Re: Issue http://www.w3.org/2000/03/rdf-tracking/#rdf-ns-prefix-confusioN

Graham Klyne <GK-lists@ninebynine.org> wrote:

>> Perhaps something like:
>> 
>>  [[
>>    Unqualified elements and attributes in RDF are discouraged
>>  ]]
>> 
>> would be better.
> 
> Hmmm... sounds like a fudge to me.  If there is a problem with unqualified
> attributes, then I think we'd do better to disallow them;  (maybe use
> language like "... MUST generate namespace qualifiers, but MAY/SHOULD
> recognize unqualified attributes"?)

Sorry for responding so late, your mail got lost in my inbox somehow...

Anyway, the result from the telecon[1] felt even fudgier to me:

    RESOLVED: We'll strongly recommend the use of namespace qualified
    attributes, and allow but deprecate unqualified attributes.

Afterwards, DanC asked[2]:

    stronly deprecate"???? I wonder what that means. Does the test pass or
    not?

I think that if we decide to come up with some system of conformance to the
new spec, unqualified attributes should not conform, but I think the
practice should be noted and processors should be allowed to support it for
backwards-compatibility.

Hmm, this tightrope-walking bit is tricky.

[1] 
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/att-0017/01-2001-
04-27.html
[2] http://ilrt.org/discovery/chatlogs/rdfcore/2001-04-27.html#T15-01-10
-- 
[ Aaron Swartz | me@aaronsw.com | http://www.aaronsw.com ]

Received on Tuesday, 8 May 2001 22:02:26 UTC