RE: A simpler solution to %7e vs %7E vs ~ in namespace comparison

We need to be clear what is the scope of this discussion.  
Though it has appeared to be about namespace names only, 
aren't there other areas where the same questions arise?
I'm thinking especially of RDF.

When considering proposals based on restrained usage, we 
need to consider all areas affected.  In case RDF is within 
the scope of this discussion, I don't think it would be 
viable to forbid hex-escapes in RDF URIs.  Consequently, we 
need equivalence rules which make clear how hex-escapes and 
other anomalous cases are to be treated.

The view of the W3C I18N WG (with which I am, regrettably, 
no longer asociated due to work commitments) has been that 
it is not viable to demand that hex-escapes be recognised 
as such in equivalence matching, other than at the point of 
dereferencing.

Though other WGs may sometimes perceive the W3C's horizontal 
Activities as imposing burdens on them, in this instance at 
least, the I18N WG and the XML Core WG (and, I guess, the 
RDF Core WG) take the same view.  [If this is no longer true, 
I'm sure I shall be corrected.]

Misha


-----Original Message-----
From: Larry Masinter [mailto:LMM@acm.org] 
Sent: 08 February 2003 07:40
To: www-tag@w3.org
Subject: A simpler solution to %7e vs %7E vs ~ in namespace comparison



Don't allow hex-escapes in namespace names.
If hex-escapes are not allowed, then it isn't
necessary to test for equivalence.

Are you aware of any current use of a namespace name with
hex escapes?

In general, don't use, as namespace names,
different URIs that are commonly considered equivalent.
For HTTP URIs in particular, prefer the shorter
and lower case versions under HTTP URI equivalence
rules. That is:

Disallow:

  http://www.example.org:80/path

Use

  http://www.example.org/path

instead. Disallow:

  http://www.EXAMPLE.org/path

Use

  http://www.example.org/path

instead.

This allows namespace name equivalence to be
accomplished by using string equality, because
not ALL URIs are allowed to be namespace names.

I've made this suggestion to 'xml-names-editor',
but since the topic seems to be alive in the TAG
still, I thought I would repeat the suggestion here.

Larry
-- 
http://larry.masinter.net



-------------------------------------------------------------- --
        Visit our Internet site at http://www.reuters.com

Get closer to the financial markets with Reuters Messaging - for more
information and to register, visit http://www.reuters.com/messaging

Any views expressed in this message are those of  the  individual
sender,  except  where  the sender specifically states them to be
the views of Reuters Ltd.

Received on Saturday, 8 February 2003 09:15:24 UTC