W3C home > Mailing lists > Public > public-lod@w3.org > May 2008

Re: ESWC 2008 Linked Data Playground

From: Hugh Glaser <hg@ecs.soton.ac.uk>
Date: Fri, 30 May 2008 00:17:27 +0100
To: "public-lod@w3.org" <public-lod@w3.org>
Message-ID: <C464F817.24DA3%hg@ecs.soton.ac.uk>

Hmm.
In a linked data world, the location of the institution would be published,
and then another linked world would be used to map place to lat/long?

This could actually happen just within dbpedia.
So looking up the institution in dbpedia will often give the p:city, which
leads to geo:lat and geo:long.
So what is needed is to sameAs
http://data.semanticweb.org/organization/humboldt-universitaet-berlin
and
http://dbpedia.org/page/Humboldt_University_of_Berlin
etc
And then follow the Linked Data road.

There are other solutions, but this is probably the simplest and safest?
Best
Hugh

On 29/05/2008 21:52, "Chris Bizer" <chris@bizer.de> wrote:

>
>
> Hi David,
>
>> Now, with all that open linked data, how much work does it take to get
>> lat/lng of 2 dozen well-known organizations? Should be trivial, right?
>
> Yes, trivial and very useful indeed.
>
> If somebody would go though the effort and put them online somewhere, we
> would be more than happy to set RDF links from the ESWC 2008 to this data.
>
> Cheers
>
> Chris
>
>
> --
> Chris Bizer
> Freie Universitšt Berlin
> +49 30 838 54057
> chris@bizer.de
> www.bizer.de
> ----- Original Message -----
> From: "David Huynh" <dfhuynh@alum.mit.edu>
> To: <kidehen@openlinksw.com>
> Cc: <public-lod@w3.org>
> Sent: Thursday, May 29, 2008 4:35 PM
> Subject: Re: ESWC 2008 Linked Data Playground
>
>
>>
>> Kingsley Idehen wrote:
>>>
>>> David Huynh wrote:
>>>>
>>>> Hi Richard,
>>>>
>>>> If you look at this version
>>>>
>>>>
>>>> http://people.csail.mit.edu/dfhuynh/projects/eswc2008/eswc2008-rdf.html
>>>>
>>>> you'll see that the RDF/XML file is linked directly. So, there's pretty
>>>> much zero cost in getting RDF into Exhibit (it took me about 5 minutes
>>>> to put that exhibit up). Exhibit automatically routes RDF/XML files
>>>> through Babel for conversion. In the first version, I did that
>>>> conversion and saved the JSON output so that Babel won't get invoked
>>>> every time someone views the exhibit. That's an optimization. Of course,
>>>> Babel isn't perfect in doing the conversion.
>>>>
>>>> Here is an iPhone mockup version for the same exhibit:
>>>>
>>>> http://people.csail.mit.edu/dfhuynh/projects/eswc2008/iphone/iphone-exhibit
>>>> .html
>>>> I only tested it on Firefox and Safari. I think the back button
>>>> functionality doesn't quite work that well, but you get the idea.
>>>>
>>>> David
>>> David,
>>>
>>> Even if you don't use RDFa to express what <> is about i.e. it's
>>> foaf:primarytopic, foaf:topic etc..
>>>
>>> In the Exhibit pages <head/> You can accompany:
>>> <link
>>> href="http://data.semanticweb.org/dumps/conferences/eswc-2008-complete.rdf"
>>> type="application/rdf+xml" rel="exhibit/data" />
>>> with
>>> <link rel="alternate"
>>> href="http://data.semanticweb.org/dumps/conferences/eswc-2008-complete.rdf"
>>> type="application/rdf+xml" />
>>>
>>> I think we need to adopt a multi pronged approach to exposing Linked Data
>>> (the raw data behind the Web Page):
>>>
>>> 1. Content Negotiation (where feasible)
>>> 2. <link rel=.../>  (for RDF sniffers/crawlers)
>>> 3. RDFa
>>>
>>>
>>> Re. point 2, I've just taken a random person "Abhita Chugh
>>> <http://demo.openlinksw.com/rdfbrowser2/?uri=http%3A%2F%2Fsemanticweb.org%2F
>>> wiki%2FAbhita_Chugh>"
>>> from <http://data.semanticweb.org> which exposes the RDF based
>>> Description of "Abhita Chugh"
>>> <http://demo.openlinksw.com/rdfbrowser2/?uri=http%3A%2F%2Fsemanticweb.org%2F
>>> wiki%2FAbhita_Chugh>
>>> via our RDF Browser without problems (we use all 3 of the methods above
>>> to seek "Linked Data" associated with a Web Document). In this case it
>>> also eliminates the need to translate anything (e.g. routing via Babel)
>>> since the original data source is actually RDF.
>>>
>>> Of course, I could take the exhibit page and slap this in myself, but I
>>> am hoping you could tweak Exhibit such that it does point 2 and maybe
>>> point 3 automatically. That would be a major boost re. Exhibit's Linked
>>> Data credentials :-)
>> Exhibit can't do #2 because it only acts on the page at runtime, so the
>> author of an exhibit must put that in herself. And that I just did for the
>> ESWC 2008 exhibits.
>>
>> BTW, Semtech 2008 has a cool Exhibit-backed event browser:
>>    http://www.semantic-conference.com/scheduler/
>> Maybe future *SWC conferences would have use for the same service.
>>
>> So, it'd be good to get lat/lng coordinates for the affiliations and then
>> plot the speakers on a map, like what I did for ISWC 2007 (just for
>> kicks):
>>
>> http://people.csail.mit.edu/dfhuynh/projects/graph-based-exhibit/graph-based-
>> exhibit2.html
>> Now, with all that open linked data, how much work does it take to get
>> lat/lng of 2 dozen well-known organizations? Should be trivial, right?
>>
>> David
>>
>>
>
>
>
Received on Thursday, 29 May 2008 23:18:16 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 31 March 2013 14:24:16 UTC