[
Lists Home |
Date Index |
Thread Index
]
Still not quite accurate, or at least, still too much to
the side of the opinion that XML is too heavy for mortals.
Given editing systems that can read a DTD and produce a
GUI, did the programmer implement that? In the sense that
they created a piece of code that can read the DTD and
create the GUI, yes? In the sense that once they have
written that code, the author is left to create the DTD, no.
Systems like MS Access enable a person of even less than
DePH skills to create databases and define schemas for them.
Same difference.
The problem here is that most XML tools and experts ignore that class
of user. The ones that begin to understand the role of
the subject matter expert (can analyze the vocabulary
domain, can analyze the business proceses that use that
domain), will be a lot more successful than those that
believe this is programmer work. It was PRECISELY to
get this sort of thing out of the hands of the programmers
that markup was adopted by some organizations. They needed
a middle ground where the SME and the system could negotiate
to produce more effective local definitions. WYSIWYG was
ultimately a failure; we needed the databases but we needed
a cheaper and more user-centric way to get the data into
and out of them.
That was the lesson of SGML. We watched the easy to
use tools such as SoftQuad quickly outpace the ArborTexts
in terms of numbers of installations.
We watched the Datalogics go down in flames.
We began to understand that the end-user was the SGML user.
We kept the faith on that and some of us began to develop
stylesheet driven systems for hypertext that let the
end user develop both layout and tags and enabled those
to be the same person/role or different.
It was working marvelously until the advent of the WWW,
HTML, and the XML coup d'etat.
len
From: Aaron Skonnard [mailto:aarons@develop.com]
Fair enough. It would have been more accurate to say "markup users are
simply using pre-defined vocabularies *implemented* by programmers."
|
