Print

Print


Sebastian:

: > One can speculate about why SGML has been such a well kept secret.  It
: > takes some work to appreciate it.  That does not means that it does
: > not work nor that there are not reliable free tools nor that it would
: > not be fairly easy once we come to understand what it can do for us to
: > generate tools that are optimized for our purposes.
: you would have enjoyed a keynote at MT 98 by Brian "Scribe" Reid, where
: he basically repeated a talk he gave in 1981. it was an effective
: demonstration that being "right" and "working" markup stuff
: has zero impact on most people, who simply dont *want* generic markup...

So may I conclude that Sebastian now realizes that my comment this past
Friday was "right"?  :-)

I'm not sure what "generic markup" means, but I do want personally
convenient markup, which I would take *not* to be "generic".

It is true that there is a vague "political" factor, apart from
technical factors, that is involved in the determination of whether an
innovation gains acceptance.

I think all of us here understand that the markup under mathml for
solving quadratic equations has not "played well" in the community.

Some will want wysiwyg authoring interfaces for writing.  Fine.
But I do not, and most of my mathematical friends do not.

There are free reliable tools for sgml (including the xml subset).
The tools that I am now using have been available since December 1995.

Most of my friends are not using them because they do not understand
them.  Sometimes they think they understand them and reject them as
"inadequate".

I thought that it might make it easier for some if there were a
LaTeX-like interface.

I have been trying to persuade my friends, mostly academics, that they
need sgml.  I think that Sebastian agrees except that he would say xml
only.  I agree with xml for browsing except I see a gain for us in
using sgml inside our shops.

In-shop stuff is private.  My gellmu dtd is didactic, not prescriptive.
In fact, my docs have always said that I expect each shop to want
its own dtd just as each LaTeX shop now wants its own favorite list
of packages and styles outside the core.

: > Note: HTML 2.0, HTML 3.2, and HTML 4 all use different SGML declarations,
: > none the default.
: >
: and no HTML browser enforces validation, does it?

That's also a web rule.  Authors and servers should behave according
to strict rules.  Clients should be very tolerant.  (And indeed clients
need to be very tolerant.)

In fact, the prevalence of invalid html gives one pause in
contemplating the future of roll-your-own xml, where documents that
are not valid will splatter across one's screen.

(We do agree that html is sgml and not xml.)

: > For this purpose one should perhaps view an SGML document type as a
: > decl/dtd pair.  Of course, there is no decl for an XML.
: oh come. there is very much a decl for XML!!! its vital for parsing
: XML with SGML tools!

Indeed.  Yes, there is *one* decl for XML.  Hence it is silly in the
case of an XML to conceptualize a document type as a dtd/decl pair.

: > Obviously, to the extent that it is sensible to adopt different formats
: > for different purposes, it is desirable to have automatic processing to
: > faciliate conversions.  Many such conversions should be fairly easy.
:                            ^^^^^^^^^                       ^^^^^^

Disagreement?  Going (1) from exchange to authoring formats or (2)
from shared-authoring to personal authoring formats under sgml does
require some thought.  Providing the thought at early dtd design
stages is a big help.  That's going "uphill".  Going downhill really
is easy.

: by which we see why LaTeX is unpopular in production workflows. that
: translates to "10% failure"

Hmmm...  I believe that the head editor of a math journal that I know
would disagree with this last statement about LaTeX.  I think that he
would say that no mathematician uses sgml or xml and that the LaTeX he
receives is almost always better than the plain TeX that he receives.
(I myself have no formal connection with that journal.)

                                   -- Bill