Weighing in on XSL / Standards

[Rick Jelliffe]

 From: David Megginson <david at megginson.com>

>XSL inherits from at least two major sources -- FOSIs, which are/were
>SGML instance-based stylesheets for typesetting SGML documents,
>specified by the USDoD; and DSSSL, which is/was a declarative
>Scheme-based language for transforming and formatting SGML documents,
>standardized by ISO.

I would say that XSL inherits from three major sources: surely calls for
schemas
to use element syntax have had some influence. The arguments that
schemas
will be easier to use and manipulate and read and treat as hypertext
than
some non-XML syntax can just as well be applied to transformation
languages.

(By the way, why isnt Michael Leventhal also saying that XSchema is a
waste of time too: one can write DOM programs to do all that validation
too!)

> Both were implemented and used in
>production-grade systems (in the case of DSSSL, only *partly*
>implemented), and FOSIs were very heavily used through much of the
>1990's.

Dave has mentioned that DSSSL was only partly implemented, but it should
be emphasized that this does not mean a failure in the DSSSL
specification:
for example, why on earth would James Clark's Jade implement Chinese
vertical writing if RTF and MIF (i.e., his target outputs) do not
support it.

It is an ISO requirement to pay attention to internationalization;  I
know that
some of the ISO WG8 saw DSSSL as an effort to gather and formalize
knowledge
about documents and document processing: ISO standards such as SGML,
HyTime
and DSSSL are not intended to be technologies (like Perl, perhaps) which
are handy
but which do not represent any theoretical or methodological gift to the
world. They
also function to raise the base-level of discussion by program designers
about what
is possible, what is needed, what is good, etc.

It is quite like the CALS DTDs, where no-one implemented all the CALS
tables model;
however, vendors got together and made a common base-list of features
they all would
support: it raised the general capabilities of all.

DSSSL came out of long discussions; the first DSSSL drafts used a
non-LISP
syntax, and (my understanding is) that adopting Scheme (LISP) gave the
list-processing
and functional power  needed to get DSSSL going. There is a strong
tradition
in the text-processing world of using LISP: Interleaf, Cost. (11 years
ago I co-wrote
a C text processing tool for XML-ish SGML that used a little LISP for
its scripting
syntax; I found it worked fine.)

>The real question is what kind of market penetration XSL will have.

Then that is largely on of how well it is marketed, not the core
technology!

Rick Jelliffe


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)