RE: Grinding to a halt on Issue 27. from Joshua Allen on 2003-04-30 (www-tag@w3.org from April 2003)

From: Joshua Allen <joshuaa@microsoft.com>
Date: Wed, 30 Apr 2003 07:19:08 -0700
To: <robin.berjon@expway.fr>, "Elliotte Rusty Harold" <elharo@metalab.unc.edu>
Cc: <www-tag@w3.org>
Message-ID: <4F4182C71C1FDD4BA0937A7EB7B8B4C108E89078@red-msg-08.redmond.corp.microsoft.com>

> No, it's not. Namespaces clearly specifies that they are to be
> differentiated based on string comparison. We know that IRIs will be 
> strings. We can safely have a dependency on them, it'll never break
the 
> spec. People using IRIs too early may get breakage, but otherwise 
> everything will be fine.

This dance of "strcmp for namespaces and canonicalization for anyUri"
has led to a status quo that is buggy and ambiguous, particularly for
any non-trivial scenario.

IMO, it's more important to do this right than to do it soon.  Unless we
do it right, we will just make things worse than they already are.

Received on Wednesday, 30 April 2003 10:19:17 UTC