atom feed42 messages in org.w3.www-tagRE: XML-SW, a thought experiment
FromSent OnAttachments
Tim BrayFeb 6, 2002 8:42 pm 
David OrchardFeb 6, 2002 10:00 pm 
Jacek KopeckyFeb 7, 2002 3:02 am 
Elliotte Rusty HaroldFeb 7, 2002 4:14 am 
Elliotte Rusty HaroldFeb 7, 2002 4:17 am 
Elliotte Rusty HaroldFeb 7, 2002 4:28 am 
Elliotte Rusty HaroldFeb 7, 2002 4:51 am 
Tim BrayFeb 7, 2002 10:12 am 
Tim BrayFeb 7, 2002 10:13 am 
Piotr KaminskiFeb 8, 2002 4:34 am 
Tim Berners-LeeFeb 12, 2002 12:03 pm 
Paul PrescodFeb 12, 2002 2:21 pm 
Norman WalshFeb 12, 2002 2:29 pm 
Dan ConnollyFeb 12, 2002 2:33 pm 
Simon St.LaurentFeb 12, 2002 3:05 pm 
Simon St.LaurentFeb 12, 2002 3:46 pm 
Martin DuerstFeb 12, 2002 4:00 pm 
Jacek KopeckyFeb 13, 2002 4:53 am 
Norman WalshFeb 13, 2002 7:00 am 
Jacek KopeckyFeb 13, 2002 7:40 am 
Elliotte Rusty HaroldFeb 13, 2002 8:07 am 
Norman WalshFeb 13, 2002 8:14 am 
Jacek KopeckyFeb 13, 2002 10:05 am 
Paul PrescodFeb 13, 2002 10:27 am 
MURATA MakotoFeb 13, 2002 6:12 pm 
Rick JelliffeFeb 13, 2002 11:38 pm 
Norman WalshFeb 14, 2002 6:07 am 
Jun FujisawaFeb 16, 2002 5:44 am 
Tim Berners-LeeFeb 22, 2002 7:56 am 
Eric van der VlistFeb 22, 2002 8:07 am 
Eric van der VlistFeb 23, 2002 1:23 am 
Piotr KaminskiFeb 23, 2002 3:18 am 
Elliotte Rusty HaroldFeb 23, 2002 9:06 am 
Paul PrescodFeb 23, 2002 9:31 am 
Eric van der VlistFeb 23, 2002 12:01 pm 
Piotr KaminskiFeb 23, 2002 3:47 pm 
Elliotte Rusty HaroldFeb 23, 2002 4:26 pm 
Piotr KaminskiFeb 23, 2002 5:21 pm 
Elliotte Rusty HaroldFeb 23, 2002 7:36 pm 
Tim Berners-LeeMar 4, 2002 11:40 am 
Tim Berners-LeeMar 4, 2002 1:14 pm 
Rick JelliffeMar 25, 2002 7:00 am 
Subject:RE: XML-SW, a thought experiment
From:Elliotte Rusty Harold (elh@metalab.unc.edu)
Date:Feb 7, 2002 4:51:32 am
List:org.w3.www-tag

At 10:01 PM -0800 2/6/02, David Orchard wrote:

Tim,

This is really great stuff. While I think that PIs should also be lopped off,

I wouldn't. PIs are quite useful in practice. Eliminating them would hurt way too much. It is useful to be able to step out of the markup for processor-specific uses like PHP and stylesheets.

and XInclude for entity replacement and an optional XML Schema validation level added, I can certainly live with this.

These can be layered quite cleanly on top of well-formed XML for those who need them. I don't see any reason for putting them in the core spec. With regard to schemas (and DTDs for that matter) I do wonder where default attributes fit in. Just part of the PSVI but not the pre-SVI I suppose. --