summaryrefslogtreecommitdiff
path: root/info/ltx3pub/l3d002e.tex
diff options
context:
space:
mode:
authorNorbert Preining <norbert@preining.info>2019-09-02 13:46:59 +0900
committerNorbert Preining <norbert@preining.info>2019-09-02 13:46:59 +0900
commite0c6872cf40896c7be36b11dcc744620f10adf1d (patch)
tree60335e10d2f4354b0674ec22d7b53f0f8abee672 /info/ltx3pub/l3d002e.tex
Initial commit
Diffstat (limited to 'info/ltx3pub/l3d002e.tex')
-rw-r--r--info/ltx3pub/l3d002e.tex1599
1 files changed, 1599 insertions, 0 deletions
diff --git a/info/ltx3pub/l3d002e.tex b/info/ltx3pub/l3d002e.tex
new file mode 100644
index 0000000000..52e5d5f036
--- /dev/null
+++ b/info/ltx3pub/l3d002e.tex
@@ -0,0 +1,1599 @@
+%%% ====================================================================
+%%% @LaTeX3-article{ LaTeX3-LTX3-002e,
+%%% filename = "l3d002e.tex",
+%%% archived = "ctan:/tex-archive/info/ltx3pub/",
+%%% related-files = "part of l3d002.tex",
+%%% author = "David Rhead",
+%%% doc-group = "Project core team",
+%%% title = "Some ideas for improving {\LaTeX}\\ General",
+%%% version = "1.1",
+%%% date = "18-Mar-1993",
+%%% time = "20:19:36 GMT",
+%%% status = "public, official",
+%%% abstract = "Ideas and suggestions from David Rhead for
+%%% improving various areas in LaTeX",
+%%% note = "prepared for the workshop at Dedham 91",
+%%% keywords = "",
+%%% project-address = "LaTeX3 Project \\
+%%% c/o Dr. Chris Rowley \\
+%%% The Open University \\
+%%% Parsifal College \\
+%%% Finchley Road \\
+%%% London NW3 7BG, England, UK",
+%%% project-tel = "+44 171 794 0575",
+%%% project-FAX = "+44 171 433 6196",
+%%% project-email = "LTX3-Mgr@SHSU.edu",
+%%% copyright = "Copyright (C) 1993 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@vm.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:
+%%% ftp.tex.ac.uk
+%%% ftp.dante.de
+%%% ftp.shsu.edu
+%%% 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@ftp.shsu.edu
+%%% 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 = "28553 1599 11400 79963",
+%%% 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.",
+%%% }
+%%% ====================================================================
+
+\chapter{Citations and reference-lists}
+\label{citing}
+
+\section{Some e-mail comments}
+
+\begin{footnotesize}\begin{verbatim}
+From: Sebastian Rahtz <spqr@UK.AC.SOTON.ECS>
+Date: Tue, 7 Aug 90 15:33:13 bst
+
+... writes:
+ > One of the most common mistakes that I see from LaTeX users is typing
+ > \maketitle before \begin{document}; since this causes horrible things to
+ > happen, perhaps we should make sure that this causes an error.
+and why not put \bibliographystyle before \begin{document}, where it
+belongs..
+\end{verbatim}\end{footnotesize}
+\begin{center} --- \end{center}
+\begin{footnotesize}\begin{verbatim}
+From David Rhead ...
+Date: 13 Aug 90 17:39:39
+
+Sebastian suggested
+ "and why not put \bibliographystyle before \begin{document}, where it
+ belongs.."
+
+This might need thinking through, bearing in mind that certain types of
+document may have more than one list of references. For example:
+Manuals produced by software houses, e.g. the SPSS-X documentation.
+ Such manuals may, in effect, divide citations into 2 categories:
+ 1. references to other manuals produced by the software house,
+ which might be cited using a "short title" scheme
+ 2. references to other literature, which might be cited using
+ an author-date scheme.
+ E.g. the SPSS-X Introductory Statistics Guide generally uses
+ author-date, but gives the full reference to the SPSS-X User Guide
+ in its preface and thereafter refers to it as "SPSS-X Users Guide".
+ The software house may list its own publications in the preface to
+ its manual, putting the list of references to other literature
+ at the back of the manual. To support this sort of thing would
+ require something like
+ \documentstyle{manual}
+ \begin{document}
+ ...
+ % \xxx represents a command that embeds \bibliography-like
+ % information in a preface. "software-house" represents
+ % a style that software houses seem to like where the
+ % reference is embedded in an explanatory paragraph.
+ \xxx[Our other manuals]{software-house}{our-manuals}
+ \chapter{...}
+ ...
+ \bibliography[Other people's stuff]{author-date}{other-literature}
+ where \xxx and \bibliography are assumed to have
+ * an optional argument to specify a title
+ * an argument to specify the scheme (like \bibliographystyle does)
+ * an argument to specify the bib files.
+Books. It might be sensible to divide a book's references into 2 lists, e.g.
+ "References" and "Further reading". E.g. the draft revised British
+ Standard for theses suggests having "Bibliography" as well as "List
+ of references". It might also be sensible to have different styles
+ for the 2 lists: perhaps a concise style for the "References",
+ but a style that prints some extra information about the
+ "Further reading". The author might want to specify something like
+ \documentstyle{...}
+ \begin{document}
+ ...
+ \bibliography[References]{author-date}{specific-books}
+ \bibliography[Further reading]{annotated}{general-books}
+ ...
+ [Such sub-division seems to be countenanced by the gurus, e.g. Chicago
+ Manual of Style (p. 425) and Butcher's Copy-editing (p. 183,192).]
+Unfortunately, such considerations lead to more questions like "How does
+one arrange that \cite gives (...) around author-date citations, but not around
+short-title citations?" and "Can a root file have several bbl files and what
+would they be called?", and I don't know the answers.
+
+At the moment, I just think that it would be better to refrain from putting
+\bibliographystyle before \begin{document} until the implications for
+documents that have more than one list of references have been thought through.
+Otherwise there might be a change to "\bibliographystyle before
+\begin{document}" at LaTeX 3.0 followed by another change (e.g. to
+"bibliography style as argument of \bibliography") in some subsequent LaTeX.
+\end{verbatim}\end{footnotesize}
+\begin{center} --- \end{center}
+\begin{footnotesize}\begin{verbatim}
+From: Sebastian Rahtz <spqr@UK.AC.SOTON.ECS>
+Date: Tue, 14 Aug 90 10:00:33 bst
+
+ > "and why not put \bibliographystyle before \begin{document}, where it
+ > belongs.."
+David Rhead has some sensible cautionary remarks, and I would concur
+with his underlying thesis that the bibliography support in
+LaTeX/BibTeX has definite lacunae. He identifies the problem
+
+ > This might need thinking through, bearing in mind that certain types of
+ > document may have more than one list of references. For example:
+
+and suggests a formulation along the lines of
+ > \bibliography[References]{author-date}{specific-books}
+ > \bibliography[Further reading]{annotated}{general-books}
+
+this is interesting, but nobody will thank anyone for changing the
+basic syntax of a LaTeX command in the near future. I would suggest
+retaining \bibliography as it is documented, and implementing David's
+suggestions with others as new commands. ... such as separate
+bibliographies for chapters - whatever LL or OP say about it being so much
+work to do a multi-chapter book that sets of bibliographies being not
+much more work, I don't see why we shouldn't get it in one day. People
+*do* want it. I am on my fourth conference proceedings in as many
+years, and I dont enjoy building the slightly complicated Makefile to
+get all the references in the right place up to date.
+
+but I continue to say that the user who puts \bibliographystyle before
+\begin{document} is behaving intuitively, and should therefore be
+allowed to do it. lets keep `bibliography' for the relatively simple
+academic type that LL and OP envisaged, and invent a new term
+(\reference?) for what David Rhead is talking about.
+\end{verbatim}\end{footnotesize}
+\begin{center} --- \end{center}
+\begin{footnotesize}\begin{verbatim}
+From: "Nelson H. F. Beebe" <Beebe@EDU.UTAH.SCIENCE>
+Date: Sat, 18 Aug 90 17:15:10 CET
+
+I just hit what I view as a design flaw in thebibliography
+environment. It was precipitated by the following BibTeX
+bibliography entry in the .bbl file:
+
+\bibitem{Steele:floating-point-output}
+Guy~L. {Steele Jr.} and Jon~L. White.
+\newblock How to print floating-point numbers accurately.
+\newblock {\em ACM SIG\-PLAN Notices}, 25(6):112--126, June 1990.
+\newblock In electronic mail dated Wed, 27 Jun 90 11:55:36 EDT, Guy
+Steele reported that an intrepid pre-SIGPLAN 90 conference
+implementation of what is stated in the paper revealed 3 mistakes:
+
+ \begin{enumerate}
+ \item
+ Table~9 (page 125):\par\noindent
+ \begin{tabular}{ll}
+ for & {\tt -1:USER!({"}{"});} \\
+ substitute & {\tt -1:USER!({"}0{"});}
+ \end{tabular} \par\noindent
+ and delete the comment.
+ \item
+ Table~10 (page 125):\par\noindent
+ \begin{tabular}{ll}
+ for & {\tt fill(-k, {"}0{"})}\\
+ substitute & {\tt fill(-k-1, {"}0{"})}
+ \end{tabular}
+ \item
+ Table~5 (page 124):\par\noindent
+ insert {\tt k <-- 0} after assertion, and also delete {\tt k
+ <-- 0} from Table~6.
+ \end{enumerate}
+
+The effect of this is that the next bibliography entry gets
+number 4, one more than the last enumerate counter. The reason
+lies in latex.tex:
+
+% The thebibliography environment is a list environment. To save the
+% use of an extra counter, it should use enumiv as the item counter.
+
+For now, I will switch to an itemize instead of enumerate.
+
+The question for this list is, is the saving of an extra counter
+at what is usually almost the end of a document anyway worth this
+design gotcha? I suggest not.
+\end{verbatim}\end{footnotesize}
+\begin{center} --- \end{center}
+\begin{footnotesize}\begin{verbatim}
+From: MITTELBACH FRANK <PZF5HZ@EARN.DRUEDS2>
+Date: Fri, 5 Oct 90 17:43:46 CET
+
+I would like to foward a mail from Nico which was a reaction
+to some discussions at the Cork meeting.
+
+------------------------ forwarded mail -------------
+>
+> here are some first thoughts on BibTeX and front matter.
+>
+> BibTeX
+> ------
+> BibTeX is unfortunately not a database management system --
+> unfortunate in the sense that users have to think of a way to
+> *really* manage the database, i.e. add, change, remove, sort, select,
+> import and export entries. On my Atari I use a simple database
+> program -- a real GEM program with mouse and buttons -- to maintain a
+> literature database. I've written a conversion program that generates
+> a .bib file from this database. This system works and is also used ny
+> my ex-colleagues at the university.
+>
+> In general, working with BibTeX is cumbersome -- every time you add a
+> reference to your document you have to run LaTeX, BibTeX, and LaTeX
+> twice again -- and a lot of LaTeX users think it only pays off when
+> your document/bibliography is above a certain length.
+>
+> What I would like to see in LaTeX is coding of the logical structure
+> of every \bibitem. In version 2.09, BibTeX takes care of the logical
+> structure and outputs formatted text. In other words: a part of the
+> document, the one included by the \bibliography command, does not
+> contain tags for the logical structure. Furthermore, if a user
+> decides NOT to use BibTeX, he/she has to do the formatting completely
+> by hand.
+>
+> Front matter
+> ------------
+> Currently, an article starts with \title, \author and \date
+> instructions. In the Elsevier styles I have added \address,
+> \received, \revised and \accepted commands, and also a keyword
+> environment, similar to the abstract environment. By doing so, we can
+> automatically convert a LaTeX-coded document to an SGML-coded
+> document.
+> I think it is a bad idea to put the front matter information in a
+> .bib record for the above reasons and also for the following reason:
+> some parts of the front matter do not have to be re-used again in
+> bibliographical entries, and therefore do not belong in the .bib
+> database.
+> Examples: list pf previous books in a series, LCC data, dedication,
+> motto (books), keywords, date of receipt, revision, acceptance
+> (articles).
+>
+...
+
+The last comment about the bad idea probably needs some
+explanation: During the Cork conference we discussed the possibility
+to use an extended version of bibtex, which is able to produce
+several bibliogrphies at once, to format a title page (on request).
+
+The idea was that for larger documents one could set up a bibliography
+entry for the document itself with all relevant information in it.
+In the document itself the title would then be formated by a command,
+say \titlefrombib{<key>} which would make request bibtex to produce
+a file containing all necessary information for producing the title.
+This means that every journal that accepts LaTeX input would supply
+a BibTeX style file which generates from a single bib entry such
+a file. The advantages of this method would be that unused information
+would produce no problem (the bst file would simply ignore them).
+Another advantage would be that authors (for eample for TUGboat) would
+already sending a bib entry along with his document.
+
+Of course, a standard way for simple document that generates titles
+without using BibTeX should be availabe too. And the standard document
+styles should also have their bst files for producing the title.
+
+This scheme if, of course only sensible (if at all) if \BibTeX is
+able to produce several bbl files in one pass.
+
+I still think that such a feature has some promising possibilites
+and would like to hear other opinions about it.
+\end{verbatim}\end{footnotesize}
+\begin{center} --- \end{center}
+\begin{footnotesize}\begin{verbatim}
+From: David Rhead
+Date: 9 Oct 90 10:39:45
+
+Frank forwarded some mail from Nico about BibTeX etc.
+
+
+BIBLIOGRAPHIC DATABASE?
+
+I agree with Nico's comment about BibTeX not being a database management
+system. I have the impression that some research workers really want a
+system that will not only contain author, date, etc. for a paper/book,
+but also a copy of the abstract and perhaps some reminders to themselves.
+Thus, they want something that will both
+(1) help (by supplying bibliographic details) when they are in the process
+ of writing a paper
+and
+(2) help them to search through all the papers they've ever read
+ until they come to something that is about a particular subject.
+ E.g. "give me all the papers about X", "what was that paper about Y?".
+
+It's unfortunate that Lamport/Patashnik used the term "bibliographic
+database" (although I can't think of anything better offhand). It leads
+to people thinking that BibTeX will do the things that they associate
+with "database systems" these days.
+
+However, I think that to turn BibTeX into something that did more
+than (1) would be too ambitious. We're going to have enough problems
+finding someone to make BibTeX do (1) better.
+
+I think it would be better to treat (2) as a separate project, and to ask
+"is (2) best done by adding database management features to BibTeX,
+or would it be better done by adding BibTeX-like features to a
+database management system?" I don't know what the answer is.
+
+For the time being, I think users will have to be left in an unsatisfactory
+situation. If the "raw data" is kept in a "real database", the .bib file
+is just yet another intermediate file. I seem to remember that
+Sebastian Rahtz has set INGRES up so that INGRES can hold bibliographic
+information and write it out in the form of a .bib file. It sounds as
+though Nico is doing something similar with a different database
+system.
+
+It seems sensible to use a "real database system" for what "real database
+systems" are good at. Although it is clearly unsatisfactory to have a .bib
+file that merely "shadows" a "real database" I can't think of anything better
+that could be done quickly. [Perhaps a database expert might have some bright
+ideas. Could such an expert write software that took an .aux file, generated
+instructions in a "query language" to select the \cite-ed references, and
+then produced a .bbl file (or equivalent) without there ever being a .bib file?
+I don't know: I'm not a database expert. Even if they could, it
+would probably still be useful to have a standalone program like BibTeX
+that did task (1) in a database-system-independent way.]
+
+A salesman has sent me a leaflet about a piece of software called
+EndNote which apparently seems to aim to do both (1) and (2) for Mac and PC
+word-processor users. (It's not public domain, but then neither is INGRES.)
+I see that it can export information for troff's "refer". He's coming to
+see me at the beginning of November, so I'll ask if EndNote might be tailored
+to read .aux files and produce .bbl files. I expect the answer will be
+"no" or "only after a lot of work", but it does no harm to ask!
+Perhaps one could have a public-domain BibTeX for task (1) with tailored
+proprietary software for people who want (1)+(2).
+
+Conclusion: I can't think of an easy way of improving the "3 times
+through LaTeX and once through BibTeX" business at the moment. I doubt
+whether its worth making BibTeX into a proper database system, although
+it might be worth making "a proper database system" do what BibTeX does.
+
+
+CODING THE LOGICAL STRUCTURE OF EACH \BIBITEM?
+
+I can see the attraction of this. The list of references could be
+held in the root file or \input or \include-ed from a .tex file
+like everything else. There wouldn't need to be any special treatment
+of .bbl files.
+
+However, in terms of project-management, it seems very convenient to regard
+determination of the logical structure for bibliographic references
+as a separate task which can be delegated to whoever volunteers
+(if we can find them) to do another iteration on BibTeX.
+It would be up to this person to specify the logical elements
+for a reference, e.g. to agonise about whether ADDRESS is
+the fundamental concept of whether it should be PLACEOFPUBLICATION.
+
+All that needs agreeing between the LaTeX 3.0 people and the
+BibTeX worker (if we find one) is the form of interface, i.e.
+what BibTeX passes back to (a) go where the \cite was and (b) go
+in the list of references. There might need to be separate
+interfaces for "reference by number", author-date and "short title"
+[I'll have a go at suggesting what these might be in a future message]
+but beyond that, the LaTeX 3.0 people need not be concerned
+about the distinction between e.g. different author-date styles.
+Thus, the BibTeX worker's considerations of matters such as
+ADDRESS versus PLACEOFPUBLICATION can proceed in parallel with
+the LaTeX 3.0 people's work on other matters. It wouldn't matter
+much whether both projects were complete at the same time.
+
+If the LaTeX 3.0 people attempted to code the logical structure of
+each \bibitem, this would mean that LaTeX 3.0 could not be finished
+until the LaTeX 3.0 people had satisfied themselves about "What are
+the fundamental types of publication (\bibitem{knuth-84}{book} or
+\bibitem{knuth-84}{monograph}?), and what are the fundamental
+items of bibliographic information about them? Did Lamport/Patashnik
+get it right? E.g. ADDRESS versus PLACEOFPUBLICATION." The LaTeX 3.0
+people would presumably have to provide TeX code that sorted bibliographic
+details out into the order required for a particular style e.g. "reference
+by number in ACM style" (and perhaps provide a few style-options to
+show how the details could be changed for a different convention,
+e.g. author-date in APA style). I think that the LaTeX 3.0 work
+is ambitious enough without taking this analysis on too. Attempting
+to get the subdivision of \bibitem right could hold the rest of the
+project up.
+
+Continuing to delegate the work on the logical structure of each \bibitem
+to BibTeX might not be as elegant from the user's point-of-view as
+getting LaTeX to do all the work (using structure information from
+subdivisions of \bibitem and bibliography-style information from
+\documentstyle) but I think the result would be available sooner and that
+the user might prefer to have something better than LaTeX 2.09 soon, rather
+than to have perfection not so soon.
+
+The two approaches (1) "put all the BibTeX work into LaTeX" (to take
+account of the logical structure of each \bibitem), and (2) "make
+BibTeX into a proper database management system" seem to be pulling
+in different directions. I don't think one can do both (otherwise
+you'd end up with LaTeX as a bibliographic database management system),
+although one could do neither.
+
+For LaTeX 3.0, I'd be inclined to leave the contents of each \bibitem
+(or the successor to \bibitem) as a "black box", to be filled in by the
+user or by BibTeX. If someone does the analysis for BibTeX 2.0 (say),
+the question could be considered again if there is ever a LaTeX 4.0 (!)
+
+[This all assumes that the SGML-ers have not analysed the structure
+of a list of references and hence that someone has to do the analysis.
+If the SGML-ers have done the anlysis (for a DTD, perhaps), could they
+publish it?]
+
+Conclusion: I'd specify an interface between LaTeX 3.0 and BibTeX
+that would support the "reference by number", author-date and (if
+possible) "short title" schemes, but delegate the task of supplying
+\bibitems (or whatever) to that specification to whoever updates BibTeX
+and its .bst files.
+
+
+FRONT MATTER INFORMATION IN A .BIB RECORD?
+
+The gurus of "how to do a list of references" seem to agree that
+bibliographic details should be as they appear on the title page of
+the article, book, etc. But there are many caveats:
+* Several books by the same author in one bibliography should
+ follow the same style (Chicago Manual of Style, p. 441).
+* There are potential problems with names like Tchaikovsky,
+ which may appear in different forms (Chaikovsky) on different
+ title pages, even though the works are all by the same person.
+ [British Standard BS 1629, p. 5]
+* The part of the name not on the title page may be enclosed
+ in square brackets (Chicago, p. 441).
+* If the name on the title page is a pseudonym, the author's
+ real name may be given in the bibliography in square brackets.
+ (Chicago, p. 442).
+* Capitalization, punctuation, etc. of a title may be differ in a
+ bibliography from the conventions on the title page (Chicago, p. 447).
+ Similarly, compulsory line-breaks may be wanted on the title-page
+ but not in the list of references (\\, Lamport's book, p. 164).
+ For another example, consider "LaTeX: A Document Preparation System":
+ that's not how it appears on the title page.
+* It may be necessary to use discretion about whether to regard
+ a subtitle as part of a title or to abbreviate a long title
+ (ISO standard 690, p. 5).
+* A bibliography may give "place of publication" in a form that is
+ different to that on the title page, using discretion about:
+ - whether to list all places where the publisher has offices
+ or just one place
+ - whether to give further information (if the place of publication
+ is not widely known or could be ambiguous).
+ (Chicago, p. 456).
+* A bibliography might give a publisher's name in a form that differs
+ slightly from that shown on the title page (Chicago, p. 458).
+
+I think that having a BibTeX that can produce several bibliographies at
+once would be "a good thing". For example:
+* for conference proceedings where each contribution may have its own
+ list of references
+* for books that may have e.g. "References" and "Further reading"
+* things like the SPSS manual (and other things produced by software
+ houses), which seem to give the software-house's related publications
+ in a preface, but put "academic references" at the end.
+
+However, I don't necessarily think that the same mechanism should
+be used to "derive a publication's title-page from its .bib entry".
+Traditionally, bibliography entries have been derived from title-pages
+(with some human discretion), rather than the other way round, so its
+probably safer to have software that imitates the tradition. One might
+think of having things like
+\begin{titlepage}
+ \author[bibliography-version]{titlepage-version}
+ \title[bibliography-version]{titlepage-version}
+ \place[bibliography-version]{titlepage-version}
+ \publisher[bibliography-version]{titlepage-version}
+\end{titlepage}
+\begin{copyrightpage}
+ \copyrightholder{...}
+ \isbn{...}
+\end{copyrightpage}
+in the .tex file (where the optional arguments allow humans to exercise
+the discretion recommended by the gurus) and having LaTeX produce
+perhaps a .bibitem file that the user can append to a suitable .bib file
+(maybe after exercising a bit more discretion). LaTeX could put
+information that might conceivably be used by a bibliographer in the
+.bibitem file, but refrain from putting information that no bibliographer
+would ever want to that file. TUGboat would get a .bib entry
+with each article (but it would be derived automatically from the
+article-heading, rather than used to automatically produce the
+article-heading).
+
+[Conversely, if the Cork suggestion was adopted, and title-page
+information was produced by some future BibTeX from a .bib file, there
+would have to be some mechanism to allow for minor variations, e.g.
+ TITLE = "LaTeX: A Document Preparation System",
+ TITLEPAGETITLE = "LaTeX\\A Document Preparation System"]
+
+The problems that I mentioned in the context of "logical structure
+of \bibitem" arise here too. To write .sty files for (say) book, report,
+conference-proceedings and article, you only need to be clear about those
+categories (as well as being a TeX wizard and having a lot of time,
+perseverance and patience). To write .bst files with entry-types of book,
+report, conference-proceedings, article, you need to be clear whether
+they are distinct entry-types or not. [ISO 690 could be interpreted
+as lumping books, reports and conference-proceedings all together
+as "monographs".] You also need to be clear about the fields:
+e.g. PLACE or ADDRESS. So the LaTeX 3.0 code for title-pages would
+get held up (and hence LaTeX 3.0 as a whole would get held up)
+while someone analysed "the structure of a \bibitem".
+On the other hand, if LaTeX 3.0 wrote out a .bibitem file that wasn't
+quite what some new BibTeX expected, it wouldn't matter very much,
+and could be corrected once it was clear what was required.
+
+Conclusion: I'd like BibTeX to support multiple lists-of-references,
+but think that "LaTeX producing .bib info from titlepage info"
+might be better than "BibTeX producing titlepage info from .bib info".
+\end{verbatim}\end{footnotesize}
+\begin{center} --- \end{center}
+\section{My e-mailed \lq\lq paper''}
+
+\subsection{Introduction}
+
+Since there are now plans for a new version of \LaTeX\
+\cite{lamport-86,m+s-89}, this may be a good time to consider
+how a future version should deal with citations and reference-lists.
+
+My viewpoint is that of an advisor to authors who use \LaTeX\ 2.09.
+As such, I'm often in the position of having to decide whether \LaTeX\
+acts inappropriately or whether an author is asking for something
+inappropriate. Generally, I have the impression that \LaTeX\ 2.09 sometimes
+makes it unnecessarily difficult for people to comply with the conventions that
+are standard in \lq\lq academic publishing''.
+
+In making the suggestions that follow, I do realise that it's easier for me
+to suggest that the volunteers developing \LaTeX\ 3.0 might provide some new
+facilities than it is for a volunteer to find time to do the work!
+
+I assume that the reader is familiar with the relevant sections of
+\cite{lamport-86}.
+
+\subsection{Conventions to be supported}
+\label{conventions}
+
+\subsubsection{Citation schemes}
+\label{basic-schemes}
+
+In mainstream publishing
+\cite{bs-5605,bs-6371,butcher-81,chicago-82,gibaldi,huth,inter,iso-690,%
+oconnor,page},
+there seem to be three basic schemes for citations and the corresponding
+reference-lists:
+\begin{description}
+\item[reference by number] In this scheme, citations are normally numbered in
+ order of first citation. In particular, \lq\lq order of first citation''
+ is used by over 300 biomedical journals \cite{huth,inter}, and is
+ specified in the ISO standard \cite{iso-690}. (Thus, in \BibTeX\ terms,
+ the usual numbering sequence is that given by {\tt unsrt}.) The number
+ is used as a \lq\lq label'' in the reference-list.
+\item[author-date] There are two main forms of citation, depending on whether
+ or not the cited author's name occurs naturally in a sentence. In the
+ first case, the citation is of the form \lq\lq\ \dots\ (1972)
+ \dots\ '', whereas in the second case it is of the form \lq\lq\ \dots\
+ (Crane, 1972) \dots\ ''. There are no \lq\lq labels'' in the
+ reference-list, which is arranged in alphabetical order of authors'
+ surnames (with supplementary rules for \lq\lq tie-breaking'').
+\item[short form] The \lq\lq short form'' scheme is often used when citations
+ occur in footnotes. Typically, the first citation (or the first
+ citation in the current chapter) gives a fairly full reference but
+ subsequent citations use a short form. The \lq\lq short form'' may be
+ introduced within the first citation, or given in a table of
+ abbreviations. There are no \lq\lq labels'' in the reference-list,
+ which may be subdivided by \lq\lq type of cited document''.
+ The scheme is common in the humanities, but also seems to be used by
+ some software-houses when referring to their own publications (see, for
+ example, \cite{norusis}).
+\end{description}
+I think that, in order to make it straightforward to achieve, with \LaTeX,
+the effects that people routinely achieve with traditional publishing
+procedures, it is desirable that \LaTeX\ should provide explicit support for
+all three citation schemes.
+
+Notice that:
+\begin{itemize}
+\item It is not, in general, possible to convert a document from one scheme
+ to another (e.g., from \lq\lq reference by number'' to author-date)
+ automatically. Some re-writing is required.
+\item The number of items of information that need to be available
+ differ between the schemes. For a \lq\lq reference by number''
+ citation, it is only necessary to keep track of one item
+ (the number), whereas for the other schemes it is necessary to keep
+ track of more than one item (i.e., the author and the date, or the
+ \lq\lq fairly full form'' and the \lq\lq short form'')
+ so that they can be used separately.
+\item Occasionally, two different schemes may be used in parallel
+ within the same document. (For example, in \cite{norusis},
+ a software-house seems to use \lq\lq short form'' when citing
+ its own publications, but author-date when citing other publications.)
+\end{itemize}
+Thus, it seems to me that the three schemes are best regarded as logically
+distinct.
+
+However, within a particular scheme, there are variations of punctuation, etc.,
+that can be regarded as matters of \lq\lq house style''.
+For example, some journals that have adopted a \lq\lq reference by number''
+scheme use bracketed numerals for citations, while others use superscripts.
+Such variations can be accomodated by differences between style files.
+
+\subsubsection{Additional references}
+\label{support-additional}
+
+In addition to a list of \lq\lq works cited'', some documents
+have an additional reference-list that specifies \lq\lq further reading''
+or \lq\lq all sources consulted'' (see \cite[pp.\ 182 \& 192]{butcher-81},
+\cite[p.\ 40]{huth} and \cite[p.\ 22]{bs-4821}). There needs to be some
+provision for typesetting such additional lists.
+
+
+\subsection{Deficiencies in \LaTeX\ 2.09}
+\label{2.09-problems}
+
+\subsubsection{Citations}
+\label{2.09-citations}
+
+When used in conjunction with \BibTeX\ and \verb+\bibliographystyle{unsrt}+,
+\LaTeX\ 2.09 makes a good job of organising citations (and sorting the
+corresponding reference-list) according to the \lq\lq reference by number''
+scheme as required by many journals.
+
+Generally, however, \LaTeX\ 2.09 does not provide the breadth or depth
+of facilities needed to support the variety of requirements for
+mainstream publishing:
+\begin{itemize}
+\item It is unfortunate that the \lq\lq reference by number'' sequence obtained
+ most naturally by the do-it-yourself-er \cite[p.\ 73]{lamport-86} from
+ \LaTeX\ 2.09 (\lq\lq order of appearance within {\tt thebibliography}'')
+ is unlikely to be what the do-it-yourself-er's journal-editor requires
+ (which will usually be \lq\lq order of first citation''). The
+ do-it-yourself-er is given no warning (either in \cite{lamport-86} or by
+ the software) that sorting is likely to be needed.
+\item Although one can use style-options such as {\tt apalike} and
+ {\tt aaai} to re-define \verb+\cite+ and {\tt thebibliography}
+ for an author-date scheme, there are many obstacles placed in the
+ user's way:
+ \begin{itemize}
+ \item The existence of the style-options is not documented in
+ \cite{lamport-86}.
+ \item If one finds a style-option in a (software) archive, it
+ may need modification to produce the precise effect required.
+ \item It is not obvious how one should refer separately to two
+ items (author and date) supplied via a \verb+\bibitem+ argument
+ originally designed for one. The do-it-yourself-er might
+ have to study style files such as {\tt aaai.sty} and {\tt bbl}
+ files such as those produced by {\tt aaai-named.bst} to
+ deduce how to do this.
+ \end{itemize}
+\item The \lq\lq short form'' scheme seems unsupported.
+\item At certain points in a document, an author my need to cite
+ several works at once. It may be necessary to specify a
+ page (or section, etc\@.) for each work. For instance,
+ \cite[p.\ 404]{chicago-82} suggests references of the form
+ \lq\lq (Kelley 1896a, 10; Kelley 1896b; Kelley 1907, 3)''.
+ This is not easy in \LaTeX\ 2.09, since \verb+\cite+'s
+ optional argument applies to the citation as a whole. The
+ author cannot supply a separate \lq\lq optional argument''
+ for each work. (By contrast, {\sf EndNote}'s analogous facilities
+ \cite[p.\ 58]{endnote} seem to allow each individual work
+ to be given its own \lq\lq additional text''.)
+\item It does not seem easy to use different schemes in parallel
+ within the same document. If \verb+\cite+ and
+ {\tt thebibliography} are defined as required for one scheme,
+ they will usually be unsuitable for any other.
+\end{itemize}
+
+\subsubsection{Additional references}
+\label{2.09-additional}
+
+As stated in section \ref{support-additional} an author may need to
+typeset a list of \lq\lq further reading'' etc., in addition to
+the usual list of \lq\lq works cited''.
+
+If using the {\tt thebibliography} environment from one of \LaTeX\
+2.09's standard styles for such additional references, an author will be
+faced with the following problems:
+\begin{itemize}
+\item the title will be the same as that for the list of \lq\lq works
+ cited'', namely \lq\lq References'' for {\tt article} and
+ \lq\lq Bibliography'' for {\tt report} and {\tt book}
+\item the \lq\lq labels'' (which may be appropriate in the list
+ of \lq\lq works cited'', particularly for the \lq\lq reference
+ by number'' scheme) will also appear in the
+ additional list (where they are inappropriate), because
+ both lists use the same definition of \verb+\bibitem+
+\item by default, the \lq\lq labels'' will not be unique, since
+ the \lq\lq works cited'' list and the \lq\lq additional references''
+ list will both be numbered from one
+\item it will be necessary to supply dummy {\it cite-key\/}s,
+ purely to satisfy the syntax required for a \verb+\bibitem+.
+\end{itemize}
+
+\subsubsection{Other problems}
+\label{2.09-other}
+
+The definitions of the {\tt thebibliography} environment in \LaTeX\
+2.09's standard styles:
+\begin{itemize}
+\item issue either a \verb+\section*+ or a \verb+\chapter*+ command,
+ using a {\it heading} of \lq\lq References'' for {\tt article}
+ and \lq\lq Bibliography'' for {\tt report} and {\tt book}
+\item set {\it left-head\/} and {\it right-head\/} to either
+ \lq\lq REFERENCES'' or \lq\lq BIBLIOGRAPHY''
+\item do not arrange for a table-of-contents entry.
+\end{itemize}
+These definitions can cause problems when the {\it heading}, etc.\
+supplied by the standard style is inappropriate, or when a table-of-contents
+entry is desired.
+Admittedly, anyone who doesn't like the standard styles is free to
+take copies of the style files and modify them to suit their requirements.
+However, I have the impression that:
+\begin{itemize}
+\item among people who are competent to modify style files,
+ modification of these aspects of the standard styles is \lq\lq the
+ rule'' rather than \lq\lq the exception''
+\item those \LaTeX\ users who aren't particularly computer-literate
+ find the whole business mysterious, and seek out support staff
+ who have to modify these aspects of the style files for them.
+\end{itemize}
+Overall, this seems to be an area where \LaTeX\ 2.09 is failing to
+\lq\lq free people from formatting concerns to allow them to
+concentrate on writing'' \cite[p.\ 8]{lamport-86}.
+
+An associated problem is that modified style files may no longer be
+compatible with standard utilities such as {\tt lablst}.
+
+\subsection{Introduction of new facilities}
+\label{new-facils}
+
+It has been decided \cite{m+s-89} that \LaTeX\ 3.0 will be compatible with
+\LaTeX\ 2.09 input files. Thus, in particular, \LaTeX\ 3.0 must define
+\verb+\cite+ and \verb+thebibliography+ so that they have the same effect on
+\LaTeX\ 2.09 input files as the \LaTeX\ 2.09 definitions do. This implies that
+it would be difficult for \LaTeX\ 3.0 to (for example) define \verb+\cite+
+so that there can be an optional argument for each work in a multiple citation
+and define \verb+\bibitem+ so that it can have an \lq\lq author'' argument and
+a \lq\lq date'' argument.
+
+It therefore seems best to provide duplicates of the \LaTeX\ 2.09 facilities
+in \LaTeX\ 3.0 (for \lq\lq backwards compatibility'') but to attempt to
+provide new commands/environments in parallel so as to provide the required
+functionality. The new facilities would be regarded as the \lq\lq normal''
+facilities, would be described in the body of the successor to \cite{lamport-86}, and
+would be the natural choice for new users. The old facilities would be
+regarded as \lq\lq deprecated'' and relegated to an appendix of the successor
+to \cite{lamport-86}.
+
+Thus we can have both backwards compatibility and improved facilities
+for the future.
+
+\subsection{Division of labour}
+\label{div-of-labour}
+
+\subsubsection{Details needed for document \lq\lq as a whole''}
+
+The three basic citation schemes mentioned in section \ref{basic-schemes}
+determine certain details of a document \lq\lq as a whole''. For each
+citation, there must be an entry in a reference-list. Each entry in
+the relevant reference-list must have associated information that can be
+used in citations.
+
+\subsubsection{Details needed for reference-list, etc.}
+
+There are a lot of other details that need to be resolved.
+
+The information within each reference-list entry will probably need
+formatting according to certain rules of \lq\lq house style''.
+The information given in citations needs organising in a consistent way
+(particularly for the \lq\lq short form'' scheme).
+
+Different people may want to assemble their reference-lists in different ways.
+Some people may wish to \lq\lq do it themselves'' \cite[p.\ 73]{lamport-86} from
+a physical card-index, while some may prefer to use \BibTeX\ to get details
+from a {\tt bib} file. In some disciplines, proprietary systems such as
+{\tt EndNote} \cite{endnote} seem popular (because they help the user to search
+a database for literature to cite, as well as helping the user
+incorporate details of the literature into a document).
+Researchers may also wish to incorporate material obtained by searching
+details held on a {\sc cd-rom}.
+
+A reference-list generally needs sorting into a particular order. Since the
+list may occupy several pages, I assume that any sorting is best done outside
+\LaTeX, either by other software (e.g., \BibTeX) or manually by the author.
+
+\subsubsection{\LaTeX\ and other software}
+
+It seems best to regard the movement of text-strings (e.g., an author's
+surname) within the \lq\lq document as a whole'' as a task that is distinct
+from the arrangement of details within the text-strings, and to assume a
+\lq\lq division of labour'' in which the former task is performed by \LaTeX\
+while the latter is performed by some other software or manually by the author.
+The \lq\lq division of labour'' between \LaTeX\ 2.09 and \BibTeX\ seems to set
+a good precedent.
+
+This division of labour will lead to modular software. Once the
+interface between a reference-list and the rest of the document has
+been defined, people can use \LaTeX\ for the body of their document,
+but can:
+\begin{itemize}
+\item experiment with different software (\BibTeX, {\sf EndNote})
+ for formatting the details of their reference-lists
+\item enhance the other software (e.g., \BibTeX) independently
+ of enhancements to \LaTeX.
+\item lay their reference-lists out manually if they prefer.
+\end{itemize}
+
+\subsection{\protect\LaTeX\ 3.0: A possible user interface?}
+\label{what-to-do}
+
+\subsubsection{Specifications and names}
+
+If the reasoning given in sections \ref{conventions}, \ref{new-facils}
+and \ref{div-of-labour} is accepted, consideration needs to be given to
+the form that new commands/environments should take in order to support the
+three basic citation schemes, and to provide facilities for \lq\lq additional
+references''. In particular, it will be necessary to choose
+names other than \verb+\cite+, {\tt thebibliography} and \verb+\bibitem+ (since
+these names will be kept for the facilities provided for compatibility with
+\LaTeX\ 2.09).
+
+
+\subsubsection{Four sets of commands/environments}
+\label{4-sets}
+
+It seems to me that it would probably be convenient to have three sets of
+commands/environments for dealing with citations and the corresponding
+reference-lists, each set specifically designed to implement a particular
+citation scheme. Having three such sets gives scope for taking proper
+account of the peculiarities of each scheme, without having one scheme
+adversely affected by the peculiarities of another.
+
+To avoid the difficulties mentioned in section \ref{2.09-additional},
+it might also be worth having a specific environment for
+\lq\lq additional references''.
+
+\LaTeX\ 3.0 might, for example, have commands/environments as specified
+in the following table.
+
+\begin{center}
+\begin{footnotesize}
+\renewcommand{\arraystretch}{1.25}
+\begin{tabular}{lccccccc}
+\hline
+ & Citation & Environment for & Entry in \\
+ & & reference-list & reference-list \\
+\hline
+Reference by number & \verb+\numcite+ & {\tt numrefs} & \verb+\numentry+ \\
+\hline
+Author-date & \verb+\dcite+
+ & {\tt adrefs} & \verb+\adentry+ \\
+ & \verb+\adcite+ & & \\
+\hline
+Short form & \verb+\firstcite+ & {\tt sfrefs} & \verb+\sfentry+ \\
+ & \verb+\sfcite+& & \\
+\hline
+Additional references& --- & {\tt morerefs} & \verb+\moreentry+ \\
+\hline
+{\it Analogue at 2.09} & \verb+\cite+ & {\tt thebibliography}
+ & \verb+\bibitem+ \\
+\hline
+\end{tabular}
+\end{footnotesize}
+\end{center}
+
+Here it is assumed that:
+\begin{itemize}
+\item \verb+\numcite+ and \verb+\numentry+ have {\it key-list} and
+ {\it cite-key} (respectively) as their only mandatory arguments.
+\item \verb+\dcite+ and \verb+\adcite+ have {\it key-list} as argument.
+ \verb+\dcite+ gives a citation of the form (1972),
+ while \verb+\adcite+ gives a citation of the form (Crane, 1972).
+ \verb+\adentry+ has three arguments: the {\it cite-key},
+ the author (e.g., Crane), and the date (e.g., 1972).
+\item \verb+\firstcite+ and \verb+\sfcite+ have {\it key-list} as argument.
+ \verb+\firstcite+ gives the form of citation to be used when a work
+ is first mentioned. \verb+\sfcite+ gives the short form
+ to be used in subsequent citations.
+ The arguments of
+ \verb+\sfentry+ might include: the {\it cite-key}, the form
+ of reference to be used at the first citation, and the short form
+ to be used subsequently.
+ Whereas \verb+\numentry+ and \verb+\adentry+ can \lq\lq introduce''
+ the full reference (like \verb+\item+ starts a new item
+ \cite[p.\ 166]{lamport-86}), it may be better for \verb+\sfentry+
+ to have the full reference as an argument, so that it can be used
+ as the default \lq\lq form to be used at first citation''.
+\end{itemize}
+
+Although it would be desirable for the successors to the ``standard styles''
+to define facilities for all three citation schemes, other
+\verb+\documentstyle+s need not define facilities for all three. For example,
+a journal that wants its authors to use the author-date scheme would supply a
+style file that only provides author-date facilities.
+
+\subsubsection{Further details}
+\label{further-details}
+
+\paragraph{Reference by number}
+
+The \verb+\numcite+ and \verb+\numentry+ commands might take the form
+\verb+\numcite{+{\it key-list}\verb+}+ and
+\verb+\numentry{+{\it cite-key}\verb+}+. Notice that, since \verb+\numentry+
+is specifically designed for \lq\lq reference by number'', there is no need to
+allow an optional {\it label} argument like that for \verb+\bibitem+.
+
+To conform to the ISO specification \cite{iso-690}, the successors to the
+``standard styles'' would arrange for \verb+\numcite+ to give a citation of the form (24)
+and for \verb+\numentry+ to give a reference-list entry of the form
+\begin{description}
+\item[{\rm 24.}] {\sc Crane, D.} {\it Invisible colleges.} \dots
+\end{description}
+
+Perhaps \LaTeX\ 3.0 could use the {\tt aux} file to refine an initial estimate
+of the width of the final \verb+\numentry+'s \lq\lq label'', so that the
+do-it-yourself-er wouldn't need to supply a {\it widest-label\/} argument.
+
+\paragraph{Author-date}
+
+The commands \verb+\dcite+, \verb+\adcite+ and \verb+\adentry+ might
+be defined to have the forms \verb+\dcite{+{\it key-list}\verb+}+,
+\verb+\adcite{+{\it key-list}\verb+}+ and
+\verb+\adentry{+{\it cite-key}\verb+}{+{\it author}\verb+}{+{\it date}\verb+}+.
+The {\tt adrefs} environment would {\it not} have a {\it widest-label}
+argument, since in this scheme entries in the reference-list are unlabelled.
+
+If such a definition of \verb+\adentry+ was documented in the successor to
+\cite{lamport-86}, a do-it-yourself-er would be able to use the author-date
+system just as easily as the \lq\lq reference by number'' system.
+
+To conform to the ISO specification \cite{iso-690}, the successors to the
+``standard styles'' would arrange for \verb+\dcite+ to give a citation of
+the form (1972), for \verb+\adcite+ to give a citation of the form
+(Crane, 1972), and for \verb+\adentry+ to give a reference-list entry with no
+label.
+
+\paragraph{Short form}
+
+The \verb+\firstcite+, \verb+\sfcite+ and \verb+\sfentry+ commands might
+be defined as \verb+\firstcite{+{\it key-list}\verb+}+,
+\verb+\sfcite{+{\it key-list}\verb+}+ and
+\verb+\sfentry{+{\it cite-key}\verb+}[+{\it
+ fairly-full-form}\verb+]{+{\it short-form}\verb+}{+{\it
+ full-reference}\verb+}+.
+Such definitions would, in effect, automate Butcher's manual method of
+ensuring consistency \cite[p.\ 178]{butcher-81}.
+Having {\it full-reference} as an argument means that the full reference
+can be used as the default {\it fairly-full-form} (to be used when the work
+is first cited \cite[p.\ 407]{chicago-82}).
+The {\tt sfrefs} environment would {\em not} have a
+{\it widest-label} argument.
+
+The successors to the ``standard styles'' would arrange for \verb+\firstcite+
+to produce the {\it fairly-full-form} and \verb+\sfcite+ to produce the
+{\it short-form}.%
+\footnote{%
+This makes the pessimistic assumption that \LaTeX\ can not itself determine
+whether a citation is the \lq\lq first citation'' of a particular work.
+If someone was ingenious enough to produce code that determined whether a
+citation is a \lq\lq first citation'', {\tt\ttbackslash firstcite} would be
+unnecessary (except, perhaps, for the construction of tables of abbreviations
+\cite[p.\ 414]{chicago-82}.)%
+}
+To conform to the ISO specification \cite{iso-690}, neither
+\verb+\firstcite+ nor \verb+\sfcite+
+would add any punctuation (but, for example, a style file that implemented
+the MLA conventions would have to add brackets \cite[ch.\ 5]{gibaldi}).
+In the successors to the ``standard styles'',
+the \verb+\sfentry+ would produce an entry with no label.
+
+People producing other style files would be free to implement other
+conventions \cite[p.\ 168]{butcher-81}.
+
+Some works may have no bibliography or may have just a \lq\lq select
+bibliography'' \cite[p.\ 168]{butcher-81}. For such works, it will still
+be necessary to supply the details for use by \verb+\firstcite+ and
+\verb+\sfcite+. It may therefore be worth allowing a form of
+{\tt sfrefs} (e.g., \verb+\begin{sfrefs}[null]+)
+that holds details of cited works but does no typesetting.
+
+\paragraph{Additional references}
+\label{3.0-additional}
+
+Since the entries in an \lq\lq additional list'' will not be cited as such
+(although an \lq\lq all sources consulted'' list may contain a duplicate of
+a cited entry in a \lq\lq works cited'' list), the list will be typeset
+without \lq\lq labels''. Even in a document that uses the \lq\lq reference by
+number'' citation scheme (and so needs \lq\lq labels'' in the {\tt numrefs}
+list), there will be no \lq\lq labels'' for the entries in an additional list.
+
+It therefore seems likely that the {\tt morerefs} environment could
+be implemented as a variation of {\tt adrefs} or {\tt sfrefs}, the main changes
+being:
+\begin{itemize}
+\item a change of title (but see section \ref{reflist-scope})
+\item absence of
+ {\it cite-key, author, date, fairly-full-form} and {\it short-form}
+ arguments.
+\end{itemize}
+
+\subsubsection{Order within the reference-list}
+
+As stated in section \ref{div-of-labour}, it is probably best to leave
+any sorting of the reference-list to some other software, or to the author.
+
+However, it might be possible for \LaTeX\ to provide a warning if a
+reference-list is obviously in the wrong order. Perhaps:
+\begin{itemize}
+\item although there may be no easy alternative to numbering
+ \verb+\numcite+s in order of appearance within {\tt numrefs}
+ (even though \lq\lq order of first citation'' is usually what
+ is required), \LaTeX\ could give a warning if a \verb+\numcite+
+ gave a number that exceeded the \lq\lq biggest number produced
+ by \verb+\numcite+ so far'' by more than one.
+\item there could be a warning if an \verb+\adentry+ had an
+ {\it author} whose first letter came before the first letter of the
+ preceding \verb+\adentry+'s {\it author} in the alphabet.
+\end{itemize}
+
+\subsubsection{Citation of a specific division}
+\label{division}
+
+As stated in section \ref{2.09-citations}, provision needs to be made for
+the citation of a particular division (e.g., page, section, chapter,
+equation) of another work. The syntax of citation commands should
+not only allow several works to be cited simultaneously, but should also
+allow the relevant division of each work to be specified.
+
+From the author's point-of-view, there would be a variety of satisfactory
+ways to specify citations that are to appear as
+``[4, p.\ 10; 5; 6, p.\ 3]'', e.g.
+\begin{verbatim}
+ \numcite{smith[p. 10],brown,jones[p. 3]}
+ \numcite{smith, p. 10; brown; jones, p. 3}
+ \numcite{smith & p. 10; brown; jones & p. 3}
+\end{verbatim}
+The precise syntax would have to take
+account of the practicalities of programming a command that has to be able
+to accept pairs of arguments, where the second member of each pair is optional.
+
+Analogous facilities would be needed for author-date and \lq\lq short form''
+citations.
+
+Incidentally, since abbreviations such as p., ch., sec., and fig.\ are
+common when such divisions are specified, I think that citation commands
+should arrange for the optional arguments to be typeset with
+\verb+\frenchspacing+.
+
+\subsection{Details of reference-lists}
+
+\subsubsection{Variations within mainstream publishing practice}
+\label{mainstream-lists}
+
+Although many academic and technical publications involve only a single
+undivided reference-list, some such publications involve:
+\begin{description}
+\item[more than one list]
+ This situation can arise:
+ \begin{itemize}
+ \item when there is a list of \lq\lq further reading'' etc.,
+ as well as the list of \lq\lq works cited''.
+ This case has been covered in sections \ref{support-additional},
+ \ref{2.09-additional}, \ref{4-sets} and \ref{3.0-additional}.
+ \item when conference proceedings are produced, since each
+ contribution may have its own reference-list.
+ \item in manuals for software. For example, in \cite{norusis}, a
+ software house's own publications are introduced in the preface
+ and cited (in effect) using a \lq\lq short form'' scheme, while
+ other people's publications are listed at the end of the manual
+ and are cited using the author-date scheme.
+ \end{itemize}
+\item[subdivisions within a list]
+ Some reference-lists, particularly in the humanities, are subdivided
+ according to the source of the cited documents (see
+ \cite[p.\ 183]{butcher-81}, \cite[p.\ 425]{chicago-82} and
+ \cite[p.\ 88]{gibaldi}).
+\end{description}
+
+In some cases, an author may wish to add explanatory paragraphs describing,
+for example, how material was chosen for a \lq\lq select bibliography''
+\cite[fig.\ 15.11]{chicago-82} or information about access to (document)
+archives \cite[fig.\ 15.16]{chicago-82}.
+
+\subsubsection{\protect\LaTeX\ 2.09}
+
+\LaTeX\ 2.09 can cope with documents that have more than one
+{\tt thebibliography} environment, and seems to deal satisfactorily
+with a situation in which some \verb+\cite+ commands are to
+one {\tt thebibliography} and some are to another (provided that
+the {\it cite-key\/}s are unique). The default effect is
+to give {\it label\/}s that are not unique, which will be acceptable
+when each contribution to a \lq\lq conference proceedings'' has its
+references numbered from one, but not if \lq\lq works cited''
+and \lq\lq additional references'' are both numbered from one
+(see section \ref{2.09-additional}).
+
+The specification of {\tt thebibliography} \cite[p.\ 187]{lamport-86}
+does not allow anything other than \verb+\bibitem+s within a
+{\tt thebibliography} environment. Hence, it is not clear
+how one can introduce subheadings within a reference-list.
+(In practice, a \verb+\section*+ seems to work between
+\verb+\bibitem+s, but I suspect that it puts \LaTeX\ 2.09 into a loop
+if placed before the first \verb+\bibitem+.)
+
+Anyone trying to add explanatory paragraphs (as in \cite[fig.\
+15.11]{chicago-82} and \cite[fig.\ 15.16]{chicago-82}) will probably
+find that \LaTeX\ 2.09 objects that
+\begin{quote}
+\lq\lq{\verb+Something's wrong--perhaps a missing \item+}''.
+\end{quote}
+
+\subsubsection{\protect\LaTeX\ 3.0}
+
+Ideally, in order to provide support for the conventions that are routine
+in mainstream publishing practice, \LaTeX\ 3.0 should be able to cope with
+all the variations outlined in section \ref{mainstream-lists}.
+
+\paragraph{Multiple lists}
+
+The suggestions made in section \ref{what-to-do} would probably cater for most
+situations where a document has more than one reference-list.
+
+The distinction between {\tt morerefs} and the other environments for
+reference-lists would take care of situations where there is a list of
+\lq\lq additional references'' as well as a list of \lq\lq works cited''.
+The distinction between the \lq\lq short form'' commands/environment
+and the other commands/environments would take care of situations where a
+software house uses \lq\lq short form'' for its own publications
+and some other scheme for other publications. Conference proceedings
+will be able to have \lq\lq a reference-list for each contribution''
+if \LaTeX\ 3.0 follows the \LaTeX\ 2.09 precedent that allows
+more than one {\tt thebibliography} in a document.
+
+\paragraph{Subdivisions and explanatory paragraphs}
+
+I have the impression that, because the sciences have different conventions
+from the humanities, people using the \lq\lq reference by number'' citation
+scheme are unlikely to want the options of subdividing their reference-list
+and inserting explanatory paragraphs. Therefore, it would be legitimate to
+say (for example) \lq\lq subdivisions and explanatory paragraphs are supported
+within {\tt adrefs}, {\tt sfrefs} and {\tt morerefs} but not within
+{\tt numrefs}'', if this made the programming task easier.
+
+For example, it might be convenient to implement {\tt numrefs} as
+a \lq\lq list-making environment'' (as in \LaTeX\ 2.09) but to implement
+the other environments in some other way. Lack of support for subdivisions
+and explanatory paragraphs is unlikely to matter in {\tt numrefs}; the
+\lq\lq other way'' (more like \verb+\paragraph+, perhaps?) might make it
+easier to implement support for these facilities in the other environments.
+
+\paragraph{Other problems}
+\label{reflist-scope}
+
+One approach to some of the problems mentioned in section \ref{2.09-other}
+is for the standard styles to define the heading for the reference-list by,
+for example, \verb+\def\numrefsheading{References}+, so that anyone who wants
+to change the heading can do so by issuing a \verb+\renewcommand+ command
+somewhere before the start of their reference-list.
+
+Another approach is to work in terms of the standard publishing industry
+concept of \lq\lq back matter'' \cite[p.\ 4]{chicago-82}. Instead of having to
+understand where the {\it heading}, {\it left-head}, {\it right-head} and
+table-of-contents entry (or lack of it) originate for units such as the
+glossary (if any), the reference-list(s) and the index (if any)
+{\em separately}, an author would only have to understand how these features
+are treated {\em consistently} within \lq\lq back matter''.
+
+Although the \lq\lq back matter'' approach could be used if environments
+such as {\tt numrefs} followed the {\tt thebibliography} precedent and issued
+commands such as \verb+\chapter+ or \verb+\section+ themselves, authors
+might find the way that {\it heading}, {\it left-head}, {\it right-head} and
+table-of-contents entry materialize less mysterious if it was just the same
+for a reference-list as for (say) a glossary. This would imply that
+{\tt numrefs}, {\tt adrefs}, {\tt sfrefs} and {\tt morerefs} should
+not issue commands like \verb+\chapter+ or \verb+\section+ themselves.
+As a bonus, it would then be possible for an author to insert an explanatory
+paragraph before the reference-list, and to arrange for subdivisions.
+
+For example, if there was a {\tt backmatter} environment within which
+\verb+\chapter+ was treated as defining a unit of \lq\lq back matter'',
+an author's file might contain commands such as
+\begin{verbatim}
+\begin{backmatter}
+ \chapter{Glossary}
+ ...
+ \chapter{References}
+ \section{Primary sources}
+ \begin{sfrefs}
+ ...
+ \end{sfrefs}
+ \section{Secondary sources}
+ \begin{sfrefs}
+ ...
+ \end{sfrefs}
+ \chapter{Further reading}
+ \begin{morerefs}
+ ...
+ \end{morerefs}
+\end{backmatter}
+\end{verbatim}
+
+\subsection{Conclusion}
+
+\LaTeX\ has a large number of users, and potential users, who wish to produce
+documents that conform to the conventions that are standard in academic
+publishing. One element of their requirement is the need to conform to the
+conventions for citations and reference-lists that are usual in their
+disciplines.
+
+The choice for \LaTeX\ 3.0 may be between:
+\begin{enumerate}
+\item having more facilities for citations and reference-lists than
+ \LaTeX\ 2.09, perhaps as suggested in section \ref{what-to-do}.
+ This would imply an increase in:
+ \begin{itemize}
+ \item the amount of code needed to implement the facilities, and
+ the guru time needed for writing the code
+ \item the number of pages needed, in the successor to \cite{lamport-86},
+ to describe the facilities --- perhaps 8 pages rather than
+ the 2 pages in \cite{lamport-86}.
+ \end{itemize}
+\item no significant increase in the facilities provided for citations
+ and reference-lists. Contrary to the idea of \lq\lq freeing people
+ from formatting concerns to allow them to concentrate on writing''
+ \cite[p.\ 8]{lamport-86}, many authors (perhaps most authors) would be
+ wasting time:
+ \begin{itemize}
+ \item hacking at style-files
+ \item searching (software) archives for ready made solutions
+ \item taking up support staff's time in the search for advice
+ (with the support staff
+ in turn taking up gurus' time in their search for solutions).
+ \end{itemize}
+ Moreover, most of this time would be wasted by (or on behalf of) authors
+ who don't want anything at all exotic; they just want to conform
+ to the conventions that are standard in traditional academic publishing.
+\end{enumerate}
+I'm inclined to think that the first option would be the lesser of the
+two evils.
+
+\subsection*{Appendix: Interface with \protect\BibTeX}
+\addcontentsline{toc}{subsection}{Appendix: Interface with \protect\BibTeX}
+
+The preceding sections make some suggestions for a \LaTeX\ 3.0 user
+interface that would enable the do-it-yourself-er to conform to
+the conventions that are usual in academic publishing.
+It is also necessary to consider the implications for the interface to
+\BibTeX.
+
+\subsubsection*{Single reference-list}
+
+Inspecting {\tt bst} files gives me the impression that, if it
+was decided to adopt a scheme such as that described in section
+\ref{what-to-do}, it would be fairly easy to produce new {\tt bst} files
+to supersede existing ones. For example, a {\tt bst} file
+that implemented a \lq\lq reference by number'' scheme would write
+\verb+\numentry+ commands rather than \verb+\bibitem+ commands.
+
+If {\tt bst} files were created in this way, they would be able to
+deal with the straightforward situation when there is a single
+reference-list that contains all works cited (plus, possibly,
+any works specified by a command like \verb+\nocite+).
+
+\subsubsection*{Multiple reference-lists, all with the same style}
+
+More complicated situations can arise in which a document involves
+several reference-lists. For example:
+\begin{itemize}
+\item The editor of the proceedings of a conference might want the
+ published proceedings to have a reference-list at the end of each
+ chapter.
+\item If, as suggested in section \ref{reflist-scope}, {\tt sfrefs} was
+ implemented in a way that allowed a sequence of {\tt sfrefs} environments
+ to each be preceded by a \verb+\section+ command, then, as far as
+ \BibTeX\ is concerned, each {\tt sfrefs} environment might be a separate
+ reference-list.
+\end{itemize}
+
+In both these examples, the document would involve several reference-lists,
+but each reference-list would need to be typeset in a common style.
+I assume that the main problems would be in arranging:
+\begin{itemize}
+\item to have multiple {\tt bbl} files, or distinct divisions of a single
+ {\tt bbl} file
+\item that each reference-list takes its entries from the correct {\tt bbl}
+ file, or from the correct division of a single {\tt bbl} file.
+\end{itemize}
+
+\subsubsection*{Two reference-lists, each with a different style}
+
+People producing documents that have a second reference-list (e.g.,
+\lq\lq further reading'') in addition to the list of \lq\lq works cited''
+might want the first list typeset in one style and the second
+list typeset in another. (In particular, if the \lq\lq reference by number''
+scheme is used, the first list will have \lq\lq labels'' but the second
+list will have no \lq\lq labels''.) If the commands/environments suggested
+in section \ref{what-to-do} were implemented, the first list would use
+{\tt numrefs}, {\tt adrefs} or {\tt sfrefs}, while the second list would
+use {\tt morerefs}.
+
+In this situation, it would be necessary to communicate to \BibTeX\
+that two lists are required, but that they are to be typeset in different
+styles. Since the second list is to contain \lq\lq works {\em not\/}
+cited'', it will also be necessary to specify the works to be shown in the
+second list.
+
+If it is decided to extend the interface between \LaTeX\ and \BibTeX\ to
+cater for such situations, it will probably be necessary to consider
+defining alternatives to \LaTeX\ 2.09's \verb+\bibliography+ and
+\verb+\bibliographystyle+ commands, since it seems unlikely that the syntax
+of the \LaTeX\ 2.09 commands could be extended so as to pass the necessary
+information. One might, for example, consider syntax such as
+\verb+\bibtexcites[+{\it cites-style}\verb+]{+{\it bib-files}\verb+}+
+and
+\verb+\bibtexmore[+{\it more-style}\verb+]{+{\it bib-files}\verb+}{+{\it
+ key-list}\verb+}+.
+Here, it is assumed that:
+\begin{itemize}
+\item the \verb+\documentstyle+ would set defaults for the {\it cites-style}
+ and {\it more-style} that are to be passed to \BibTeX, but the user
+ can over-ride the defaults via the optional arguments to
+ \verb+\bibtexcites+ and \verb+\bibtexmore+
+\item \verb+\bibtexmore+'s {\it key-list} argument would be used to
+ specify the works to be included in the list of \lq\lq additional
+ references''.
+\end{itemize}
+
+If it is not practicable to extend the \LaTeX/\BibTeX\ interface to cater for
+these situations automatically, it would presumably be a matter of
+some \lq\lq human intervention'':
+\begin{itemize}
+\item to prepare a {\tt bbl} file for the \lq\lq additional references''
+\item to \verb+\input+ the {\tt bbl} file.
+\end{itemize}
+
+\section{Further e-mail comments}
+
+\begin{footnotesize}\begin{verbatim}
+From: David Rhead ...
+Date: 4 Apr 91 12:16:14
+
+Here are a few comments on Nico's comments (5th March) about my paper on
+citations and reference-lists.
+
+
+>> 2. The paper on reference lists concentrates too much on layout and not
+>> enough on structure. For a, in my humble opinion, much more valuable
+>> discussion of LaTeX 3.0, reference lists and BibTeX I'd like to refer to the
+>> talk Frank Mittelbach gave at the Cork conference last year.
+
+Frank has sent me a copy of "BibTeX reconsidered", by Reinhard Wonneberger
+and himself. I think that this is the written version of the talk that
+he gave at Cork. There seems to be a lot of common ground between my paper
+and "BibTeX reconsidered", although the former is looking at it from the LaTeX
+point-of-view while the latter is looking at it from the BibTeX point-of-view.
+[Both my paper and "BibTeX reconsidered" would like support for multiple
+bibliographies, the short title/form scheme (often in footnotes) and
+commented bibliographies.]
+
+I'd like "document support" (as Reinhard/Frank put it) for reference-by-number,
+author-date and short-form citation schemes. My paper suggested one possible
+user interface through which this support could be provided. Of course,
+other interfaces could be defined that would do the job. I think that the
+main thing is to provide satisfactory support for the 3 schemes.
+
+
+>> I'm totally opposed to the idea of having different coding schemes for
+>> different systems of citation. In my opinion, this goes completely against
+>> the basic idea behind LaTeX and SGML, namely separation of form and contents.
+>> Consider the amount of re-coding when switching from the number system to the
+>> name-year system!
+
+There are differences between the schemes that may be so significant that
+they could/should be regarded as "different in form".
+* For instance, ISO 690 gives the example
+ The notion of an invisible college has been explored in the
+ sciences (24). Its absence among historians is noted by
+ Steig (13, p. 556). It may be as Burchard (8) points out ...
+ which, if converted to author-date, would be
+ The notion of an invisible college has been explored in the
+ sciences (Crane, 1972). Its absence among historians is noted by
+ Steig (1981, p. 556). It may be as Burchard (1965) points out ...
+ where the substitution has to take account of whether the author's name
+ does or does not occur naturally in the sentence. I.e. the form of
+ citation depends on what else is in the sentence, and conversion
+ cannot easily be automated. Similarly, I'd be surprised if one can
+ guarantee to be able to change from other schemes to the
+ short-form scheme without some re-writing.
+* Reference-by-number involves keeping track of one thing (the number),
+ author-date involves keeping track of two (surname and date), short-form
+ involves keeping track of at least two things (form for first citation,
+ form for subsequent citations, perhaps also the full form as the default
+ form for first citation).
+If the schemes are different in form (i.e. involve different logical
+structures), it may be legitimate to consider having different coding
+schemes, while still aiming at separation of form from content (e.g.
+whether reference-by-number uses superscripts or brackets).
+
+However, if someone has sufficient insight to be able to propose a single user
+interface that can cater for all three schemes (plus possible "additional
+references") within one set of commands/environments, I agree that it
+would be very nice. My own attempts to define a single interface
+that would cater for all three schemes have ended up being unsatisfactory.
+Here's how they end up unsatisfactory.
+ ------------
+For the purpose of this account I'll use \refentry to mean the successor
+to \bibitem. Presumably \refentry would have to have 2 or 3 arguments besides
+the cite-key (rather than \bibitem's one extra argument) so that the arguments
+could be used for:
+- nothing, in the reference-by-number scheme (since LaTeX would supply
+ the numbers)
+- author and date, in the author-date scheme
+- first-citation-form and subsequent-citation-form, for the short-form
+ scheme.
+Thus, for reference-by-number, the user (or BibTeX) would supply
+\refentry{cite-key}{}{} ...
+but for author-date they would supply
+\refentry{cite-key}{author}{date} ...
+while for short form they would supply
+\refentry{cite-key}{fairly-full-form}{short-form} ...
+[We're already in trouble. How do we deal with what appears to be the tendency
+in short form for the "full reference" to be the default "fairly full form"?
+Do we go for \refentry{cite-key}[fairly-full-form]{short-form}{full reference},
+and if so what are the implications for the other 2 schemes?]
+Or should one go for
+\refentry{cite-key}{author}{date}{fairly-full-form}{short-form} ...
+so as to give LaTeX all the information it needs to allow completely
+automatic switching between citation schemes, separating form from content
+but placing a heavy burden on the user (who would have to provide
+at least dummy information for both author-date and short-form
+schemes, even if they are only going to use reference-by-number)?
+
+For citation commands one might have \cite and \shortcite (following
+precedents in the archives) [but see section 6.5 of my paper, about citation
+of a specific devision]. By aiming at author-date, one might be able to
+get a .tex file that also worked for reference-by-number. For example,
+if \cite and \shortcite both gave a number when a .sty file implemented
+reference-by-number, while \cite gave (author, date) and \shortcite
+gave (date) when a .sty file implemented author-date, the following
+input might work for both schemes:
+ The notion of an invisible college has been explored in the
+ sciences \cite{crane-72}. Its absence among historians is noted by
+ Steig \shortcite{steig-81[p. 556]}. It may be as Burchard
+ \shortcite{burchard-65} points out ...
+I don't think that the reverse would work, i.e. if you think in terms
+of reference-by-number while writing, you'll do things like
+ Steig \cite{steig-81[p. 556]}. It may be as Burchard
+which, if converted automatically to author-date, would give
+ Steig (Steig, 1981; p. 556). It may be as Burchard
+which gives two occurences of Steig, which isn't right.
+For short-form, you might interpret \cite as meaning "the form of citation
+used at first citation" and \shortcite as meaning "the form to be used
+subsequently". But if you do this, you'll put your \shortcites in
+places that are different from the places you'd put them for author-date
+For example, you might have (using an ISO 690 example again)
+ ... Steig \cite{steig-81}. It may be as Burchard
+ \cite{burchard-65} points out ... Steig \shortcite{steig-81}
+ has further noted ...
+for the MLA or
+ ... Steig\footnote{\cite{steig-81}} ... Steig\footnote{\shortcite{steig-81}}
+for most other publications that use short-form. [This assumes that the first
+of these citations of Steig above is actually the first citation of Steig
+in the whole document (or chapter).] But this would not give you the right
+input file for author-date, for which you would want
+ ... Steig \shortcite{steig-81}. It may be as Burchard
+Again, you might be able to convert automatically TO reference-by-number,
+but it seems unlikely that you could convert automatically FROM
+reference-by-number. The problem might disappear if LaTeX itself could work
+out which citations are "first citations". Then someone can put \cites and
+\shortcites in the places required for author-date but choose a .sty file
+that gives short-form; the .sty file would ignore the distinction between
+\cite and \shortcite and would use the fairly-full-form for "first citation"
+and the short-form for subsequent citations. But can LaTeX work out which
+citations are "first citations"?
+ ------------
+Although one can try (as above) to imagine a single set of
+commands/environments that would support all three schemes, it seems to
+lead to problems, and I worry that, in making things right for one scheme,
+they'll be made wrong for another. So I gave up on the idea of a single
+set of commands, which was why my paper went for the idea of separate
+commands/environments tailored to the 3 specific citation schemes.
+But, if someone has more success than I did in imagining how a single set of
+commands/environments could be simultaneously compatible with all three schemes,
+I'd be interested to see their proposals.
+
+Or should one be less ambitions, perhaps aiming to satisfy people who
+want reference-by-number and author-date, and not bothering about people
+who want short-form? Things like \refentry{cite-key}{author}{date},
+\cite and \shortcite might work for both reference-by-number and author-date.
+(People who want reference-by-number would have to supply author and date
+information that they might regard as redundant, unless they go
+\refentry{cite-key}{}{} or the syntax is something like
+\refentry{cite-key}[author][date].)
+
+
+>> To David's review I'd like to add that \bibitem's have no sub-division, at
+>> least not one that is indicated by explicit control sequences (`tags').
+>> Instead, the tagging of \bibitem's is done _outside_ LaTeX, which has always
+>> struck me as odd.
+
+In theory, it would be nice if the \bibitem's did have subdivisions. There
+might be subdivisions for the do-it-yourself-er that were analogous to
+the fields used by the BibTeX-er.
+
+In practice, to sub-divide the \bibitems would involve deciding what the
+subdivisions should be, which leads one into questions that cause difficulty
+in BibTeX (e.g., "Should it be address or place-of-publication?",
+"Is it really worth having booklet separate?", "Is it really worth
+having phdthesis separate from mastersthesis?", "Does volume mean
+number-in-series or subdivision-of-book?"). It may not be easy to
+answer these questions. I don't think that it is worth delaying
+LaTeX 3.0 while answers are sought. [One could always return to the
+question for LaTeX 4.0, if there was one!]
+\end{verbatim}\end{footnotesize}
+
+