Eugene Eric Kim wrote:
> On Tue, 10 Oct 2000, Eric Armstrong wrote:
> > Do we want to start thinking about extending the
> > email.dtd to cover the kinds of structures needed
> > in the OHS, do we want a separate dtd for that, that
> > somehow gets embedded into an email (namespaces?).
> Two comments:
> 1. There are two areas that really need to be addressed in the preliminary
> DTD I put together for e-mail. First, we need to figure out how to set up
> the linking for responses. Second, we need to figure out how to handle
> categorization. The link that Jack sent out a while back regarding RDF
> for e-mail might serve as a good basis for this.
> 2. I don't foresee there being only one OHS DTD. Based on the
> requirements, the only thing that XML documents in general do not address
> are categorization and OHS-style addressability (i.e. SIDs). In other
> words, the OHS should be able to handle any XML DTD with a few tweaks to
> support OHS-style categories and SIDs.
> As a result, I don't see the benefit of trying to come up with a
> completely general intermediate XML DTD for the OHS. In fact, I think
> this would be counter-productive. My thought is that we should extend
> current DTDs for standard document formats (i.e. DocBook, XHTML,
> SVG, etc.), and create and promote custom, but standard DTDs for document
> formats that do not currently have DTDs (i.e. e-mail, Augment, source
> code, etc.).
Are you aware of Jon Bosak's notes on the following email examples?
DSSSL (ISO/IEC 11079), a precursor of XSL
The file mail.sgm contains an SGML document instance marked up according to the
grammatical rules specified in mail.dtd. Each of the stylesheets mail-<n>.dsl
is applied to the document in turn to produce the Jade output files
mail-<n>.rtf and then (using a word processing program) into the PostScript
printer files mail-<n>.ps.
The English/Japanese Jade processing example
This document type definition (DTD) specifies a simple markup language for
electronic mail messages.
<!element mail - - (to, from, date, subj, body) >
<!element to - - (#PCDATA) >
<!element from - - (#PCDATA) >
<!element date - - (#PCDATA) >
<!element subj - - (#PCDATA) >
<!element body - - (para|code)+ >
<!element para - - (#PCDATA) >
<!element code - - (#PCDATA) >
In English, this says: A mail message consists of a single MAIL element. A
MAIL element consists of one TO element, one FROM element, one DATE element,
one SUBJ element, and one BODY element, in that order. TO, FROM, DATE, and
SUBJ elements consist merely of characters; they contain no further
subelements. The BODY element, on the other hand, consists of any number of
PARA and/or CODE elements in any order; there must be at least one PARA or CODE
element. PARA and CODE elements consist only of characters and contain no
This file contains a mail message marked up in the language defined by
<!doctype mail system "mail.dtd">
<date>Fri, 6 Dec 1996 21:27:55 -0800</date>
<para>I killed your process running on jurassic:</para>
USER PID %CPU %MEM SZ RSS TT S START TIME COMMAND
bosak 20109 21.6 0.2 952 656 ? O 15:11:39 246:26 perfset1-14400
<para>It had been running for 6 hours and was consuming almost 100% of
<para>You really should run this sort of program on a machine
other than the mpk17 building server.</para>
A basic stylesheet for hard copy.
The construction rule for the root element MAIL produces a simple-page-sequence
and establishes the basic look of the document.
Most of the characteristics specified for simple-page-sequence are inherited by
its flow children.
This archive was generated by hypermail 2.0.0 : Tue Aug 21 2001 - 17:57:56 PDT