Re: Bitzi File Metadata RDF Dump

> I'm certainly all for minimizing the overhead of deploying
> sources of knowledge on the web, but using HTTP URLs
> as URNs for abstract resources or location independent
> identities creates far more problems [...]

I think that you misread Dan's suggestion. At what point did Dan suggest
using HTTP URIs as URNs? He merely gave the highly valid suggestion that a
property be created to disambiguate a type of (most likely literal) value
as a bitprint.

> Perhaps what is needed is the realization of the 'vnd.'
> URN namespace subtree, or similar,

Try using informal URN namespace IDs, per section 4.0 III of RFC 2611.
That's exactly what they are meant for.

> HTTP URLs are notoriously fragile as long term identifiers.

It depends upon how you define fragile. If they were really that fragile,
then the HyperText Web would not sufficiently work! Changing a link in an
HTML page is only a little easier than changing a URI in an RDF file.
Semantic Web code is meant to be geared to being able to express
equivalence relationships so that all of this stuff is easy to update. You
don't need to fear using HTTP URLs for concepts on the Web much more than
you need to fear link rot in HyperText, but you are certainly free to go
for the non-location based alternatives. However, be wary that the more
people use a scheme, the more chance it has of being "universal". URNs
don't have as much unversality as HTTP URLs.

> Faking URNs with PURLs or similar tricks and calling them
> HTTP URLs is just avoiding the real issue here, IMO (and
> of course, puts you at the mercy of a centralized organization
> to manage those PURLs, no? ;-)

Well, yeah... but I don't think it avoids the issue. PURLs are delagated by
a central authority, so are URNs. So what?

> We certainly do need a decentralized means for defining
> URNs, [...]

Huh? You need agreement. That's not necessarily centralization... it's
consensus. But I wouldn't particularly call it decentralization either.
There's no such thing as perfect decentralization.

[...]
> Using HTTP URLs as URNs where they identify abstract resources is
> IMO a total abuse of the HTTP URI scheme and in violation of the
> explicitly defined purpose of HTTP URLs.

No, because HTTP as a protocol can be extended to identify anything. In
fact, there are already response codes suitable for a great range of
purposes.

[...]
> Those of us who are concerned with very-long term persistence
> and maximal portability of resource identities will opt for supporting
> and helping to optimize the centralized agencies necessary for their
> use.

That must be why I support PURLs, URNs, informal URN registration
practices, TAG URIs, and came up with the PTS namespace then, eh?

[...]
> You *can* be an HTTP URL cowboy if you like, but that won't
> bring law and order to the frontier of the semantic web...

I think that accusing Dan of being an HTTP URL cowboy is rather absurd, but
at least it made me laugh :-) URLs work enough to be useful within the
context of their use, and if you don't believe me, try browsing the Web.
Some of the context of that use includes naming things, just the same as we
need to name things in RDF. But yes, I agree that we should not be
restricted to using HTTP URIs, and that the alternaitves are far more
alluring in many cases.

Cheers,

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .

Received on Wednesday, 26 September 2001 13:56:37 UTC