W3C home > Mailing lists > Public > public-rdf-dawg-comments@w3.org > June 2004

Re: On Requirement 3.8 (Bookmarkable Queries)

From: <massimo@w3.org>
Date: Wed, 23 Jun 2004 17:56:28 -0400 (EDT)
Message-ID: <3714.128.30.5.194.1088027788.squirrel@128.30.5.194>
To: kendall@monkeyfist.com
Cc: massimo@w3.org, public-rdf-dawg-comments@w3.org

> On Wed, Jun 23, 2004 at 05:14:12PM -0400, massimo@w3.org wrote:
>>
>> On Requirement 3.8 (Bookmarkable Queries), cf.
>> http://www.w3.org/TR/2004/WD-rdf-dawg-uc-20040602/#r3.8 ,
>> a question of clarification:
>> Does "express as a URI" means a new class of URIs will
>> be created? Or does it rather mean "embeddable in a URI"
>>
>> Obviously the first interpretation (esp. as a MUST) is
>> rather demanding. In the case of the first interpretation,
>> given the costs of creating a new URI scheme, what is
>> the rationale/benefit of going with the new URI scheme
>> rather than with the classic container approach?
>
> I'm not speaking for the original proposer of this requirement, but
> the understanding of the WG (which I think I understand and can
> represent) is "embeddable in an HTTP URI". No one on the WG has
> suggested we register a new URI scheme.
>
> There is an outstanding ACTION item to rework 3.8 to remove this
> ambiguity.

Thanks Kendall,
that clarifies my doubt (in the best possible way actually).
If this requirement passes, you might consider adding (not
in the req. doc. but in the language doc) warning
notes on the danger of long-URLs (eg the 16k limits, etc), so
that people know that long DAWG queries can become potentially
dangerous when URLed, and not be working on some systems. Anyway,
that's obviously an implementation note and something to
ponder, not a stopover to go on with the requirement (if the
wg decides to approve it).

-M
Received on Wednesday, 23 June 2004 17:56:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:14:47 GMT