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

RE: SGML complexity (was: RE: Re: Recognizing...)

• From: "Len Bullard" <cbullard@xxxxxxxxxx>
• To: "'Tatu Saloranta'" <cowtowncoder@xxxxxxxxx>, <xml-dev@xxxxxxxxxxxxx>
• Date: Wed, 30 Aug 2006 22:43:25 -0500

How many of the original XML parsers were open source?  With all the noise
this list makes about open source, are you telling me that keeping up SGMLS
or its successor is just too difficult for the markup community?

I don't buy it.  As I said, this isn't a slam on XML.  XML is a Good Thing.
This is a "if you plan to reinvent SGML anyway, maybe it's time to fess up
to needing it and use it rather than making up stories to keep from
admitting that perhaps it was also a good thing".  But let's not have
another 'kill all the lawyers' misadventure just to get the necessaries into
XML because that ensures we'll have to do it that way every generation. "As
the twig is bent..."  

Dusting off SGML and upgrading it would be cheaper than modifying XML.  At
this point, the shoe is on the other foot because there are too many
installed XML systems to fix all of them and there aren't that many
installed SGML systems by comparison.   On the other hand, this is the kind
of thing that ISO has to do or we have to burgle them again.

So we will live with XML, warts and all.  Karma. :-)

len


From: Tatu Saloranta [mailto:cowtowncoder@y...] 

--- Len Bullard <cbullard@h...> wrote:

> On the other hand, the source for SGMLS was out
> there.  Would you write a

Sure, and that is certainly better than starting from
scratch. But at the same time, unless SGMLS does
everything one needs from a parser (and works on all
platforms etc), someone needs to maintain the code.
And maintaining/developing a system that solves a
difficult problem is in itself a difficult problem.
That is: it's not much easier to work on such a system
than writing the system. How many people really know
how, say, gcc++ works, and how to work on it?

So I think that the challenge of working with
something that has expressive power of SGML will
always be there, independent of number of existing
tools and open code base. And given that, using such
systems will be expensive, one way or the other.

-+ Tatu +-


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

• Follow-Ups:
  • RE: SGML complexity (was: RE: Re: Recognizing...)
    • From: Tatu Saloranta
  • RE: SGML complexity (was: RE: Re: Recognizing...)
    • From: peter murray-rust

• References:
  • RE: SGML complexity (was: RE: Re: Recognizing...)
    • From: Tatu Saloranta