Re: Tag ontology RFC

From: Richard Newman <r.newman@reading.ac.uk> · Date: Sat, 26 Mar 2005 10:35:09 +0000

Danny,

On Mar 26, 2005, at 10:16, Danny Ayers wrote:
>> i.e. you exploit some additional knowledge about the sources you want
>> to integrate (in this case that you know that del.icio.us's feed
>> contains tags) to generate the integrating triples.
>
> Yep, that makes a lot of sense. It would be nice to have that
> integrated without stepping outside RDF/OWL (I really must read docs
> ;-)

I expect that with inference support it would be quite straightforward 
to chain domains and ranges; assert that del.icio.us's feed is a 
TagFeed, and boom (though I've not done that level of ontology design 
for 6 months, so I'm probably wrong :))

There're always rules to fall back on if that fails/takes too 
long/isn't possible due to the absence of a DL engine.

>> I really have to get some code done for this... perhaps a little 
>> Easter
>> project for myself. Everyone seems to be releasing del.icio.us-alikes
>> recently (I'm thinking of del.irio.us)!
>
> Heh, I've been working very hard to resist doing it with the code I'm
> currently working on, not a priority just now. But I've love to see
> how some experimentation with tags and RDF turns out. Happy Easter!

Then I hope to get a Lisp + Araneida + Wilbur + Ajax tag-based Web 
application done in a few hours later today ;)
Happy Easter indeed!

> Regarding del.irio.us, I just signed up to the del.icio.us list [1]

Very interesting! Another one for my inbox, thanks.

> and the first post that came through is below [2]. I reckon the points
> made are probably very significant in the context of RDF deployment -
> there's a lot of potential in Open Data...

Are you on the DataLibre list? It has occasional discussions on 
microcontent that I always try to butt back to RDF :) It's run by Steve 
Mallett, who's the author of del.irio.us, so there's a tie-in.

<http://www.osdir.com/mailman/listinfo/datalibre-discuss>

A lot of the issues facing microcontent are the same as for RDF, so I 
think both sides could learn from all this. At least, that's how I 
justify cluttering up sw@w3 with this lot ;)

> [[
> Hopefully this first public re-implementation of del.icio.us will move 
> this
> debate forwards - it's been stuck in a loop for months.
>
> The core value of del.icio.us is now purely the data set and the 
> user-base. The
> users own the data set in a distributed way.

That's the DataLibre talking ;) I quite agree, of course. This is one 
reason why I think it's vital to have date and tagger involved in the 
relation; it makes dragging the data around the Web much easier. 
Somewhere on the core feature list really ought to be "get MY RDF".

> Distributed storage / syndication / aggregation with some kind of
> lisence/agreement about ownership of and contribution to the data set 
> as a
> whole would enable anyone to develop an interface with whatever 
> features,
> without dissipating the core value.
>
> Then we can kill google.

Now that's a frightening goal! :)

Cheers,
-Richard