summaryrefslogtreecommitdiff
path: root/usergrps/uktug/baskervi/5_3/barron.tex
blob: 224f7d23202ee4b7fd178d1df59a2bdfc56886ab (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
\title{Portable Documents: Why Use SGML?}
\author[David Barron]{David Barron\\
Department of Electronics and Computer Science\\
University of Southampton}
\begin{Article}
\section{Introduction}
In this article we present a few ideas as a framework for the
discussion of portable documents. We address a number of questions:
\begin{itemize}
\item What are portable documents?
\item Who needs them, and why?
\item How to produce them, now and in the future
\end{itemize}

\section{Documents}
Traditionally, a document was a file (or a deck of cards),
and consisted solely of text.
Today, documents are typically {\em compound}, a
mixture of text and graphics (bit-map or line art) that can be 
rendered on paper or screen. Additionally, they may include
hypertext links (in which case they can only be viewed on screen).
A recent development is the ability to incorporate video 
and sound in a compound document, either embedded 
within the document or linked by a pointer:
such a document is a {\em multimedia} document. Hypertext-style links 
may also be included to form a {\em hypermedia} document: evidently,
multimedia and hypermedia documents can only be `read' on a suitably
equipped computer system.

World Wide Web (WWW) documents are a special case of
compound hypermedia documents 
where the links are to other documents elsewhere on the Internet
They can be regarded as virtual documents, in the sense that the
whole document never exists as a single identifiable object.
More generally, we can define a {\em virtual document} as
a structured collection of information from which instances of 
documents and other resources can be derived. Examples include:
\begin{itemize}
\item The Oxford English Dictionary which exists as a database
from which are derived various printed editions (Shorter, Concise, 
Pocket etc.), as well as the CD-ROM version 
\item Critical editions of a literary text, where a single source
`document' contains all the variations, and can be printed out using
different variants as the base text
\end{itemize}

\section{Portability}
The definition of portability that we shall use in this discussion is
the ability to transmit the document digitally (over a network,
or on a disk or CD-ROM) and re-create a faithful rendering of the document
after transmission, if need be
on a different hardware and/or software platform
from that on which the document was originally created.
It is important to observe that there are three different forms in
which the text and graphics in a document might be re-created:
\begin{itemize}
\item with absolute visual  fidelity
\item   with approximate visual  fidelity
\item   retaining content only
\end{itemize}

\section{Who needs portable documents, and why?}

Three different needs for portable documents can be adduced
\begin{enumerate}
\item Publishers need them in order to distribute electronic 
books and journals
\item Communities with common interests
who need to  share information need them. An example is a
scientific research community whose members use diverse hardware and
software
\item Librarians responsible for digital archives need 
portable documents, since they cannot assume that a particular
hardware/software platform will exist in perpetuity
\end{enumerate}

\section{Examples of successful portability}
\begin{itemize}
\item Computer science researchers and software manufacturers
distribute documents as PostScript files. This works well
if the fonts employed are restricted to the basic 35, and
the use of Adobe Acrobat (pdf files) increases portability when other
fonts are used.

\item The Physics pre-print library at Los Alamos National Laboratory
is used by many physicists world-wide: over 10,000 retrievals per day
are reported. The archive holds pre-prints in \LaTeX\ and PostScript 
formats (figures in PostScript only). This is successful because the
Physics community has for some years used \TeX\ as its preferred means
of exchanging information.

\item WWW documents are highly portable, since
their rendering is (almost entirely) determined by the 
browser software, and the use of a common mark-up language (HTML)
ensures portability
\end{itemize}

\section{Achieving portability}
At first sight it appears that portability might be achieved
by agreeing standards (e.g. \LaTeX, PostScript, ODA, HTML).
At present there is too much choice, and no obvious winner,
especially in hypermedia documents. This is a sign of an
immature technology. Another important fact to take into account
is that it is difficult to impose standards in some environments
e.g. acadaemia, where personal preferences lead to the equivalent of
religious wars.

Particular problems in achieving portability arise from varying
fonts and character codes e.g. in handling European languages.
Unicode will go a long way towards solving the character 
codes problem.

\section{Why use SGML?}
SGML provides a formal and portable definition of document structure.
SGML syntax can define a hierarchical structure of embedded document parts, 
and can associate a type with each component in the hierarchy.
By associating a rendering definition with each type of component,
it is possible to achieve a portable document. In particular,
SGML provides a uniform archive format for a library of portable 
documents.

\subsection{An example}
Suppose it is required to maintain a library of technical 
documents in an environment where some authors use \LaTeX,
whilst others use Microsoft Word. We can define an SGML 
DTD for the document structure, together with \LaTeX and Word 
styles to define the rendering.
This opens up three possibilities:
\begin{enumerate}
\item Author in SGML and use a tool to produce a \LaTeX\  or
Word version from which the printed version can be produced.
\item Author in \LaTeX\ and use a tool to translate to SGML to
produce the archive copy
\item Author in Word and use a tool to translate the RTF form to 
SGML to produce the archive copy
\end{enumerate}

In addition to the SGML version of the documents, the 
archive must contain the Word and \LaTeX\ style files and the 
translation tools. Once this is done, anyone can collect a 
document, the required style files and tools and produce a copy of 
the document.
This will of course only work for text documents. For any 
document with graphics content, and for hypermedia 
documents, more is required. This is possible in 
principle, but much remains to be done

\section{The future}
A combination of SGML and OpenDoc is probably the best way forward. 
OpenDoc provides an architecture for portable documents:
it treats a
document as a container for a collection of `parts', each of which can
have other parts embedded within it. Each type of part has 
associated programs to edit and render it, so that
documents can be re-created with varying degrees of fidelity
depending on the availability of rendering software
for the particular varieties of parts that it includes.

OpenDoc is a dynamic architecture, and assumes that a new type 
of part may occur at any time.
In principle SGML can be used to describe the static 
structure of an OpenDoc document, providing the final link in the
portability chain. 
\end{Article}
Sir -- Philip Taylor is to be complimented on a fine display of pedantry 
in the best academic tradition, the kind of tradition that gives academics
a bad name amongst normal folk. In computing we use lots of everyday 
words with specialised meanings, and most of us find no difficulty
in using the context of an utterance to achieve any necessary disambiguation.
With regard to his criticism of my use of the term "multimedia document",
I agree that I don't plug my computer into a multiways socket. But then,
I don't attend a performance of an operum at Covent Garden, either.

Yours sincerely