Danny Ayers | 3 Feb 11:31 2009
Picon

[foaf-dev] Minor inconsistency in spec doc

see http://xmlns.com/foaf/spec/#term_interest

[[
Property: foaf:interest
interest - A page about a topic of interest to this person.
Status: 	testing
Domain: 	foaf:Person
Range: 	foaf:Document

The foaf:interest property represents an interest of a foaf:Agent,
...
]]
the last text line here suggests the domain is foaf:Agent, though the
RDF version has foaf:Person

Personally I think it should be foaf:Agent throughout, a bit more
flexible, though won't lose any sleep with foaf:Person.

Also it would be handy to include a reasonably rich example of a
profile, to help copy & paste creation.

Cheers,
Danny.

--

-- 
http://danny.ayers.name
Olivier GENDRIN | 3 Feb 11:41 2009
Picon

Re: [foaf-dev] Minor inconsistency in spec doc

On Tue, Feb 3, 2009 at 11:31 AM, Danny Ayers <danny.ayers@...> wrote:
> Personally I think it should be foaf:Agent throughout, a bit more
> flexible, though won't lose any sleep with foaf:Person.

I agree, it would allow to indicate the foaf:interest of a company, a NGO, ...

--

-- 
Olivier G.
http://www.lespacedunmatin.info/blog/
Dan Brickley | 3 Feb 11:49 2009

Re: [foaf-dev] Minor inconsistency in spec doc

On 3/2/09 11:41, Olivier GENDRIN wrote:
> On Tue, Feb 3, 2009 at 11:31 AM, Danny Ayers<danny.ayers@...>  wrote:
>> Personally I think it should be foaf:Agent throughout, a bit more
>> flexible, though won't lose any sleep with foaf:Person.
>
> I agree, it would allow to indicate the foaf:interest of a company, a NGO, ...
>

Good catch Danny. I agree it would be useful to have this for foaf:Agent.

Any objections to broadening the rdfs:domain from foaf:Person?

Dan
Danny Ayers | 3 Feb 12:14 2009
Picon

[foaf-dev] Twitter Pack of semwebbers (interest domain Agent desirable)

Marjolein Hoekstra has compiled a Twitter Pack of folks that are
interested in the Semantic Web [1] - essentially it makes it easy to
subscribe to the whole lot. I just whipped up a dirty little Python
script [2] to express this in RDF/XML [3].

What would be nice would be to have the corresponding Twitter bio
material in RDF/FOAF too - pretty sure this should be doable via the
Twitter API but I leave this as an exercise for the reader :-)

In general expressing a Twitter user as a Person is a bit clunky,
there are quite a few Twitter Agents around, e.g.
http://twitter.com/Earthquack

Cheers,
Danny.

[1] http://twitterpacks.pbwiki.com/Semantic-Web
[2] http://twitterpacks.pbwiki.com/f/twitpack2rdf.py
[3] http://twitterpacks.pbwiki.com/f/sempack.rdf

--

-- 
http://danny.ayers.name
Olivier GENDRIN | 3 Feb 12:34 2009
Picon

[foaf-dev] foaf:accountAccesspage

Hi list !

Looking at Danny RDF version of the semwab twitter pack [1], I think
that we may need a foaf:accountAccesspage in foaf:OnlineAccount, in
order not to have to guess the account page URL just with
foaf:accountServiceHomepage and foaf:accountName

Because :
<foaf:OnlineAccount>
  <foaf:accountServiceHomepage rdf:resource="http://twitter.com"/>
  <foaf:accountName>lespacedunmatin</foaf:accountName>
</foaf:OnlineAccount>

is http://twitter.com/lespacedunmatin

but

<foaf:OnlineAccount>
  <foaf:accountServiceHomepage rdf:resource="http://flickr.com"/>
  <foaf:accountName>olivier_g</foaf:accountName>
</foaf:OnlineAccount>

is http://flickr.com/people/olivier_g/

and

<foaf:OnlineAccount>
  <foaf:accountServiceHomepage rdf:resource="http://www.linkedin.com"/>
  <foaf:accountName>olivierg</foaf:accountName>
</foaf:OnlineAccount>
(Continue reading)

Damian Steer | 3 Feb 12:35 2009
Picon

Re: [foaf-dev] Twitter Pack of semwebbers (interest domain Agent desirable)

Danny Ayers wrote:

> What would be nice would be to have the corresponding Twitter bio
> material in RDF/FOAF too - pretty sure this should be doable via the
> Twitter API but I leave this as an exercise for the reader :-)

It's worth looking to see if they are identi.ca (or, more generally,
laconica) users, since they have foaf profiles.

I'm not aware of any method to get an identi.ca id from twitter,
however. Anyone aware of any conventions around that?

Damian
Dan Brickley | 3 Feb 12:39 2009

Re: [foaf-dev] Twitter Pack of semwebbers (interest domain Agent desirable)

On 3/2/09 12:35, Damian Steer wrote:
> Danny Ayers wrote:
>
>> What would be nice would be to have the corresponding Twitter bio
>> material in RDF/FOAF too - pretty sure this should be doable via the
>> Twitter API but I leave this as an exercise for the reader :-)
>
> It's worth looking to see if they are identi.ca (or, more generally,
> laconica) users, since they have foaf profiles.
>
> I'm not aware of any method to get an identi.ca id from twitter,
> however. Anyone aware of any conventions around that?

You might try Google Social Graph API. I looked a while ago, but it 
didn't seem to have a very fresh index of these sites though.

Dan
Story Henry | 3 Feb 12:55 2009
Picon

Re: [foaf-dev] Minor inconsistency in spec doc


On 3 Feb 2009, at 11:49, Dan Brickley wrote:

> On 3/2/09 11:41, Olivier GENDRIN wrote:
>> On Tue, Feb 3, 2009 at 11:31 AM, Danny  
>> Ayers<danny.ayers@...>  wrote:
>>> Personally I think it should be foaf:Agent throughout, a bit more
>>> flexible, though won't lose any sleep with foaf:Person.
>>
>> I agree, it would allow to indicate the foaf:interest of a company,  
>> a NGO, ...
>>
>
> Good catch Danny. I agree it would be useful to have this for  
> foaf:Agent.
>
> Any objections to broadening the rdfs:domain from foaf:Person?

no objections from me.

> Dan
> _______________________________________________
> foaf-dev mailing list
> foaf-dev@...
> http://lists.foaf-project.org/mailman/listinfo/foaf-dev
B.K. DeLong | 3 Feb 13:50 2009
Picon

Re: [foaf-dev] Twitter Pack of semwebbers (interest domain Agent desirable)

It would be interesting to write something that would create a short
of "x people you are following are also following y" in general
(unless that exists).

How are the semwebbers identified in this case? By an interest or keyword cloud?

On 2/3/09, Danny Ayers <danny.ayers@...> wrote:
> Marjolein Hoekstra has compiled a Twitter Pack of folks that are
> interested in the Semantic Web [1] - essentially it makes it easy to
> subscribe to the whole lot. I just whipped up a dirty little Python
> script [2] to express this in RDF/XML [3].
>
> What would be nice would be to have the corresponding Twitter bio
> material in RDF/FOAF too - pretty sure this should be doable via the
> Twitter API but I leave this as an exercise for the reader :-)
>
> In general expressing a Twitter user as a Person is a bit clunky,
> there are quite a few Twitter Agents around, e.g.
> http://twitter.com/Earthquack
>
> Cheers,
> Danny.
>
> [1] http://twitterpacks.pbwiki.com/Semantic-Web
> [2] http://twitterpacks.pbwiki.com/f/twitpack2rdf.py
> [3] http://twitterpacks.pbwiki.com/f/sempack.rdf
>
> --
> http://danny.ayers.name
> _______________________________________________
(Continue reading)

Benjamin Nowack | 3 Feb 13:57 2009

Re: [foaf-dev] foaf:accountAccesspage


I think I saw foaf-exporting sites use the profile pages as 
direct identifiers for accounts, e.g.

<foaf:OnlineAccount rdf:about="http://twitter.com/lespacedunmatin">
...

and IIRC, DanBri was suggesting (or questioning?) this approach
here and there recently.

I'm using this in my apps, too (deliberately ignoring the 
"can a URL denote a FOAF account?" metaphysics ;).

Best,
Benji

--
Benjamin Nowack
http://bnode.org/
http://semsol.com/

On 03.02.2009 12:34:30, Olivier GENDRIN wrote:
>Hi list !
>
>Looking at Danny RDF version of the semwab twitter pack [1], I think
>that we may need a foaf:accountAccesspage in foaf:OnlineAccount, in
>order not to have to guess the account page URL just with
>foaf:accountServiceHomepage and foaf:accountName
>
>Because :
(Continue reading)


Gmane