• To: XML Developers List <xml-dev@l...>
• Subject: RE: XML-with-datatypes (was....)
• From: Vladimir Gapeyev <vgapeyev@s...>
• Date: Thu, 13 Oct 2005 19:52:40 -0400 (EDT)
• In-reply-to: <15725CF6AFE2F34DB8A5B4770B7334EE0720754C@h...>
• References: <15725CF6AFE2F34DB8A5B4770B7334EE0720754C@h...>

On Thu, 13 Oct 2005, Bullard, Claude L (Len) wrote:

> A refactorization of XML that focused on fixing the
> known syntax problems (layering) without introducing
> another layer is welcome.  Welding in datatypes is
> precisely the kind of unwelcome change one wants to
> avoid.  (Vladimir:  I'm not saying what you are
> suggesting is ridiculous.  I don't see the advantage
> of moving the namespace to core, but please keep
> that discussion going.)

[Clarification: I was talking about putting a mechanism for datatype 
identification into the core; namespaces are already there.]

Thanks for the encouragement, but I am puzzled: I feel you suggest that 
this welding-in of datatypes would _hurt_ someone.  That's what I don't 
understand: how a backward-compatible add-on extension in the style of 
namespaces or xml:ID can do that?  Those who don't need it, happily 
don't have to bother about its existence.

More likely my suggestion does not solve the problem well, solves a 
wrong problem, or there isn't enough need to have the problem solved!

VG

• References:
  • RE: XML-with-datatypes (was....)
    • From: "Bullard, Claude L (Len)" <len.bullard@i...>

• Prev by Date: Re: XML-with-datatypes (was....)
• Next by Date: Open Document Format -- Successor to XHTML? (interview with Gary Edwards)
• Previous by thread: RE: XML-with-datatypes (was....)
• Next by thread: Re: XML-with-datatypes (was....)
• Index(es):
  • Date
  • Thread