[
Lists Home |
Date Index |
Thread Index
]
>As for Mozilla (dmoz.org?) and RSS 1.0, to what extent are these simply XML
>formats that just happen to be RDF-friendly, as far as most users
>are concerned? In other words, what can I do with RSS 1.0 or dmoz.org
>data because it is RDF?
Point of information : Mozilla is the open source browser, the codebase
behind Netscape (which uses RDF), dmoz.org is the Open Directory (which uses
RDF).
I think even the most rabid RDF fanatic (RRDFF?) would have to admit that
pretty much anything you can do with RDF could be done using custom XML, but
I reckon the biggest practical benefit is that the RDF allows extensibility
with minimal extra work. You gave me an easy example with RSS 1.0 and
dmoz.org: lets say I'm a news syndicator that covers a really wide range of
topics. The basic syndication feed could be done using RSS, and each news
item could have its topic identified using a dmoz.org category. Ok, there
will be processing work involved at both the distribution and reception
sides, but the expression of the information is a no-brainer and what's more
the information will be expressed in an umambiguous way that existing and
future RDF apps will understand - if the news items had a permanent page
then it could be fed into the dmoz system essentially unchanged, and the
feeds should be readable using any half-decent existing RSS reader, just the
additional info will be ignored.
>(Please don't say "use any RDF tool to process
>it" :-)
I came close...
>I don't have no stinkin' RDF tools around because I can't think of
>anything useful to do with them!... I'm asking what concrete
>applications/ontologies/whatever can be brought to bear on these specific
>data sources)
The chicken and egg situation has hampered things a lot - no-one developing
tools because the RDF-expressed data wasn't there and vice versa. However
initiatives like Adobe's are in effect filling up the backend, and
server-side tools are slowly beginning to appear. There still hasn't been
much activity at the client-side/desktop (with the exception perhaps of
Netscape) but the fact that people like Mitch Kapor (of Lotus 1-2-3 fame)
are looking to use RDF in their applications would suggest that it is only a
matter of time.
A couple of server-side examples that can be played with now : the TAP demos
[2] (Google augmented with semantic search) and MusicBrainz (music/CD
database - only has one entry for 'Nocturnal Emissions' so far, so still a
long way to go ;-)
Cheers,
Danny.
[1] http://blogs.osafoundation.org/mitch/000007.html
[2] http://tap.stanford.edu/tap/demos.html
[3] http://www.musicbrainz.org/
|
