RE: working around the identity crisis

On 16.11.2004 16:58:03, Miles, AJ (Alistair) wrote:
>
>
>> The technical problem is that using frag ids raises issues when you
>> have a large set of terms and then try to efficiently GET a term's URI
>> in order to receive a description (or if you would like to
>> provide term
>> descriptions at a term's URI).
>
>Just to say that, as Dave Reynolds pointed out a little while ago on
>public-esw-thes list, this is not necessarily true, as you can do things
>like
>
>   http://my.org/knowlegebase/chemistry/water#concept

>    http://my.org/knowlegebase/chemistry/ice#concept

>or
>    http://my.org/knowlegebase/chemistry/water#Water

>    http://my.org/knowlegebase/chemistry/ice#Ice

>
>... see his posting at
>
>http://lists.w3.org/Archives/Public/public-esw-thes/2004Sep/0016.html

heh, sure. but this is a "fake" solution, as the different URIs *are*
built using slashes ("../water", "../ice"). Dave simply adds an
additional fragment. GETting "../water", "../water#concept", or
"../water#Water" will always return the same result (unless you're
using something like uriqa). These are all different URI(ref)s, but
you wouldn't be able to define a shared namespace for multiple terms
which would still allow to *efficiently* (added on purpose) GET a
(single) term's description. This may not neccessarily apply to SKOS
concept schemes, but it does apply to RDF term sets.

>Not that I'm expressing any sort of opinion on whether hash or slash is best
>;)
Me neither ;)
I actually like  #-vocabs very much as they are usually using the
rdf:ID construct which makes them easier to read (and yes, "view-source"-
friendliness *can* help the SemWeb grow ;). but I see the problems that
arise when someone tries to create #-URIs for individuals. using slash,
but auto-adding fragIDs, could be a nice solution both camps could be
happy with, though. (provided that tools supported such a feature. I'll
have a thought or two on that..)

regards,
benjamin

--
Benjamin Nowack

Kruppstr. 100
45145 Essen, Germany
http://www.bnode.org/


>Al.
>
>
>>
>> >If this has been discussed to death before, please feel free
>> to tell me
>> >so and/or provide a pointer to such a discussion.
>> one related thread ("pound sign vs. slash as final URI delimiter")
>> starts at [1]
>>
>> [1]
>> http://lists.w3.org/Archives/Public/www-rdf-interest/2004Feb/0117.html

>>
>> regards,
>> benjamin
>>
>> --
>> Benjamin Nowack
>>
>> Kruppstr. 100
>> 45145 Essen, Germany
>> http://www.bnode.org/

>>
>> @ DERI Galway from 2004-10-01 to 2004-12-02
>> http://www.deri.ie/

>>
>>
>>
>> >
>> >Cheers,
>> >
>> >Jeen

Received on Tuesday, 16 November 2004 18:17:44 UTC