RE: SD5 - Namespaces - New Version 2

OK. Good.  It will be nice when we can enforce various conventions, but
if SGML does not give us a way to do that yet, I'm willing to work with
what we have.

--Andrew Layman
   AndrewL@microsoft.com

> -----Original Message-----
> From:	Steven J. DeRose [SMTP:sjd@eps.inso.com]
> Sent:	Friday, May 23, 1997 12:18 PM
> To:	w3c-sgml-wg@w3.org
> Subject:	RE: SD5 - Namespaces - New Version 2
> 
> At 11:47 AM 05/23/97 -0700, Andrew Layman wrote:
> >What happens?  Nothing special.  You have a long, strange name.
> >However, I suspect that you mean to imply something that I miss.
> >
> >Frankly, I would prefer that colon were reserved to separate the
> >namespace part of a name from the rest, but I gather this would be
> hard
> >to fit into the confines of SGML.
> 
> Not really; SGML can easily be set to allow colon in names; in its
> current
> state it just could not *validate* that you only use the colon that
> way.
> But, it can't validate that you use semicolon on the end of entity
> references or that you only use #PCDATA at the beginning of a content
> model,
> either. Those are all XML conventions that restrict the range of SGML
> alternatives that XML may also use.
> 
> Steven J. DeRose, Ph.D., Chief Scientist
> Inso Electronic Publishing Solutions
>    (formerly EBT)

Received on Friday, 23 May 1997 19:26:02 UTC