W3C home > Mailing lists > Public > www-mobile@w3.org > May 2002

RE: Proposal: Values for UAProf properties

From: Butler, Mark <Mark_Butler@hplb.hpl.hp.com>
Date: Wed, 29 May 2002 18:08:33 +0100
Message-ID: <5E13A1874524D411A876006008CD059F0450187D@0-mail-1.hpl.hp.com>
To: www-mobile@w3.org

Hi Johannes, Vidhya, Nick and Tayeb

A couple of points:

> If you want to make a proposal for updating the UAProf 
> specification and
> you are a member of the WAP forum, I strongly recommend that 
> you make a
> Change Recommendation to the WAP forum's uaprof list
> 'wap-uaprof@mail.wapforum.org', attend a Wireless Application Group
> meeting and get it pushed through. 

Unfortunately the wap-uaprof email list no longer exists. I've queried this
with the WAP Forum and comments on UAProf are now supposed to got to
WAP-WAG, although I know Nick has posted questions to this list (about the
confusion between namespace URIs and schema URLs) but has not received a
reply. Your suggestion about going to a meeting is undoubtedly a good one
though - I may need to do this at some point just to talk about CC/PP and
JSR 188. Also I have tried to submit change recommendations to the WAP Forum
and they have never acknowledged them so I guess they have been ignored. 

As for the discussion about what the schema indicates, there are clearly
differences between the schemas retrievable at the URLs and the schemas in
the WAP documentation. Another problem is the proliferation of schemas:
every UAProf phone I come across seems to use a different namespace, which
is leading to a configuration nightmare.  As IBM have noted this will also
lead to problems when you try to merge requests that used to different
versions of the uaprof vocabulary.

What can we do to avoid this? Well one idea I've had would be to create a
schema that merges the existing UAProf schemas. Where properties have
changed component, resolution rule or data type between versions these would
either be marked as i) undefined or ii) all of the above. This schema would
then be frozen. Then if people want to add new properties, they would have
to use a new namespace and provide a schema there. I think it is much better
to leave existing properties in the existing namespace than copy them to a
new namespace every time a new version of the vocabulary is created. This
just creates a backward compatibility problem. 

So I have a question for people here: do people agree that UAProf has
reached a point where it is finished or do they think it still needs further
work? If you do not want to make your opinion public, send it to me directly
and I will tally the result to anonymise opinions?

thanks, best regards

Mark H. Butler, PhD
W3C CC/PP WG Chair
Research Scientist                HP Labs Bristol
Internet: http://www-uk.hpl.hp.com/people/marbut/
Received on Wednesday, 29 May 2002 13:10:55 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:16:03 UTC