[
Lists Home |
Date Index |
Thread Index
]
On 2002-02-19 23:21, "ext Paul Prescod" <paul@prescod.net> wrote:
> Michael Brennan wrote:
>>
>> ...
>>
>> This is a minor point, but it points to a more important point. The biggest
>> fly in the ointment, here, is the common use of "http:" URIs that don't
>> point to anything. I personally consider that to be very bad practice. I
>> don't think we can achieve consensus unless we agree that abusing
>> well-defined URI schemes for abstract URIs is a bad practice. If it's
>> abstract and does not point to anything, use a URN.
>
> From a usability point of view I'm on your side. And in fact I tried to
> register an abstract URN namespace for XML namespaces back when it was
> hard to do that.
>
> But consider that once you've made the choice to make a URI "abstract"
> there is no way to go back. You can't just pop up a web page and change
> your mind. If RDDL or something else takes over the universe you are
> stuck.
Your argument seems to presume that HTTP is and will be the only
way to get information about resources.
While I admit that HTTP is here now, and it works, that doesn't
mean that it is the optimal solution over the long run for
every application relating to knowledge about resources.
HTTP is a darn good hammer, but KR and the SW need more than
nails.
(RDDL, with some tweaks, makes a workable nail in the interim,
though)
Patrick
--
Patrick Stickler Phone: +358 50 483 9453
Senior Research Scientist Fax: +358 7180 35409
Nokia Research Center Email: patrick.stickler@nokia.com
|
