behavior, if necessary.

    \item One could increase the number of different sizes for accents.

    \item One could add big delimiters and their extensible versions,
      without any problem! Things will be slightly
      better adjusted in script and scriptscript style. Macros could
      be made (available as a style option) to keep the old behavior,
      if necessary.

    \item One could increase the number of different sizes for
      delimiters, and one could probably reduce the height of the
      extensible module in order to make the growing of delimiters
      more gradual.

    \item One could add some vertical extensible arrows!  Things will be
      slightly better adjusted in script and scriptscript style.

    \item One can add small and large `big operators' without any
      problem!

    \item Big improvement: one could add loads of other glyphs (symbols,
      etc.$\ldots$) that would come in all three sizes.

    \end{itemize}

  \subsection{If the \texttt{cmex} encoded font is only loaded in one
  size}
    \begin{itemize}
    \item One could add big delimiters and their extensible versions
      without any problem!

    \item One could increase the number of different sizes for
      delimiters, and one could reduce the height of the extensible
      module in order to make the growing of delimiters more gradual.

    \item One could add some vertical extensible arrows!

    \item One could add large operators without any problem!

    \item One could add wide accents without any problem, and the
    present behavior of wide accents would not change. But if wide
    accents are meant to match the script and scriptscript styles, then
    wide accents must go in another font that would be loaded in
    different sizes.

    \item One could increase the number of different sizes for accents.

    \item One could add other stuff, but it would only come in one
      size.

    \end{itemize}

  \subsection {If a \texttt{cmex}$'$ encoded font is loaded in three
      sizes}
        The imaginary \texttt{cmex}$'$ encoded font, previously
        described in this document is considered here.

  One could add the same things as when \texttt{cmex} is loaded in three
  sizes. The only difference is: if no macro programming were done, the
  text style and display style will produce the same `big operators'.
  In script and scriptscript style the `big operators' would be in
  different sizes from one another and smaller than those in text
  style.






\section{Conclusions}
If one loads \texttt{cmex} in three different sizes, many things are
improved, and with a \cn{mathchoice} the initial behaviour of large
operators could be kept, or available as a style option.

If \texttt{cmex} is kept in a single size, one must decide whether to
put wide accents in or not.







\section{The beginning of my \texttt{cmex10.pl} file}

\begin{verbatim}
(FAMILY CMEX)
(FACE O 352)
(CODINGSCHEME TEX MATH EXTENSION)
(DESIGNSIZE R 10.0)
(COMMENT DESIGNSIZE IS IN POINTS)
(COMMENT OTHER SIZES ARE MULTIPLES OF DESIGNSIZE)
(CHECKSUM O 37254272422)
(FONTDIMEN
   (SLANT R 0.0)
   (SPACE R 0.0)
   (STRETCH R 0.0)
   (SHRINK R 0.0)
   (XHEIGHT R 0.430555)
   (QUAD R 1.000003)
   (EXTRASPACE R 0.0)
   (DEFAULTRULETHICKNESS R 0.039999)
   (BIGOPSPACING1 R 0.111112)
   (BIGOPSPACING2 R 0.166667)
   (BIGOPSPACING3 R 0.2)
   (BIGOPSPACING4 R 0.6)
   (BIGOPSPACING5 R 0.1)
   )
(CHARACTER O 0 ...
\end{verbatim}



\subsection{Comments about the \texttt{cmex10.pl} file}
\begin{itemize}
\item The xheight is not equal to zero.
\item The space is equal to zero.

\item With the following:
\begin{verbatim}
(CHARACTER O 100
   (CHARWD R 0.875003)
   (CHARHT R 0.039999)
   (CHARDP R 1.760019)
   (VARCHAR
      (TOP O 70)
      (BOT O 73)
      (REP O 76)
      )
   )
\end{verbatim}
that is in the \texttt{.pl} file, one can produce something that looks
like a growing integral:
$$ \left\lmoustache \frac{3.q}{\frac{3\pi.r^2}{3.q.b.c}} \right.$$

\item The pieces used to construct the horizontal curly braces are not
linked in any way.

\item The bottom pieces of the extensible parentheses are overloaded
  for \cn{rmoustache} and \cn{lmoustache}. One of these could be linked
  (charlisted) with the integrals, so that \cn{left}\cn{bigint} could
  produce a growing integral like the delimiters.

\item The bottom pieces of the curly braces ('072 and '073) are also
  overloaded for \cn{lgroup} and \cn{rgroup}.

\item The middle pieces of the curly braces are overloaded for
  \cn{arrowvert} and \cn{Arrowvert}. Other single and double
  extensible bars with different spacing.

\item The extensible module of the curly braces is overloaded for
  \cn{bracevert}. A fat vertical bar.

\item The extensible modules of the parenthese are overloaded to
  produce more fat vertical bars.

\item More overloading: the construction pieces of the extensible
  brackets are also used for the $\lceil \rceil \lfloor \rfloor$. Thus
  the top left bracket piece ('062) identifies the left bracket; the
  top right bracket piece identifies the left bracket; the
  bottom left bracket piece identifies the $\lfloor$ extensible
  version; the bottom right piece identifies the $\rfloor$; the left
  extensible module identifies the $\lceil$; and the right extensible
  module identifies the $\rceil$. This over-loading may not be desired.

\item For the wide accents and the curly braces the depth is nil.

\item All the delimiter glyphs in \texttt{cmex} are set with a very
  small height and a big depth. This is because the radical
  primitive is also used for delimiters. For radicals, the \texttt
  {.tfm} height of the glyph is used to determine the size of the
  \cn{hrule}.

\item The extension modules do not have any height at all. Same for
  the arrow heads.

\item The four integrals have italic corrections.

\item Small versions of operators have a nil height, whereas
  big versions have a small height and a big depth:
  \begin{verbatim}
  (CHARACTER O 116
   (COMMENT This is the small \bigotimes)
   (CHARWD R 1.1111145)
   (CHARDP R 1.000013)
   (NEXTLARGER O 117)
   )
  (CHARACTER O 117
   (COMMENT This is the big \bigotimes)
   (CHARWD R 1.511116)
   (CHARHT R 0.100001)
   (CHARDP R 1.500012)
   )
  \end{verbatim}
  What is more, in the metafont code,
  both big and small versions of bigops are under the baseline.

\item There are no kerns or ligatures in \texttt{cmex}.

\end{itemize}








\section{Characters under the baseline}

\subsection{Which?}
It is understood and agreed that the radical glyphs need to be
virtually completely under the base line. \TeX\ uses their small
height to measure the thickness of the radical's over line. But
delimiters and `bigops' are also placed under the baseline for no
obvious reasons: both types of glyphs are just centered on the maths
axis.

\subsection{`Bigops' and metafont code}
\begin{description}
\item[The metafont padded operator.]
Although both big and small versions of bigops are placed completely
under the baseline (height=0), the big version ends up ---~in the
\texttt {tfm} file, with a non-nil height. Many people agree that the
reason for this is that the metafont code for the large version of
`bigops' contains the \cn {padded} macro. This last places some extra
space around the glyphs.  This extra space would serve for separating
the `bigops' from the limits they may take.

Large integrals do not have any padded macro, and thus in the \texttt
{tfm} file, they are placed completely under the base line.  The
reason for the difference between integrals, and other `bigops' could
be that the limits of the former are usually placed next to the glyph,
and not on top. Even when limits are placed on top of the integral,
the results do not look too bad, because the integral is very narrow.

The reason for `bigops' being set under the baseline is still unknown.
Whether or not they would be correctly centered on the math axis,
if they were placed over the base line is not sure either.

\item[Changing the metrics in the metafont code.] It is interesting to
  see how \TeX\ would manage if the `bigops' were placed over the base
  line.
  The best way to find out, is to change the metafont code of \texttt
  {cmex}, and see$\ldots$ I first did the test on \cn{bigoplus}
  because it was a good candidate ---~simple metafont code. I have
  never used metafont before. I copied all the necessary files into my
  directory, and changed their names to `\texttt {my*}'. I then did
  the following modifications in \texttt {mybigop.mf}
  (look for \%\%\%\%):
  \begin{verbatim}
    cmchar "\textstyle circle-plus operator";
    beginchar(oct"114",20u#,10/6dh#,0); %%%% was ,0,10/6dh#)
    adjust_fit(0,0); pickup pencircle scaled stem;
    lft x6=hround u; x2=w-x6; top y8=h; bot y4=-d; %%%% was top y8=0
    ...

    cmchar "\displaystyle circle-plus operator";
    beginchar(oct"115",27.2u#,14/6dh#,0); padded 1/6dh#;
         %%%% was ,0,10/6dh#)
    adjust_fit(0,0); pickup pencircle scaled curve;
    lft x6=hround u; x2=w-x6; top y8=h; bot y4=-d; %%%% was top y8=0
    ...

  \end{verbatim}

  When I ran Metafont on it there were no problems. After having put
  all the relevent files where they were meant to go, I gave the
  following to \TeX:

  \begin{verbatim}

     Old cmex in text style: $\bigoplus i \coprod$
     Old cmex in display style: $$-\bigoplus i\coprod \mathchar"034D$$

     % change font:
     \font\myfont=myex10
     \textfont3=\myfont
     \scriptfont3=\myfont
     \scriptscriptfont3=\myfont

     New myex in text style: $\bigoplus i \coprod$
     New myex in display style: $$-\bigoplus i\coprod \mathchar"034D$$
     The minus sign gives the height of the math axis, while the bottom
                 of the word `base' gives that of the baseline.

     \bye

  \end{verbatim}

  Here is the output:

     Old cmex in text style: $-\bigoplus i \coprod$ \\
     Old cmex in display style:
         $$-\bigoplus i\coprod \mathchar"034D base$$

     % change font:
     %\font\myfont=myex10
     %\textfont3=\myfont
     %\scriptfont3=\myfont
     %\scriptscriptfont3=\myfont

     New myex in text style: $-\bigoplus i \coprod$ \\
     New myex in display style:
         $$-\bigoplus i\coprod \mathchar"034D base$$
     The minus sign gives the height of the math axis, while the bottom
                 of the word `base' gives the baseline.

  The results are quite obvious: in both case the \cn{bigoplus} is
  correctly centered. On the 300 dpi printer I have here there is no
  visible difference. What is more the \cn{mathchar"034D} proves that
  the metrics of the \cn{bigoplus} have changed: in the first case the
  glyph is placed under the baseline, and in the second case it is
  placed over the baseline. The difference is not visible if this
  document is printed with the wrong fonts.
\end{description}
\end{filecontents}



\documentclass{l3ms002}

\usepackage{l3d007}
\usepackage{ifthen}

\setlength{\emergencystretch}{2em}

\hyphenation{pre-sent fa-mi-ly}

\renewcommand{\abstractname}{\Large Acknowledgement}


\typeout{******************************************}
\typeout{* }
\typeout{* This document makes use of three fonts}
\typeout{* which you might not have on your system.}
\typeout{* These are:}
\typeout{* \space\space ecrm1000 \space\space T1 encoded Computer Modern}
\typeout{* \space\space msam10   \space\space AMS symbol font A}
\typeout{* \space\space msbm10   \space\space AMS symbol font B}
\typeout{* }
\typeout{* In that case type <RETURN> to the error message}
\typeout{* The output will not contain the font charts but}
\typeout{* but will otherwise be readable.}
\typeout{* }
\typeout{* The document will also produce a number of underfull}
\typeout
       {* and overfull boxes. Please ignore them (or volunteer to help)}
\typeout{* }
\typeout{******************************************}

\newcounter{sleep}
\whiledo{\value{sleep}<1000}{\stepcounter{sleep}}

\begin{document}





\title{Technical Report on Math Font Encoding}
\author{Justin Ziegler}
\date{Started on June 13, 1993\\
                        Last change: June 1, 1994\\
                        Organisational updates: August 23, 2000\\
      Printed: \today \\
      Filename: \fn{l3d007.tex} }

\maketitle



\newcommand{\NFSS}{\textsf{NFSS}}

\chapter*{Foreword}

I'm pleased to present the final report on ``Math Font Encoding''
produced by Justin Ziegler for the \LaTeX3 project to the public.

Justin has worked for three months at the Johannes Gutenberg
University Mainz. His work was generously sponsored by GUTenberg
(The French \TeX{} Users Group) and by the ZDV of the University
of Mainz (Data Processing Center), the latter providing Justin with
office space and taking care of the administrative details.

In the past years a lot of work went into integrating  new fonts into
the \TeX{} system. Only five years ago, typesetting with \TeX{}
basically meant typesetting in Computer Modern. Nowadays many users can
choose  (at least theoretically) from several thousands of fonts.
Today, \NFSS{} is the standard font selection in \LaTeX{} and due to
this mechanism and the fontinst-package by Alan Jeffrey virtually every
PostScript font, in fact, every font for which a \texttt{tfm}-file can
be obtained, can be used, out of the box, with \LaTeX.

But for these thousand text fonts there are only five font families for
use in math formulas
to go with them. Even worse, every of these math font sets are encoded
in a different way making it nearly impossible even for an expert \TeX{}
user to use different fonts for math in different jobs.

The work undertaken by Justin is the first of several steps to solve
the problems at hand, the final goal being the development of a system
that allows the user
to change math fonts as painlessly as it is now possible with text
fonts.

Based on Justin's analysis and his proposal, the \LaTeX3 Project is now
undertaking to provide a prototype implementation for math fonts,
starting with the Computer Modern fonts as well as the Euler Math fonts.
We expect this implementation to be available for public usage
during 1995.


\begin{flushright}
Mainz, December 6, 1994\\[5pt]
Frank Mittelbach\\
Technical Director \LaTeX3 Project
\end{flushright}




\begin{abstract}

\vskip1cm
  I wish to thank the many people without whom my stay in
  Germany would not have been possible, and the work I did would not
  have been done. This includes:

  \begin{description}
  \item[GUTenberg] who financed my stay;

  \item[Ehoud Ahronovitz] for helping me with the administrative side
    of things, for giving me the opportunity of coming here, and
    spending extra time with me to make sure that everything went
    all right;

  \item[Frank Mittelbach] for his friendly welcome, for the
    organization, time and guidance;

        \item[Bernard Gaulle] the past president of GUTenberg, for the
                organization and logistics;

  \item[Klaus Merle] for lending all the material that I used;

  \item[Chris Rowley] for the organization, and help;

  \item[Stefan Steffens] for answering patiently all my stupid
    questions, and helping me integrate Mainz and the university;

  \item[The \LaTeX3 project] which partially financed my stay in England
    for the Aston conference;

  \begin{sloppypar}
  \item[Barbara Beeton and Alan Jeffrey] who commented my papers, and
    answered more stupid questions;
  \end{sloppypar}

  \item[J\"org Knappen] who gave me advice on the project, and
    with whom I discovered the Mainz night life;

  \item[All the computer center employees] for making my stay more
    pleasant;

  \begin{sloppypar}
  \item[All the people who took the trouble to answer my mail,] for
  their time  and effort;
  \end{sloppypar}

  \item[Donald E. Knuth] who created \TeX.

  \end{description}

\end{abstract}


\setcounter{tocdepth}{1}
\tableofcontents




%                                This is the introduction




\chapter{Introduction / Overview}

\section{The technical environment}
I worked in the ZDV of the university of Mainz in Germany.  In German
ZDV stands for ``Zentrum f\"ur Daten Verarbeitung''. Which means: Data
Processing Center.
This is where the main --- soft and hardware --- maintenance team
works.

I worked on an X-terminal like a lot of other people in the university.
For writing my documents I used GNU Emacs together with the
AUC\TeX\ package.



\section{A few Definitions}%    ***************** first section
        %jwzinria: one could get rid of this subsection ?
  \subsection{\TeX: a page description language}
  The best definition I can find for \TeX\ is: ``one third compiler,
  one third interpreter, and one third word processor''. It was
  written in $\fam 1 1975$ by D.~E.~Knuth and a group of students.
  One of its main features is its portability. A document written
  on one machine can be used on another machine.  Knuth also
  insisted that \TeX\ would not change. So a document written in $\fam
  1 1980$
  is still usable in $\fam 1 1990$.

  The language defined by \TeX\ is very specific, in so far as it is
  designed to describe a page layout. \TeX\ processes the page like a
  rectangle, or more exactly like a box, that can be filled with
  smaller boxes. These smaller boxes can similarly be filled with
  smaller boxes, and etc... The smallest box one can get is a
  vertical~/~horizontal line, or a character (a glyph), or just some
  space. \TeX\ has variables in which one can put boxes, or different
  types of numbers. One can define functions ---~usually called
  macros~--- in a way similar to lisp. The if-then-else statement is
  there, and combined with recursion it can be used to make loops.

  In spite of its limitations due to its specificity, \TeX{} defines
  a Turing machine. The syntax is very disagreeable, but one can get
  used to it: somebody wrote a basic interpreter in \TeX. The only
  difference between \TeX{} and a usual compiler, is that \TeX{} stops
  the compilation when it gets to the pcode, and just puts it into a
  file.  This file, called the device independant file, can then be
  sent to a printer, a screen, or any other printing device.

  Today many people use \TeX. All \TeX\ users have got together and
  created TUG: \TeX\ Users Group.

  \subsection{Plain \TeX} \label{plain-tex}
        %jwzinria: one could get rid of this subsection ?
  Plain \TeX\ is the standard set of macros and definitions that comes
  with \TeX. It is written in \TeX.

  \subsection{\LaTeX{}: a document description language}
        %jwzinria: one could get rid of this subsection ?
  Just as \TeX\ is a language to describe pages, \LaTeX\ is a
  language designed for describing whole documents, and their logical
  structure. The idea is that it lets the user concentrate on the
  contents of the document rather than the formating commands
  necessary for the document to look good. Thus it uses the
  logical mark-up concept. It was written by Leslie Lamport in 1985.
  Technicaly, \LaTeX\ is ``only'' a cluster of macros written in \TeX.
  This means that a \LaTeX\ user has still got access to most of the
  \TeX\ language.  \LaTeX\ includes the following facilities:
  \begin{itemize}
       \item Cross referencing.
       \item Automatic construction of a table of contents.
       \item Automatic construction of an index.
       \item Bibliography referencing.
       \item Basically the same math mode as \TeX.
  \end{itemize}



  \subsection{The \LaTeX3 project}
        %jwzinria: one could get rid of this subsection ?
  During the 1989 TUG conference at Stanford, the decision was taken
  to produce an improved and expanded version of \LaTeX, that was
  going to be called \LaTeX3.

  The major difference in the new version will be the addition of a
  good interface through which designers can specify how classes of
  documents should be formated.

  Frank Mittelbach is the technical director of the project; he and
  Chris Rowley are responsible for the management.


  \subsection{Fonts, glyphs, and slots}
  \TeX\ would not be able to produce any nice documents if it did not
  have any fonts. One cannot get a nice looking `A' or `$\mathcal{A}$',
  or any other letter if nobody has previously designed it.

  All \TeX\ really does, is produce a file that contains a set of
  instructions. Each instruction looks like the following: ``place
  here the picture that is in such and such a file, in position number
  $x$.'' \textem{The files that contain all the pictures (the letters
    and other symbols), are called ``fonts''. All the pictures that are
    in a font are called ``glyphs''.} Every glyph in a given font has
  a specific and known position. \textem{I shall use the word ``slot''
    to refer to a given position in a font.} Some slots can be empty,
  but most of them contain a glyph.




  \subsection{Font encodings} \label {about-encoding}
  When \TeX\ refers to the glyph number $x$, it must know which glyph
  is in position number $x$. This knowledge is contained in the
  encoding.  In some cases one could say that the letters are in the
  ASCII order. But this is not sufficient, because the
  ASCII code does not include all the glyphs that people wish
  to put in their documents. Therefore, one must link every single
  font with a given encoding, and make the encoding known by \TeX.
  Many different encodings exist, sometimes even for the same group
  of glyphs. But there are also many fonts that use the same encoding.

  A mathematical definition of an encoding could be the following:
  \textem{An encoding is a set of glyph names in a given order.}

  \subsection{The ``Computer Modern Fonts''}
  When D.~E.~Knuth created \TeX, he also created a set of fonts called
  the
  \textem{Computer Modern Fonts}. Most of them were based on an encoding
  that is called the Computer Modern Encoding throughout this document.

        All file names of Computer Modern Fonts start with the two
        letters `\texttt{cm}'.


  \subsection{Metafont: a font description language}
  Metafont is a language / program especially designed to describe glyph
  shapes, and more generally whole fonts.  It was used to generate all
  the  ``Computer modern'' fonts. The Metafont user must describe or
  ``program'' the curves for each glyph. Then Metafont produces an array
  of black and white dots for each glyph. The dots can be made as small
  as necessary to fit the precision of the printing device.


  \subsection{\TeX\ version 3}
  In the beginning of $\fam 1 1990$, under a lot of pressure (from the
  \TeX\ User Community), D.~E.~Knuth produced a new version of \TeX.
  \TeX\ version 3 was born. The main improvements were the following:

  \begin{itemize}
  \item Up to 256 glyphs per font. The previous versions of \TeX\ could
    only use the first 128 glyphs of a font.


  \item Virtual fonts. A normal font has all its
    glyphs in a file, and this file is in actual fact the font.
    Virtual fonts enable people to group 256 glyphs taken from many
    different fonts, and make \TeX\ think it is using one normal font.
    For instance, one could make a virtual font with lowercase letters
    in bold, and uppercase letters in italic. The user would work as
    if he was using one font, but the results would in actual fact be
    a combination of two fonts. A very good example implementation of
    virtual fonts is the creation of ``Small Caps'' fonts: the
    uppercase letters could come from a roman upright font at 12
    points, whereas the lowercase ones could come from a roman
    upright at 10 points.

    Virtual fonts enable still more ingenious things, like replacing
    glyphs with a set of \TeX\ macros.  One can then
    consider, for example, automatic raising or lowering of some
    letters.

  \item Better hyphenation. \TeX\ version 3 can have up to 256
    different hyphenation tables, and can produce good automatic
    hyphenation even when a word contains accents. The latter was not
    possible in previous versions. More generally the hyphenation
    mecanisms have been improved.

  \item The new ligature mecanism is more powerful. The result of a
    ligature is no longer only one glyph, but can be a set of
    glyphs...

  \item Special ligatures can be done at the beginning and at the end
    of words. Thus when a given letter is at the end of a word, its
    shape can be different from the shape it would have in the middle
    of a word.

  \item Better automatic adjusting of interword space.

  \item More little details that make everybody happy...
  \end{itemize}


  \subsection{The ``DC Fonts''}
  Although D.~E.~Knuth included a lot of ``European glyphs'' in his
  Computer
  Modern fonts, more were needed. In $\fam 1 1989$ \TeX\ users got
  together in Cork, and designed some new fonts called the \textem{DC
  Fonts}.
  Thanks to the new features  of \TeX\ version 3 (256 glyphs per font
        encoding),
  DC fonts included for example more special letters for Catalan and
  Scandinavian languages.

        The DC Fonts used what is now called the \textem{Cork encoding.}
        All DC  fonts file names start with `\texttt{dc}'.


\section{My work}
%\section{My work}
One of \TeX's nicest features is its ability to typeset mathematical
formulae. There has now been over ten years of experience typesetting
mathematical material with \TeX.  During this time, \TeX's math mode
has been used to set a wide variety of material, including traditional
mathematics, categorical diagrams, chemical reactions, computer
programs and textual material such as `$5\frac12\%$ or $\fam0M^{lle}$.

In recent years, with the arrival of the Cork standard for typesetting
European text, and the Virtual Font standard, the fonts available for
use in \TeX\ have radically changed. The current situation is that
there are over 14,000 text fonts available for use in \TeX, but only
five math fonts:
\begin{itemize}
\item Computer Modern
\item Computer Concrete with Euler
\item Lucida Math
\item Lucida New Math
\item Math Time
\end{itemize}
Each of these fonts use different encodings, and each comes with its own
selection of \TeX\ macros.

Although the Cork encoding is rapidly being established as the
standard encoding for European Latin text, there is no similar
encoding for mathematics.  The result is:
\begin{itemize}
\item complex macro packages for using each math font.
\item it is difficult to set mathematics with Cork text, since the Cork
   encoding does not include the uppercase Greek.
\item installing PostScript math fonts such as Mathematical Pi is very
   difficult.
\end{itemize}

Furthermore, the present math encoding includes glyphs like old-style
digits, and game card suits $(\spadesuit)$ that just do not belong in
a math encoding. On the other hand, many new glyphs have been
designed and should be included in the math encoding.

To solve these problems, a new math encoding, using all the power of
\TeX\ version 3, is needed. For this reason I have been trying to
re-organize all the glyphs that are needed to typeset mathematical
formulae with \TeX, according to various technical constraints.

The new math encoding that I am helping to produce is hopefully going
to be part of the \LaTeX3 package, and comes as one of the general
improvements of \LaTeX.

First I learnt to use \TeX. In a second stage, I had to study and
understand the technical constraints that apply on the grouping of
mathematical glyphs in a font. Only then could I actually start
thinking about which glyphs should go where. I intensively used \LaTeX\
---~so that I permanently had an up to date record of what had been
done ---~and email, to communicate with the people I
was working with.


















\chapter{The \TeX nicalities of math typesetting}



\section{A brief description of \TeX's math facility}
  \begin{description}
  \item[Logical markup like \LaTeX.] For the design of \TeX's user
    interface, one of Knuth's concerns was that in the source code of
    a mathematical document the formulae should be readable in a
    linear manner. Thus when a mathematician thinks, he says to
    himself: ``$n$ over $n-1$'', and when a \TeX nician works, in order
    to produce the result $\frac{n}{n-1}$ he just has to type:
    \verb|n \over {n-1}|.

    The user is no longer bothered by trying to get this bit of text
    higher than this other bit of text.  He just gives \TeX\ the
    logical meaning of what should be typeset, and it is correctly
    placed.

  \item[The two math modes.] There are two ways to enter \TeX's math
    mode, which produce slightly different results with the same
    input. One mode is called the \textem{display mode,} and produces
    \textem{display style,} while the other is called \textem{text
      mode,} and produces \textem{text style.} The following input:\\
    \verb|    $ \int_0^1 \frac{1}{x}\;dx $| \\ produces
    \textem{text style:} $\int_0^1 \frac{1}{x}\;dx$, which can be
    mixed with text, whereas \\
    \verb|    $$ \int_0^1 \frac {1}{x} \;dx $$| \\ produces \textem
    {display style:}
    $$ \int_0^1 \frac {1}{x} \;dx $$ which is automatically
    centered and surrounded by space.

  \item[Automatic size change according to meaning.] When the user
    says to \TeX: ``this letter is a superscript'', or ``this number is
    a subscript'', \TeX\ automatically typesets the letter (or the
    number) in a smaller font size. \TeX\ does that same size adjustment
    for setting limits on glyphs like $\sum$, or $\smallint$.

  \item[Automatic placing for sub/superscript and for limits.] At the
    same time as \TeX\ changes size automatically when the user
    specifies a sub- or superscript, \TeX\ also raises and lowers the
    resulting text. When placing limits over a $\sum$, for example,
    \TeX\ automatically centers them over the sum:
    $$\sum_{i=0}^{i=n} i = \frac {n(n+1)} {2}$$

  \item[Size change for big operators.] One can see in the previous
    example that the two $\sum$ signs (one in the text and one in the
    example), are not set in the same size. \TeX\ changes the size of
    some big operators when they are set in a centered environment
    like that example is. The integral also changes size.

  \item[Automatic spacing and math classes.] As one can see in the
    previous example, \TeX\ also spaces various glyphs in a special
    way. For instance the space around the $+$ sign is quite large,
    whereas the space between the $n$ and the open parentheses is
    comparatively reduced. Turning off the automatic mathematical
    spacing for the $+$ sign would produce the following: $(n \mathord
    + 1)$ versus $(n+1)$.

    On a \TeX nical point of view, the math spacing is done by
    dividing all mathematical glyphs into classes. For each class
    \TeX\ has different spacing rules. Thus a class 1 glyph followed
    by a class two glyph would not induce the same spacing as a class
    1 followed by a class 3. There is no point in giving all the spacing
    rules here. The different classes are listed below\footnote{Thanks
      to Victor Eijkout for the comments.}:
    \begin{enumerate}
    \item \textem{Ordinary:} lowercase Greek characters, and those
      symbols that are just `symbols';
    \item \textem{Large operators:} integral and sum signs, and `big'
      objects such as \cn{bigcap}, or \cn{bigotimes}. Large operators
      are centered vertically, and they may behave differently in text
      style, and in display style\footnote{See below for
        explanations.}.
    \item \textem{Binary operators:} plus, minus, and look
      alikes;
    \item \textem{Binary relations:} equal, less than, subset, and
      friends;
    \item \textem{Opening symbol:} opening brace, bracket,
      parentheses, etc$\ldots$
    \item \textem{Closing symbol:} closing brace, etc$\ldots$
    \item \textem{Punctuation:} most punctuation marks, with an
      exception or two;
    \item \textem{Variable family:} described further on in section
      \ref {var-fam}.
    \end{enumerate}

  \item[More symbols/glyphs.] Last but not least, \TeX's math facility
    gives the user easy access to special symbols: Greek letters,
    $\aleph$\footnote{$\aleph$ is a Hebrew letter, not a Greek one.}
    $\cap$, $\subset$, and many others that are often used in
    mathematical formulae.

  \end{description}


\section{Math styles}
  When Knuth wrote ``The \TeX book'', he extended the `display', and
  `text style' terminology. If \TeX\ is typesetting sub- or superscript
  material, one says that it is in \textem{script style.} Furthermore,
  if \TeX\ is typesetting sub- or superscript when it is already in
  script style, one says that it is in \textem {scriptscript style.} The
  style terminology must not be confused with the size terminology that
  is described further on: text size, script size, and scriptscript
  size.


\section{Font families}
  \subsection{What are font families? / a definition}
  In math mode, \TeX\ does not load fonts in the same way as it does
  in text mode. For maths, Knuth thought best to organize the fonts in
  families, and give each family a number. One font family can contain
  three fonts.

  The normal use is to load in a single family the same font in
  three different sizes. One size for the main text, one size for
  superscript and subscript, and one size for the exceptional
  super-superscript, or super-subscript. A good example ought to make
  things clear: $$\int_0^\infty\;e^{\alpha.x^\alpha}\;dx = \;\;?$$ It is
  clear that the $x$ is smaller than the $e$, and that the second
  $\alpha$ is smaller than the first, which is the same size as the $x$.

  \subsection{The organization of mathematical glyphs}
  In the present version of \TeX\ the mathematically used glyphs are
  organized in 4 families:
  \begin{description}
  \item[Family 0: Computer Modern Roman (\texttt {cmr})] This is a
    normal upright roman text font. It is loaded in a math family
    in order to typeset things like $\log$ or $\sin$. The other reason
    for which it is loaded into a math font family is that it
    contains the uppercase Greek alphabet, so that the user can
    typeset $\Psi$ and $\Gamma$, or even $\Upsilon$. A few other
    symbols are also taken from \texttt{cmr}: `;' `=' `( )' `[ ]' `:'
    `+' $\ldots$ See figure in appendix \ref
    {app-fonts}.

  \item[Family 1: Computer Modern Math Italic (\texttt {cmmi})] The
    \texttt {cmmi} font is one of the special math fonts. For a
    non-expert user, its letters look just like normal \textit {italic}
    letters. But in actual fact they are slightly different in their
    shapes, especially the lowercase. The reason for the letters being
    different is so that the variable $a$ can be easily differenciated
    from the article `a' used in ``a horse'' for example.

    Whereas \texttt {cmit}\footnote {The normal italic Computer
      Modern font.} contains ligatures, \texttt{cmmi} does not, and
    includes instead the Greek lowercase and uppercase alphabets in
    italic.

    A strange feature of \fn{cmmi} is that it contains some
    old style digits. Thus one can write $\fam1 1789$ or $\fam1 1942$
    which are quite different from 1789 and 1942. But these digits are
    never used in maths, so they do not belong in a font that is
    designed for use in maths.

    The \fn{cmmi} font also includes some other useful\footnote
    {Only for scientists though.} symbols / glyphs that one can see on
    the corresponding figure in appendix \ref {app-fonts}.

  \item[Family 2: Computer Modern Symbols (\texttt {cmsy})] One can
    find in this font the calligraphic alphabet that some scientists
    use: $\cal{A B C D E F G H}$ $\cal{ I J K L M N O}$ $\cal{ P Q R S
      T U V X Y Z}$; plus lots of other symbols that only
    mathematicians could want to use: $\cap \cup \ominus \otimes
    \bigtriangleup \exists \; \forall \subset \le \succ \leftarrow
    \ldots$ See figure in appendix \ref
    {app-fonts}.

  \item[Family 3: Computer Modern Extensibles (\texttt {cmex})] All
    three sizes in this family are the same.  \texttt {cmex} mainly
    contains symbols that change size, automatically.
    One can produce:
    \[ \left\{
    \begin{array}{ll}
    u(x,y,z,t) & = u_{0}(x,y,t) + U(x,y,z,t) \\
    v(x,y,z,t) & = v_{0}(x,y,t) + V(x,y,z,t) \\
    w(x,y,z,t) & = w_{0}(x,y,t) + W(x,y,z,t) \\
    w'(x,y,z,t) & = w'_{0}(x,y,t) + W'(x,y,z,t)
    \end{array}
    \right. \]\label {extens-example}%
    with four or ten lines, and the `\{' will get bigger and bigger of
    its own accord, without the user specifying anything more. \texttt
    {cmex} also contains wide accents, so one can produce: $\widehat{a}$
    $\widehat{ar}$ $\widehat{arg}$. I have previously spoken about the
    automatic size change of some operators, whether in text, or in
    display, style. These double sized `big operators' are in \texttt
    {cmex}:
    $\bigcap\ and\ \coprod$ in text style, and in display style:
    $$\bigcup \  and\ \coprod\ and \ldots$$
    The total contents of \fn {cmex} is
    shown in a  figure appendix \ref {app-fonts}.

    Most of the glyhs in \texttt{cmex} have a stange metric
    particularity, that makes them \TeX\ specific. Thus no other
    typesetting system can use those glyphs. Vice versa \TeX\ could not
    use those glyphs if they were made for another typesetting system.
    I spent a certain amount of time trying to understand all the
    triks hidden in \texttt {cmex}, and wrote a document on the topic
    (see appendix \ref {app-rep-cmex}). The math font group was then
    able to take decisions concerning the replacement of \fn
    {cmex}.

  \item[The AMS symbol fonts: \texttt {msam} and \texttt {msbm}.] Many
    more mathematical glyphs, and an extra blackboard bold alphabet.
    They are not part of the standard \TeX, and are not loaded
    automatically in a family, but they are used on many sites. They
    were designed for the AMS: American Math Society, for use with
    \TeX, and are now very widely spread. Their contents is shown in
    figures, appendix \ref {app-fonts}.

  \end{description}


  \subsection {How does \TeX\ identify glyphs? }
    \begin{description}

    \item[Glyph names.] In Plain (see section \ref {plain-tex}) many
      glyph names are defined. They refer to some of the numerous
      glyphs \TeX\ can typeset.

      The user can also define his own names for glyphs. To a
      glyph name must be associated a family number, and a position
      in the given family. On top of that \TeX\ likes to know which
      class the glyph belongs to. As well as the classes that have
      already been defined, there is an extra one:

    \item[The \textem {`variable family'} class and the \cn{fam}
      variable.] \label{var-fam} This class has nothing to do with
      spacing, and, to my mind, treating it as a class is one of
      Knuth's mistakes. It is used in particular for letters, but it
      could have other uses. If the calligraphic, upright, and italic
      letters all have the same position in their respective fonts,
      one does not want to define a different name for each letter in
      each shape. Instead, \TeX\ has a \cn{fam} variable, that
      contains the number of the current family where glyphs should be
      taken from. So when a glyph is of class \textem {`variable
        family'}, it is taken from the family number \cn{fam}. But
      that is not enough. Some times the \cn{fam} variable can be
      equal to $-1$, and there is no family number $-1$.  In such a
      case a default family number is used. So together with the class
      and the position, one can assign the default family number for
      each glyph name. When a glyph is not defined as being variable
      family, it always comes from the same family, and its family
      number is linked to its name in the same way as the class
      number.

      Example: when the user enters math mode, \cn{fam} is equal to
      $-1$, the letters come from the default family. By typing:
      \verb|$abda$| which produces `$abda$', one can see that the
      default family for letters is family number 1 (See family
      descriptions). If the user assigns the family variable to $0$
      then the letters will come from family $0$. Thus
      \verb|$\fam0 abda$| produces `$\fam0 abda$'. (See family
      descriptions).
    \end{description}







\section{Font metric files: The ``\texttt{.tfm}'' files}
  \subsection{A theoretical overview}
  When \TeX\ is typesetting a page, and making all the calculations
  that are necessary for this, it does not need the actual picture of
  the glyphs. All \TeX\ needs at this stage is the dimensions of the
  glyphs, and other numerical data. That information is in the
  ``\texttt{.tfm}''\footnote{\texttt{tfm} stands for ``\TeX\ font metric
    file.''} files, and every font has one.  Without it, the font is
  unusable as far as \TeX\footnote{There are some slight exceptions to
    this rule: in some cases a given font can use another font's
    ``\texttt{.tfm}'' file. But the visual results are not very good.}
    is  concerned.

  For mathematical typesetting \TeX\ uses all the information that a
  ``\texttt {.tfm}'' file can give. One of the first things I had to do
  was to study and understand the machinery hidden in the math fonts
  ``\texttt {.tfm}'' files. From a general point of view a font metrics
  file can contain the following data: \footnote {This is not
    restricted to \TeX.  Although the file formats maybe slightly
    different, Postscript type fonts and others use similar
    metric files.  One can find programs to convert the files
    from one format to another.}

  \begin{description}
  \item [Font dimensions.] These are global parameters
    for the whole font. In a normal text font one would find the
    slant (positive on an italic or slanted font), the size of the
    interword space, other interword spacing parameters, more general
    spacing parameters, and the x-height. The latter is the height of
    the `x' glyph, and is used for correct accent positioning.

    The fonts in family 2 and 3 are a little special as far as font
    dimensions are concerned. \TeX\ looks in family 2 and 3 for more
    font dimensions than usual. This extra information is used for
    special math spacing.

  \item [Glyph dimensions.] Each glyph has a height, a width and a
    depth specified in the ``\texttt{.tfm}'' file. The height of the
    box that surrounds a glyph is equal to the height of the glyph
    plus the depth of the glyph, whereas the width is that of the
    glyph. I think it is important to say that \textem{these values
      are theoretical, and can be quite different from the real size
      of the glyph.} Thus some glyphs are bigger than their box. A
    good example of this is the italic `f': \textit{f}. The top right
    end, and the bottom left end stick out of the box. The right hand
    side of a given glyph box is also the left hand side of the
    next\footnote{The box on the right of the first one.} glyph box.

  \item [Kerns.] \label {about-kerning} They are necessary for the
    \textem{automatic adjustment } of the spacing between two glyphs.
    Many non professional electronic typesetting systems have for a
    long time ignored this refinement of traditional typesetting. The
    problem is the following: for visual comfort all the letters of
    the alphabet cannot be spaced in the same manner. For instance
    when an `A' is followed by a `V', the two letters must be brought
    closer together to produce `AV' versus `A{V}'. In other cases
    letters must be separated a little to produce `aj' versus `a{j}',
    or `f!' versus `f{!}'. Otherwise the spacing does not look correct
    compared to the spacing of surrounding letters.  In the
    ``\texttt{.tfm}'' file, for each glyph one can specify kerns with
    every other glyph \textem{of the font}. \textem{When two glyphs
      that are kerned in the ``\texttt{.tfm}'' file are found side by
      side in the right order, \TeX\ automatically brings them closer
      together, or farther away.}

  \item [Ligatures.] \label {about-ligs} Here again, the idea is to
    improve visual comfort, and reading. Some letters when followed by
    other particular letters do not look right. In this case the two
    glyphs side by side must be replaced by another glyph that will
    look much better. This is called a ligature. The best and very
    well known example occurs when an `f' glyph is followed by an `i'
    glyph.  The non-ligatured glyphs look like `f{i}', and
    \textem{\TeX\ automatically replaces} them with the ligature that
    looks like `fi'. In the ``Collection La Pleiade'', one can see many
    other ligatures if one looks hard enough.

    In \TeX\ version 3 the concept of ligatures is more general. It
    can use more than two letters, and has other interesting new
    features.

  \item [Italic corrections.] For this I can only quote Frank
    Mittelbach:
    \begin{quote}
      ``At the points where one switches from slanted or italic to
      upright, the glyphs usually come too close together, especially
      if the last slanted/italic glyph has an ascender\footnote{Here
        is something that has not been defined. The following letters
        have ascenders: l,k,h,f,t,b,d, in lowercase. One can guess
        what descenders are.}. The proper amount of extra
      white space that should be added at this boundary is called the
      `italic correction'. Its value depends on individual glyph
      shape, and is therefore stored in the ``\texttt{.tfm}'' file for
      each glyph. [...]  For an upright font the italic corrections
      are usually null. [...]  In slanted and italic fonts, the italic
      corrections are usually positif...''
    \end{quote}
    Example: in the word {\it dif}ferent, the first f runs into the
    second one. Whereas in the word \textem{dif}ferent, a little space
    is left between the two f's. That space is the f's italic
    correction.

  \item [`Skewchar' kerning.]  The skewchar is a specific character
    that is used for placing mathematical accents. In math mode, when
    an accent is placed on a glyph, the accent is first centered on
    top of the glyph's box, and then shifted rightwards by the amount
    of the kern between the glyph and the skewchar.

    Each font should have its own skewchar. For most characters, the
    ``{\tt .tfm}'' file specifies the kerning of each letter with its
    skewchar. This is true for the computer modern fonts, but other
    font designers may have chosen not to use this feature.

    Why choose one skewchar rather than another? This is because the
    character $\mathchar"017F$ chosen by Knuth does not have any other
    kerning that could have been disturbed by the skewchar kerning.
    This choice may not always be good for all fonts, because it
    depends on what the character in position '127 is.  Thus a font
    designer might choose another skewchar and put the necessary
    kernings in the ``{\tt .tfm}'' file. Accent glyphs can be used as
    skewchars, because they are not usually subject to kerning from
    other glyphs.''


  \begin{sloppypar}
  \item[Charlists.]  \label {about-charlists} Charlists enable several
    characters in a font to be linked together.  The \texttt {cmex}
    font uses charlists a lot: by just typing \texttt{charlist oct
      "000": oct "020": oct "022": oct "040": oct "060"} in the
    metafont source code, one links in order of increasing size all
    the left parenthesis that are in the font. Thus with this
    information contained in the ``\texttt {.tfm}'' file, \TeX\ can
    find the parentheses that has the correct size for what is
    currently being typeset.

    Charlists are used for:
    \begin{itemize}
      \item Linking variable-size delimiters,
      \item Linking variable-width accents,
      \item Pairing the ``big operators'' that are typeset in
        different sizes in display style, and text style.
    \end{itemize}
  \end{sloppypar}

  \item[Extensibles.] \label {about-extens} Extensible glyphs can
    change size vertically (not horizontally), according to the
    context. A good example is given in section \ref {extens-example}
    where the \{ grows automatically.

    \begin{sloppypar}
    An extensible glyph is identified with one of its pieces. One simply
    has to decide which piece is going to be used for this
    identification.  In the following example: \texttt {extensible
      oct"060": oct"060", 0, oct"100", oct"102";} --- which appears in
    the metafont code of \texttt {cmex}, the first oct"060" is the
    identifier of the whole extensible glyph.  The next three
    characters are the top, middle, and bottom pieces of the glyph
    whose identifier is oct"060". The last character code is that
    of the piece to be repeated as many times as necessary between the
    top and middle, and between the bottom and middle pieces. All
    pieces are optional except the repeatable piece.
    \end{sloppypar}

    This mecanism is also used for the construction of the radical
    sign. But it only works for glyphs that grow vertically.
    Therefore the horizontal braces and the horizontal extendable
    arrows cannot use this facility.

\end{description}




  \subsection{Example: analysis of `\texttt{cmmi}' metrics}
  %\footnotetext{\texttt{cmmi} stands for Computer Modern Math italic,
  %  and is the default font for typesetting mathematics.}
  I shall use here the usual \TeX\ notation for writing octal numbers.
  Thus all numbers preceeded by a little quote sign like '77 are in
  octal.
  \begin{itemize}
  \item Most characters in `\texttt{cmmi}' are kerned with the
    skewchar.

  \item Many Greek uppercase and lowercase letters are kerned to: `.'
    `,' and `/' respectivly '72, '73, and '75. This takes us right up
    to position '50.

  \item Characters from '50 to '73 are not kerned at all. This
    includes: funny horizontal half arrows, two hooks for the arrow
    construction set, two triangles, the old style digits, the `.', the
    `,' and the `$<$'.

  \item The `$/$' sign is kerned with 1\footnote {The digit.}, A, M,
    N, Y, Z. Nothing to say about `$<$' and $*$ and $\partial$.

  \item Then come the uppercase Latin letters. They are not kerned
    among each other.  They are not kerned either with the lowercase
    letters. Just like the Greek letters, some of them are kerned with
    `.' `,' `$/$'.

  \item In my \fn{.pl} file, it looks as though N and X have got two
    different kerns with '75.  (Not yet any explanation for this.) The
    3 musical signs are not kerned with anything. The horizontal
    parentheses are not either.

  \item The lowercase Latin letters are not kerned with each other,
    except `$d$' that is kerned with $Y, Z, j, f$. Some of them are
    kerned with `,' `.'  `$/$' in a way similar to that of uppercase
    letters.

  \item The last characters are not kerned at all.
  \end{itemize}

  For compatibility reasons, all these kerns will have to be in the
  new encoding.

  See appendix \ref {app-rep-cmex} for a complete description and
  analyses of \fn{cmex10.tfm}.


%  \subsection{Large operators}
%  \subsection{Delimiters}


















\chapter{Dividing all the glyphs into groups}

\section{More vocabulary}\label{MoreVocab}
  \begin{description}
  \item[An ``encoding table''.] This conveys the traditional meaning
    of an encoding (see section \ref {about-encoding}). That is to say
    a set of 256 glyphs in a given order.  The expression ``encoding
    table'' is usually abbreviated: ``encoding''.

  \item[A ``slot'':] the usual word used for referring to a
    position in an encoding. A slot can contain a glyph, or be empty.
    It is represented by an integer between 0 and 255. A slot is
    \textem{not} a family in spite of the usage some people make of this
    word.

  \item[The ``math kernel''.] This terminology is used to specify the
    minimal group of fonts that is necessary for the math facility to
    work, as described in the \TeX\ documentation\footnote{And
      LA\TeX, AMS\TeX, etc, documentation.}.  In D.~E.~K.'s
    package (Plain) the math kernel consists of the families
    numbered from 0 to 3. Together with the kernel, many other fonts can
    optionally be loaded and used.

  \item[A ``math encoding'':] considered here as a
    whole. Not just one 256-glyph encoding table, but a set of $x$
    encoding tables, where $x$ is greater or equal to the number of
    fonts in the math kernel.  I will sometimes refer to this concept
    with the abbreviation ``M-encoding''.

  \item[The ``default alphabet'':] the alphabet that is used
    when a user types \texttt{\$abc\$}. With Plain \TeX's math
    encoding that produces $abc$.

  \item[``Glyph compatibility'':] two encodings (or M-encodings) are
    glyph compatible, if they contain the same glyphs. The latter do
    not systematically have to be in the same positions. However
    identical glyphs must have the same
    metrics. The kerning and ligaturing information must
    also be identical in both M-encodings.


\end{description}

\section{General approach}
Taking all the glyphs one by one, and putting them in a font encoding
would have been to easy, and above all not satisfactory.  Instead one
must divide all the necessary glyphs into groups and subgroups, and then
tried to match groups in individual encoding tables according to all
the constraints.

For instance, a typical group is the Latin alphabet: it includes the
uppercase letters A-Z, and the lowercase letters a-z.
Mathematicians often use accents on letters. For this they
need a dotless `i' (looks like: `$\i$') and a dotless `j'
(looks like: `$\j$') with every Latin alphabet. Thus the Latin
alphabet group contains the uppercase and lowercase letters, the
dotless `i' and the dotless `j'.

The grouping is based on different types of constraints: some
technical, and some based on glyph usage. These are detailed in the next
section.

\section{Grouping constraints}
Before being able to group the glyphs, grouping rules had to be
established. \footnote {I've put in appendix
\ref {app-accents}, \ref {app-rep-cmex}, and \ref {app-delims}, three
of the documents that I wrote for this purpose.}

At first I did not realise the importance of the design similarity
constraint for the person designing the font. Thus the first two
proposals did not really take it into account at all.

  \begin{description}
    \item [Kerning.] See section \ref {about-kerning} for a definition
    of  kerning. The kerning information for a given font can be found
        in it's \texttt{tfm} file.  Thus two glyphs from the same
        encoding table can be kerned together, but the letter `f'
        belonging in a given encoding table, cannot be kerned with a
        glyph (the open parenthesis for instance) belonging in another
        encoding table. A group resulting from this constraint is: the
        group of glyphs that must be kerned with the default alphabet.
        This group and the alphabet will have to live in the same
        encoding table. This constraint is considered to be one of the
        most important.

        In fact this type of grouping is not so much grouping together
        all the glyphs that must be kerned, but putting together in one
        group the glyphs that need to be kerned with another group. In
        order to facilitate the counting.

  \item[Ligaturing.] See section \ref {about-ligs} for a definition of
        ligaturing.  In a similar manner to kerning, ligatures request
         that various glyphs live in the same encoding. If the letter
        `f' is to be ligatured with the letter `i' and produce the `fi'
        ligature, then those three glyphs `f',`i', and `fi' must live
        in the same encoding. In actual fact ligatures are not really
        used in math fonts. But they may be necessary one day. So
        empty slots should be left for ligatures where possible.

  \item[Design similarity:] another reason for which the letter A must
  live in the same font encoding as the letter B, and all the other
  letters. All the glyphs in a normal text encoding are designed to be
  visually compatible with each other. This should also be the case in a
  math encoding.  But all the compatible glyphs cannot live in the same
  font. There are simply too many of them. So one has to make a
  choice. Which glyphs must be alike? A lot of groups result from this
  constraint, which even comes into play when putting the groups
  together into encodings. A good example is the sim group.  `Sim' is
  the name given to the glyph: $\sim$. Many mathematical symbols contain
  such a sim. $\approx$ cannot be separated from $\sim$ because they
  must look alike, and for that they must be designed by the same
  person.  Even more, the $\sim$ and the $\approx$ should be produced in
  metafont using the same sub-routine, with the same parameters. This
  also explains why it is important that the letters of a given style
  all live together.


  \item[Charlists.] The reader is advised to re-read section \ref
    {about-charlists} if he no longer remembers what charlists are. The
    information that such and such a glyph is part of a charlist is in
    the \texttt{tfm} file.  Therefore charlists are also restricted to
    one font. Because of this all glyphs that are intended to be
    linked in a charlist must be put in the same font. Concerned
    by this restriction are:
    \begin{itemize}
    \item Wide accents, which are linked with a charlist in order of
      increasing size,
    \item Big delimiters: same as accents,
    \item The two sizes of big operators which are linked,
    \item All the different sized radicals.
    \end{itemize}

  \item[Extensibles.] In case of memory deficiency the reader is
    advised to take another look at the relevant passage in section
    \ref {about-extens} again. As for charlists, the extensible
    information is part of the \texttt {tfm} file. The different
    pieces of an extensible glyph must therefore live in the same
    font. Concerned by this restriction are:

    \begin{itemize}
      \item Extensible delimiters (not all delimiters are extensible).
        This constraint is doubled by the fact that an extensible
        delimiter is often the last element of a charlist. Thus many
        glyphs must live together.

      \item Radicals: the last element of the radical charlist is an
                   extensible: it grows as high as necessary. In
                   the same way as delimiters, the glyphs used to
                   build the extensible radical are a subgroup of
                   the radical charlist group, and therefore must
                   live with the other members of the charlist.

      \item Vertical arrows or bars\footnote {But not horizontal
          arrows.}.
    \end{itemize}

  \item[Constructed symbols.] Some glyphs in a font are especially
    designed to be put next to each other. Good examples are the
    horizontal arrows, and the horizontal curly braces. Because of
    their horizontal characteristic, the extensible mechanism cannot
    be used. So the $$\underbrace {\mathrm {horizontal\ curly\ brace}}$$
    is built up with abbuting glyphs. These glyphs must be of the same
    weight, and very well adjusted in order to fit together properly.
    They must therefore live in the same font.

\end{description}



\section {Constraint importance}
The design constraint is less important than the kerning constraint.
Whereas keeping empty slots for ligaturing has very little
importance compared to the two former constraints.

Charlists' and extensible lists' members \textbf {must} stay together,
without exception. One could establish the following order of
importance:

  \begin{enumerate}
  \item Extensibles,
  \item Charlists,
  \item Constructed symbols,
  \item Kerning,
  \item Ligatures,
  \item Design similarities,
  \item Empty slots for ligatures.
  \end{enumerate}



\section{A few groups}

\begin{itemize}
\item The Greek letter sets,
\item The Greek-like glyphs,
\item The Latin letter set,
\item The Latin-like material
\item The digits,
\item The vertical arrows,
\item The horizontal arrows,
\item The accents, wide, double, underaccents,
\item The core symbols: must live with the default alphabet,
\item The subset group,
\item The greater than group,
\item etc ...
\end{itemize}

A lot of the above groups were still divided into smaller groups in
order to make things fit in the encoding tables. Compromises had to be
made, in order to respect the constraints set by compatibility.



\chapter{Making encoding tables}
Similarly to the constraints governing the grouping, the constraints
governing the division into encoding tables listed below were not at
all obvious, and had to be thought of, and fully understood.

The construction of encoding tables largely depends on the main goals of
a new math encoding.

\section{The constraints of group grouping}
\begin{description}

  \item[Glyphs access.] (alphabets, variable family) This is
    another technical constraint due to the way \TeX\ accesses glyphs.
    It is also a user interface constraint, because the idea is to
    make alphabets easily accessible to the user.

  Due to the variable family mechanism (explained in section \ref
  {var-fam}), It is very practical for the user that font
  encodings contain only one alphabet. Thus when the fonts are
  loaded into the families, different letters can be accessed by
  changing the \cn{fam} variable, and typing the usual letters on
  the keyboard. For instance, when \cn{fam} is equal to $-1$, the
  default family is used. When \cn{fam} equals 2 the user can
  get the script alphabet. For this the user need only type
  \verb|$\fam=2 A,B,C$| and the letters $\cal{A,B,C}$ are
  produced.

  The alternative would be to have many alphabets in one encoding.
  In that case, to access script letters A,B,C for example, the user
  would have to type \verb|$\scriptA,\scriptB$|. That would be much more
  difficult to read, and less practical.

  This constraint --~due to glyph access~-- sets the shape of the whole
  M-encoding and has a very high priority.

  \item[Font access.] This only concerns the font that will replace
    \texttt {cmex}. For compatibility reasons, the math font group
    decided that it would be reasonable to try and replace \texttt
    {cmex} by a font that can be loaded in one size, \textem {and} in
    three sizes. Therefore the \texttt{cmex} replacement can only
    take:
    \begin{itemize}
    \item Wide accents,
    \item Big delimiters,
    \item Big Operators,
    \item Radicals (with a small change),
    \item Vertical extensible arrows.
    \end{itemize}
    It would be too long to justify the decision here, but the relevant
    document is in appendix \ref {app-rep-cmex}. One of the consequences
    of this is that one cannot put an alphabet in \texttt {cmex}'s
    replacement encoding. An aphabet must be available in all three
    sizes.
    Other glyphs are also victims of this limitation.

  \item[Kerning.] Obviously, if glyphs in two separate groups must be
                kerned, then those two groups must live together.

  \item[Design similarity.] This is a designor's constraint and
  therefore has low priority. Because of this low priority, it often
  happens that big design similarity groups are subdivided into smaller
  ones. In such case one must try in sofar as is possible to put the
  smaller groups back together.

  \item[Bold face.] Mathematicians and physicists often use boldface
    glyphs. These can either be directly available in some of the
    encodings, whereby the encoding will specify: here should go a
    bold uppercase `A' ---~and that could be next to a non-bold glyph;
    or none of the encoding tables specify whether or not the glyphs
    are bold, and a bold version of the whole M-encoding or of each
    encoding table can be made ---~as with text fonts.

           To reduce the total number of glyphs in the M-encoding, the
           second possibility has been chosen. But this induces another
           constraint on the global M-encoding: the individual encoding
           tables must be designed in such a way that the most commonly
           used bold glyphs are put together.

  \begin{sloppypar}
        \item[Compatibility with other font-using programs.] Since the
           invention of ASCII code, the first 32 slots of fonts were
           often not used for glyphs, but reserved for control
           codes. Today many programs are still not designed to use the
           first 32 slots of a font. Thus fonts should not contain any
           glyphs in those slots.
           But this would be a big waste for \TeX, because it can use
           glyphs in slots below 32.

           However, if the glyphs in the critical slots do not have any
           kerning relation-ship with other glyphs in the font, then the
           former can be put in another font, and be used with little
           difficulty even in problematic software. This seemed a fairly
           good compromise, so it was decided to fill slots below 32
           with glyphs that do not have any kerning with the others, and
           could thus be separated from them.

    On the same lines: some programs are unable to use fonts that do
    not have a space in position 32. To solve this problem, only one
    slot in concerned, so it was decided to include a space in every
    font. This should not be a problem.

  \item[Grouping \TeX\ specific glyphs: another compatibility
    issue.] \hfil
    The present \texttt{cmex} font/encoding contains glyphs
    that cannot be used by other typesetting systems, because they
    are set in a strange way. Similarly \texttt{cmsy} contains one
    glyph that is set in a strange way: the radical sign.  Therefore
    the whole of \texttt{cmsy} is unusable for other programs. Such a
    mistake must not be reproduced.

        It is hoped that the new \TeX\ math encoding will set a
        standard, that will not only be used by \TeX, but by all
        systems that typeset mathematical formulae. If everything goes
        according to plan, in the next few years many math fonts will
        exist, for many different systems, and they will all use the
        same M-encoding. Thus it will be very easy to use the same
        fonts on different systems. One day a \TeX\ user will be able
        to take a mathematical font from Microsoft Word, and convert it
        easily in order to use it with \TeX.

    If \TeX\ specific glyphs are grouped in one font, there will only
    be one problematic font. As it happens, all \TeX\ specific glyphs
    are more or less geometric, so they could be used next to
    different math fonts. On the other hand, if \TeX\ specific glyphs
    are spread around in many fonts, then many ``imported'' fonts will
    not be usable by \TeX\ without major changes.

    From a commercial point of view, if a font designer creates a math
    font for Adobe, the work necessary for adapting it to \TeX\ must
    be reduced to the minimum.  Otherwise nobody will provide any new
    math fonts for \TeX.

    \TeX\ specific glyphs are the following:
    \begin{itemize}
    \item The delimiters,
    \item The large and small `bigops',
    \item The radicals.
    \end{itemize}


  \item[Compatibility with Plain and \LaTeX.] Let us consider a user
    that has typed a document with the present math encoding, and in
    so doing has saturated the available families. If the new math
    encoding does not garantee Plain and \LaTeX\ glyph compatibility
    with a \textem {maximum of 4 fonts,} then the document will
    not be able to run with the new math encoding: not enough
    families. Thus one should make the first four encoding tables of
    the global M-encoding glyphs compatible with the Computer
    Modern cluster: cmr, cmmi, cmex, and cmsy.


  \item[Compatibility with AMS\TeX, AMS\LaTeX, and LAMS\TeX.]  Let us
    consider this time a user that has typed a document with the
    existant AMS\TeX\ or AMS\LaTeX\ package, and in so doing has
    saturated the available families.  If the new math encoding does
    not give AMS\LaTeX\ and AMS\TeX\ glyph compatibility with less
    than 6 encoding tables, then that document will not run with the
    new math encoding, for lack of family reasons.

    The first 6 encoding tables must be one way glyph compatible with
    the fonts provided in the AMS packages.
  \end{sloppypar}

  \item[Trying to give the Plain \TeX\ user a logical cluster of new
    glyphs.] No comment.

\end{description}

\section{The Aston-LC math encoding}
This is one of the proposals first thought of, but it is
not the one finally chosen, because it had many problems.

LC stands for latin core. The main characteristic is the separation of
the Greek letter sets from the Latin ones.  In keeping these two sets
separate, we give the greek letters an identity of their own, thus
making them quite independent of the rest. The idea goes in the
direction of orthogonal grouping. All the encodings that contain
letters would have them in the Cork encoding positions, thus making
access very simple. In fact this positioning concept will be taken
farther: Cork encoded glyphs that are in the new encoding, will keep
their Cork position.

\subsection{The encoding tables}
  \begin{description}
  \item[The text symbols: the TS encoding.] Here would be included the
  old  style numerals, and most of what is to be taken out of the
  present math encoding, because it does not belong with the rest of the
    math glyphs. Other symbols could be added in this encoding.

    This encoding is not part of the M-encoding, but it will contain
    symbols that previously were accessed via the math fonts. In
    normal usage, this font will not be loaded in a family. It will
    simply be loaded as a normal text font.

  \item[The base: a Cork encoded latin text font.] The main use of
    this font would be to typeset function names like \texttt
    {\string\log}. The idea being that the user can actually choose
    this font among the existing Cork encoded fonts. Thus `sin' can
    actually be typeset in the same style as the text, or in another
    special style to match the rest of the math glyphs.

  \item[The core: the MC encoding.] It would not contain any Greek
    glyphs (unlike \texttt {cmmi}). The basic accents (only one size)
    would be here, next to the default numerals. It would also include
    all the upper and lowercase default latin alphabet, all
    of the symbols that are most commonly used, and glyphs that must
    be kerned with the default alphabet.

  \item[The Greek alphabets: the MG encoding.] This encoding table
    would contain all the upper and lowercase Greek letters in
    upright and italic, plus some variable shape Greek letters, also
    in upright and italic, and some numeric Greek letters. Any other
    Greek related glyphs would also live in MG. If place is still
    available, one could include some symbols. An advantage of
    putting the italic Greek and upright Greek together, is that both
    are often requested in medium and in bold weight.

  \item[The extensibles: the MX encoding.] This encoding would look
    very much like the present \texttt {cmex} encoding: the usual
    extensible characters, together with some new ones. It could
    include any characters that have strange \TeX\ features like big
    descenders. Thus glyphs that are not compatible with the
    outer world would be kept together.

  \begin{sloppypar}
  \item[The math symbols: the MS1, MS2, MS3... encodings.] Each of
    these encodings would contain a set of Latin letters, like for
    instance script or blackboard bold, in upper or lowercase, or
    both, together with a set of matching accents if needed. In some
    cases a place should also be reserved for a set of matching
    numbers.  The rest would be filled up with symbols. There could be
    an  MS$_i$ encoding for:
    \begin{itemize}
    \item Calligraphic,
    \item Script,
    \item Open,
    \item Old german, (Fraktur)
    \end{itemize}
  \end{sloppypar}

  \end{description}



\subsection{Other requested typefaces}
  \begin{itemize}
  \item A ``text-like'' italic or slanted font for computer science
    identifier-names and the like.  This would be Cork encoded.
  \item A ``bold upright'' for use as variables -- e.g. vectors in
    physics notation rather than the arrow over an italic letter. This
    would be Cork encoded.
  \item Bold italic for use as variables: an MC or Cork encoding.
  \item Bold Old german (occasional).
  \item Bold script (occasional).
  \item Sans serif lightface (occasional): Cork encoded font.
  \item Sans serif boldface (occasional): Cork encoded font.
  \item Bold symbols: the same encodings loaded in bold.
  \item Ultra bold symbols: the same encodings loaded in bold.
\end{itemize}



\subsection{Summarizing the family occupation}
The following encodings are needed in the kernel:

\begin{enumerate}
\item A Cork encoded upright text font.
\item An MC encoded font containing the default alphabet, digits,
  accents, and symbols.
\item An MS$_1$ encoded symbol font for calligraphic/script.
\item An MX encoded extensible font.
\item An MG encoded font for Greek italic and upright.
\item An MS$_2$ encoded symbol font for Open and symbols.
\item An MS$_3$ encoded symbol font for Old german and symbols.
\end{enumerate}

This occupies 7 families, and leaves 9 free for anything else, (like
bold or sans...) and makes many symbols available.




\subsection{Pros and cons}
This proposal did not respect the limit of 4 and 6 families
(compatibility with Plain\TeX\ and AMS\TeX), nor did it enable the Latin
and Greek to be kerned together, nor could the Greek be kerned with the
same symbols as the Latin alphabet, unless these were
repeated. Generally, to get the equivalent of Plain \TeX, one would have
had to load 5 families, and to get the functionalities of AMS\TeX, one
would have needed to load 7 families.

One of the advantages was the orthogonality of the individual encoding
tables, i.e. there were no strange mixes like Latin and Greek, or
anything of the sort.

The main reasons for rejecting this proposal are:
\begin{itemize}
\item it is a big family consumer. In particular bold Latin and Greek
  would occupy two extra families, and they are frequently requested.
\item it does not enable kerning between the Greek and punctuation
  which is needed for compatibility ---~the punctuation is in
  a separate encoding table from the Greek.
\end{itemize}

The next proposal is more attractive...



\section{The Aston LGC math encoding}

LGC stands for Latin Greek core.  One of the main features of this
proposal is that the Greek and Latin alphabets have been put
together. In one font they could be upright, and in the other they could
be italic. A good reason for doing things this way is that the font
dimension called slant may give a few unexpected problems if italic and
non italic glyphs are mixed in the way that they would have been in the
Aston LC math encoding.


\subsection{The encoding tables}
\begin{description}

\begin{sloppypar}
\item[The text symbols: the TS encoding.] This would be the same as in
  the previous proposal, and would be used in a similar manner.

\item[The base: a Cork encoded latin text font.] This would be the same
  as in the previous proposal, and they would be used in a similar
  manner.
\end{sloppypar}

\item[The core: the LG encoding.] Instead of the MC encoding (in the
  previous proposal), the core could be duplicated. Once in upright,
  and once in uppercase.  The LG encoding would contain one instance
  of both Latin and Greek letter sets. So two LG encoded fonts would
  be used (upright, and italic).

  As far as the other slots are concerned, they could be filled in
  with the most used math symbols (similarly to the MC encoding),
  these would then appear once in upright, and once in bold. An
  alternative to such a duplication would be to make an LG1 encoding
  that would contain different symbols from an LG2 encoding, and these
  would always be in upright, whereas the letters would be specified
  as italic in LG1, and upright in LG2.

  \textbf{Note.} The user could choose whether he wants to load both LG1
  and LG2, or only one of the two.


\item[The extensibles: the MX encoding.] It would be the same as in
  the previous proposal, and it would be used in a similar manner.

\item[The math symbols: the MS1, MS2, MS3... encodings.] {\sloppy These
    would be the same as in the previous proposal, and they would be
    used in a similar manner.}

\end{description}

\subsection{And the rest?}
Similarly to the previous proposal, many other fonts could be loaded
in all the free families.


\subsection{Summarising the Family occupation}
\begin{enumerate}
  \item An LG encoded font containing Latin and Greek italic. (This
    could be LG1 if necessary. See explanations above.)
  \item An MS$_1$ encoded symbol font for calligraphic/script.
  \item An MX encoded extensibles font.
  \item An LG encoded font containing latin and Greek upright. (This
    could be LG2 if necessary. See explanations above.)
  \item An MS$_2$ encoded font for Open and symbols.
  \item An MS$_3$ encoded font for Old German and symbols.
\end{enumerate}

Only 6 families are occupied. This leaves 10 families free for
anything else, (like bold or sans...)  and makes many symbols
available.



\subsection{Pros and cons}

One of the advantages of this proposal is that kerning can be done
between Latin and Greek (as long as they are in the same shape), and
between Greek and other symbols present in the encoding such as
punctuation. Also when bold is requested, one gets the bold Latin and
the bold Greek in the same font table, which again consumes less
families than having the two separate.

This proposal occupies less families than the previous one.

Reasons for abandoning the Aston LGC math encoding:
\begin{itemize}
  \item The user must be able to choose the look of his log, sin, and
    friends.  He may want them to be either text compatible, or
    compatible with the other math alphabets and the rest of the math
    glyphs in general. The choice must be left open, and the math font
    designer must not impose his decision on the user.
  \item A solution to the previous problem is to include another
    font for this purpose, as in the previous proposal. But then the
    family occupation rises up to 7, and three Latin alphabets are
    loaded, of which one (the LG upright) is probably not going to be
    used much. Thus a lot of precious space is wasted.
  \item The `Yaasp' proposal is much more attractive.
\end{itemize}




\chapter{The proposed YAASP encoding}
\label{app-yaasp}
\begin{quote}
   This chapter is the final proposal that was made. It is also the body
   of the official document that was produced. The reader may find some
   similarities with the previous sections, for instance some of the
   definitions can be found in section \ref {MoreVocab}. Also a lot of
   the points discussed in the global policy section have already been
   discussed.

\end{quote}

\input{l3d007a.tex}



\chapter{The glyph groups}

\input{l3d007b.tex}




\chapter{Final conclusions}

The `Yaasp' proposal, which is the final proposal made, is given in
chapter \ref{app-yaasp}.

Working on the \LaTeX3 project in Mainz was very interesting for many
reasons:
\begin{itemize}
\item People next to me were working on net management and system
  maintenance. I used email intensively for communicating
  with other people working on the project. I used tar files and
  other programs to send large amounts of information to other people.

  It was a very good introduction to the network oriented studies I
  will be doing in my last year. I don't think that I really knew before
  what the network was.  Now I have a better idea.

\item I installed a test version of NFSS2. It was a good
  introduction to software installing, and enabled me to discover a
  few more UNIX tools. I hope I will be able to use this knowledge for
  installing various packages for \TeX\ and especially emacs in the
  Ecole des Mines de Saint Etienne.


\item The work I did was to a large extent research work and thus
  involved many topics for which no previous experience was
  available. For this reason the work seemed to go slowly, and we often
  had to go back to the drawing board and re-think points that we
  thought were already finished and done with. All of this, of course,
  was made worse by the fact that a lot of the communication was done
  via email.

   However, the final result was very positive. By the end of the three
   months, a complete proposal for a new math font set-up was produced.
   For Frank Mittelbach technical director of the \LaTeX 3 project,
   this is a good achievement, and a big step forward. The next stage is
   to try and implement the proposal, and start testing it.


\item Last but not least, I greatly improved my English and my German.
  I learnt about another country, about its educational system, and
  about its habits, which one can only grasp by working in the country.
  I learnt how to integrate in a foreign environment, and how to deal
  with a few distressing problems: it was not easy to keep calm when
  my car packed up the week-end before I had planned to go back to
  France.  In one's own country garage mechanics aren't easy people to
  deal with; matters get even worse when it is in a foreign country.

\end{itemize}

The whole experience was very enriching in many domains. The difficulty
I had in speaking German proved to me that the teaching of languages in
the \textem {Ecole des Mines de Saint Etienne} is not good enough, and
must be improved and given more importance.


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% appendix starts here.
\appendix


\chapter{Analysing \TeX's positioning of \cn{mathaccent}s}
\label {app-accents}
\input{l3d007c.tex}


\chapter{A close look at extensible characters}
\label {app-delims}
\input{l3d007d.tex}


\chapter{Replacing \texttt{cmex} ?}
\label{app-rep-cmex}
\input{l3d007e.tex}






\chapter{Fonts and font encodings}
\label{app-fonts}

The first 4 figures given on the next few pages are the standard fonts
used in plain \TeX\ for maths.
\begin{itemize}
\item \textbf{Computer Modern Roman:} loaded in family 0 shown on figure
  \ref {fnt-tbl-cmr}.
\item \textbf{Computer Modern Math Italic:} loaded in family 1 shown on
  figure  \ref {fnt-tbl-cmmi}.
\item \textbf{Computer Modern SYmbols:} loaded in family 2 shown on
  figure \ref {fnt-tbl-cmsy}.
\item \textbf{Computer Modern EXtensibles:} loaded in family 3 shown on
   figure \ref {fnt-tbl-cmex}.
\end{itemize}
%
Figure \ref {fnt-tbl-dcr} shows the DC-encoding with which the new
math encoding is designed to live. Unlike the \fn{cmr} encoding, the
Cork encoding does not include any Greek glyphs, this prevents its use
in family 0 for maths. But an upright text font is needed in family 0,
for mixing sub- and super-script in text. This problem has until now
prevented the wide spreading of the DC-fonts.

The next two fonts shown in figures \ref {fnt-tbl-msam} and \ref
{fnt-tbl-msbm} are the AMS fonts, designed especially for use in
maths.
%
\begin{figure}[b]
  \dofonttable{cmr10}
  \caption{The \fn{cmr} encoding: 128 glyphs.}
  \label{fnt-tbl-cmr}
\end{figure}
%
\begin{figure}[b]
  \dofonttable{cmmi10}
  \caption{The \fn{cmmi} encoding: 128 glyphs.}
  \label{fnt-tbl-cmmi}
\end{figure}
%
\begin{figure}[b]
  \dofonttable{cmsy10}
  \caption{The \fn{cmsy} encoding: 128 glyphs.}
  \label{fnt-tbl-cmsy}
\end{figure}
%
\begin{figure}[b]
  \dofonttable{cmex10}
  \caption{The \fn{cmex} encoding: 128 glyphs.}
  \label{fnt-tbl-cmex}
\end{figure}
%
\begin{figure}[b]
  \dofonttable{msam10}
  \caption{The \fn{msam} encoding: 128 glyphs.}
  \label{fnt-tbl-msam}
\end{figure}
%
\begin{figure}[b]
  \dofonttable{msbm10}
  \caption{The \fn{msbm} encoding: 128 glyphs.}
  \label{fnt-tbl-msbm}
\end{figure}
%
\maxz=256 \maxiz=255
\begin{figure}[b]
  \dofonttable{ecrm1000}
  %\dofonttable{cmssdc10}
  \caption{The \fn{dcr} encoding: 256 glyphs.}
  \label{fnt-tbl-dcr}
\end{figure}
%

%\begin{figure}[b]
%  \dofonttable{cspex10}
%  \caption{The \fn{cspex} and Saint Mary Road encodings.}
%  \label{fnt-tbl-mary}
%\end{figure}



%\chapter{Requirements analysis}
%\label{app-require}











%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% end
\end{document}
\chapter{Daily detail of what I did (for personal record)}

\begin{itemize}

\jzitem{First week}
A lot of reading: Tex for the impatient...
Getting accustamed to the local environment: Emacs Bash Mwm...
Getting Telnet and mail connection with France.
Started my social report.
Looked a little at the code of latex.
Played around with latex. (wrote two letters)
Played around with the Tex boxes.
Read the chapter 7 from Frank's coming book.
Started to read a little more about the math fonts in \TeX.

\jzitem{Sunday 13th June}
Started this report. Got familiar with Frank's integrated documenting
system. Got to letter f in the tex index of commands. (evening)

\jzitem{Monday 14th June}
Trying to install NFSS2 in my environment.
Read a lot of the installation documentation.
Found the punk font, it was here.
Read to letter k of the tex index.(evening)

\jzitem{Tuesday 15th June}
Last details of installing nfss2.
Testing various features of NFSS2.
Playing with nfss2: first page of my report.
Read to letter p of the tex index.

\jzitem{Wednesday 16th June}
Still correcting a few problems with the installation of nfss2.
Installing the punk font in NfSS2:
made file nfpunk.dst based on nfpandor.dst.

\jzitem{Thursday 17th June}
Still installing the punk font in NfSS2.
Read mail about maths.
Went to get my car.
End of afternoon with Frank.
Read to letter t of the tex index.

\jzitem{Friday 18th June : time flies}
Ended the installation of the punk fonts.
Read the article from tugboat about the punk fonts.
Did the documentation for the pk fonts.
Played around with NFSS2 and texlatex: getting my presentation page ok.

\jzitem{Saturday 19th June}
Ended the tex index.
Read more about char in the Tex by topic.

\jzitem{Sunday 20th June}
Read still more about the Tex fonts and math fonts.
Tested char and loops =$>$ loops make problems in latex.
Tried postcript fonts.

\jzitem{Monday 21st June} Tried to install the dunhill font with
nfss2: OK.  Maybe make a ``\texttt{.sty}'' file like punk. I find it
is a nice font.  Used raise and negative kerns for a joke.  Reading
more on math fonts.

\jzitem{Tuesday 22nd June}
Still more reading on mathfonts.-Families-mathcode-mathchar-greek
and testing. Finaly solved all problems due to the loop.

\jzitem{Wednesday 23rd June}
Testing on skewchar, and more reading about maths.

\jzitem{Thursday 24th June}
More testing with skewchar.
Meeting with FMI

\jzitem{Friday 25th June}
All day : writing the paper about accents.

\jzitem{Week end}
nothing about tex or latex
oh yes: reading about fonts. (cahier gutenberg)

\jzitem{Monday 28th June}
Morning more testing with accents.
Afternoon: Meeting with FMI and Jorg K.

\jzitem{Tuesday 29th June}
Writing a summary of yesterday, and trying to figure out how to do the
work.

\jzitem{Wednesday 30th June}
Reinstalling NFSS2. Sending NFSS2 to Estonia. Reading all the mail
printed by Joerg.

\jzitem{Thursday 1st July}
More thinking about maths. Sending mail to people. Reading about the
Tex files.

\jzitem{Friday 2nd July}
More mailing.

\jzitem{monday 5th July}
More mailing, meeting with Frank.

\jzitem{Tuesday 6th July}
More mailing, reading about the extensible chars.

\jzitem{Wednesday 7th July}
More mailing. Sending stuff to joerg. Dealing with Aston.
Continuing The paper on extensible chars.

\end{itemize}
\end{document}


% Local Variables:
% mode: latex
% TeX-master: t
% TeX-command-default: "LaTeX2+"
% End: