[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Programming for Markup vs. Markup for Programming

• To: xml-dev@l...
• Subject: Programming for Markup vs. Markup for Programming
• From: "Simon St.Laurent" <simonstl@s...>
• Date: Fri, 02 Aug 2002 11:17:06 -0400

This is kind of a followup to the somewhat heated "Markup Perspectives" 
thread, and hopefully it will clarify what I mean even if it fails to make 
people happy.

When XML 1.0 began, it used some fairly simple criteria to create a markup 
technology that was (relatively) easy for programmers to implement and 
use.  I think XML failed the Desperate Perl Hacker test, but it succeeded 
to the point where lots of parsers became available and tools around those 
parsers became available.

Since then, we've seen a substantial change in how we look at the 
relationship between programmers and markup.  Instead of programmers 
implementing tools for working with markup (the kind of programming I do 
whenever I can find time), we have programmers (and vendors who sell to 
programmers, etc.) insisting that XML be recast to meet their needs.

Instead of understanding the pros and cons of markup and abiding by them, 
this crew sees markup as mere object serialization or database 
representation and attempts to pile on the set of understandings needed for 
those tasks, with only cursory attention given to what markup was before 
they came to it, how its structures evolved, or what markup is good at.

I say it's time to kick those bums out.  If they want to use markup, 
they're welcome to do so, but without inflicting piles of understandings 
from their own systems on the rest of us.  W3C XML Schema is a scandal 
because of the mash these folks have given us, and the W3C itself appears 
unable or unwilling to do very much about it.

Developers who want concise representations of strongly-typed information 
should be asking themselves whether verbose textual formats are genuinely 
compatible with their needs.  Sure, ASN.1's a mess, but maybe that just 
means it's time for that community to do to ASN.1 what XML did to 
SGML.  Instead, they appear to creating ever more problems for markup while 
solving their own problems inefficiently at best.  This doesn't seem 
sustainable.

I'd be very happy at this point to have fewer tools and fewer people 
interested in using XML in exchange for people actually focusing on quality 
markup, interoperability, and pushing forward with the "less is more" 
spirit that animated XML 1.0.


Simon St.Laurent
"Every day in every way I'm getting better and better." - Emile Coue

• Follow-Ups:
  • Re: Programming for Markup vs. Markup for Programming
    • From: Mike Champion <mc@x...>
  • Re: Programming for Markup vs. Markup for Programming
    • From: Mike Champion <mc@x...>
  • Re: Programming for Markup vs. Markup for Programming
    • From: David Carlisle <davidc@n...>

• Prev by Date: RE: Markup perspective not code (was RE: Re: URIs, concrete)
• Next by Date: Re: Global/Local attributes
• Previous by thread: Truncation parsing &lt;
• Next by thread: Re: Programming for Markup vs. Markup for Programming
• Index(es):
  • Date
  • Thread