Lists Home |
Date Index |
10/8/2002 9:56:15 AM, Sam Hunting <firstname.lastname@example.org> wrote:
>XML is great, I would never trash it, but it doesn't meet all
>requirements. I see the emergence of little languages developed by markup
>people as a very encouraging, grass roots sign.
I argued against that for a long time, but have Seen the Light. I had thought
that the "network effect" of ubiquitous XML processors would motivate
people to buck up and learn the basics of XML. For better or worse,
this ain't happening: XML is overkill for simple, non-hierarchical data
(c.f. "RSS 3.0") and underkill for complex text that doesn't fit into
a neat hierarchy (c.f. LMNL). People are inventing alternate syntaxes to ease the
pain, and putting in a transformation step in their processing pipeline
to XML-ize or HTML-ize it. (Wiki ML comes to mind along with others that
Eric mentioned, along with the apparent demise of the XML syntax for XQuery).
Which gives me a segue to plug http://www.artima.com/intv/perfect.html on
yet another mailing list: (I think ERH plugs in in the XOM stuff too).
'To me, thinking from the user in is the designer's job. If you just want to implement,
you should work for somebody who knows how to design. If you want to design, you have
to ask, "Who am I designing for? What do they want to accomplish? And what do they
have at their disposal to accomplish it?" '
I suspect that people can be more easily taught to think in terms of structured
text that an implementation can turn into XML than in raw XML, and that what
they have at their disposal are text editors. (Hopefully word processors will
soon produce configurable XML easily, see the Office 11 trade press articles,
Corel's acquisition of SoftQuad, etc.) But until these are ubiquitous, it's
going to be the little languages that people actually use to produce XML.
Question: this was a much-touted advantage of SGML, but did/do off-the-shelf
SGML processors (Epic, XMetal, SP, etc.) support this kind of thing? Or was it
something that people wrote custom code to handle for every "little language"?