Re: Crowdsourcing request: Google People Finder Data as RDF

Hi,

I am interested to help for this project.  I have about than 10 years
experience with semantic web technology and it is my dog food everyday. I
had the idea of doing it during the Haiti Earthquake. I looked at the People
Finder Interchange Format (PFIF) that is used by google
http://zesty.ca/pfif/1.2/ . The problem is XML format is mainly its fixed
structure and its difficulty to extend it for specific purpose (like
address).
I would be interested to work on developing core ontology that would fix the
defect of PFIF and then use it as a foundation to develop extensions. There
are other ontologies that could be taken in account such as Sahana
http://ontology.nursix.org/sahana-person.owl.    I think it is important we
do it right that just going to a straight conversion from PFIF format. It
requires some effort but it should pay off in the long term.
I would appreciate if you can tell me where to start (forum, wiki, code base
...etc)

Best regards
Stephane Fellah



On Thu, Mar 4, 2010 at 1:12 PM, Bill Roberts <bill@swirrl.com> wrote:

> Hi Aldo - I'd like to help, but I see you posted your mail a few hours ago.
>  Do you have updated information on what still needs done?  Do you have a
> wiki or similar to coordinate volunteer programming efforts?
>
> Regards
>
> Bill
>
>
> On 4 Mar 2010, at 14:06, Aldo Bucchi wrote:
>
> > Hi,
> >
> > As most of you heard things were a bit shaky down here in Chile. We
> > have some requests and hope you guys can help. This is a moment to
> > prove what we always boast about: that Linked Data can solve real
> > problems.
> >
> > Google provides a prople finder service
> > (http://chilepersonfinder.appspot.com/) which is right now
> > centralizing some ( but not all ) of the missing people data. This
> > service is OK but it lacks some features plus we need to integrate
> > with other sources to perform analysis and aid our rescue teams /
> > alleviate families.
> >
> > This is serious matter but it is indeed taken a bti lightly by
> > existing software. ( there is a tradeoff between the amount of
> > structure you can impose and ease of use in the front-line ).
> >
> > What we would love to have is a way to access all feeds from
> > <http://chilepersonfinder.appspot.com/> as RDF
> >
> > We already have some databases operating on these feeds, but we're
> > still far away a clean solution because of its loose structure ( take
> > a look and you'll see what I mean ).
> >
> > Who wants to take a shot at this?
> >
> > Requirements.
> > - Take all feeds originating from <http://chilepersonfinder.appspot.com/
> >
> > - Generate an initial RDF dump ( big TTL file )
> > - Generate Incremental RDF dumps every hour
> >
> > The transfromation should do its best guess at the ideal data
> > structure and try not to loose granularity but shield us a bit from
> > this feed based model.
> >
> > We then take care of downloading this, integrating with other systems,
> > further processing, geocoding, etc.
> >
> > There's a lot of work to do and the more we can outsource, the bettter.
> >
> > On Friday ( tomorrow ) there will be the first nation-wide
> > announcement of our search platform and we expect lots of people to
> > use our services. So this is something really urgent and really,
> > really important for those who need it.
> >
> > Ah. Volunteers are moving all this data into a Virtuoso instance that
> > will also have more stuff. It will be available soon at
> > http://opendata.cl/ so stay tuned.
> >
> > We really hope we had something like DBpedia in place by now, it would
> > make all this much easier. But now is the time.
> > Guys, the tsunami casualties could have been avoided it was all about
> > mis-information.
> > Same goes for relief efforts. They are not optimal and this is all
> > about data in the end.
> >
> > I know you know how valuable data is. But it is now that you can
> > really make your point! Triple by Triple.
> >
> > Thanks!
> > A
> >
> > --
> > Aldo Bucchi
> > skype:aldo.bucchi
> > http://www.univrz.com/
> > http://aldobucchi.com/
> >
> > PRIVILEGED AND CONFIDENTIAL INFORMATION
> > This message is only for the use of the individual or entity to which it
> is
> > addressed and may contain information that is privileged and
> confidential. If
> > you are not the intended recipient, please do not distribute or copy this
> > communication, by e-mail or otherwise. Instead, please notify us
> immediately by
> > return e-mail.
> >
>
>
>