Re: Linked Data available

Hugh,

On 7 Nov 2007, at 23:08, Hugh Glaser wrote:

(snip -- most issues are fixed now)

> We essentially have these datasets so that we can provide a way of  
> providing
> an ontology-mediated unified view of multiple, distributed, semantic  
> web
> repositories (a true semantic *web* application).
> So we are putting our resources into this RKB Explorer.
> The prototype was at http://resist.ecs.soton.ac.uk/explorer/

Yes, the prototype makes it very easy to make sense of the available  
data. Good job on hiding the RDF, I didn't see a single URI leak  
through into the user interface. (In fact, you were a bit *too*  
thorough in that regard -- a link to an RDF version of the currently  
selected item would be great.)

> Having built that, however, the re-engineering so that we were not  
> caching
> the RDF locally anymore, and adhering to linked data standards, took  
> much
> longer than we would have liked.Since the datasets have been pretty  
> much
> ready for a while, it seemed a shame not to announce them (things  
> are never
> really ready).

I agree. Hope you didn't mind getting early feedback ;-)

> I think it is hard to say what would be a useful "root" for things  
> like NSF
> or DBLP.

Lists of conferences, journals, years of publications, ... There are  
possibilities.

> So our story for the navigation is twofold:
> 1) The data is public, and so it should be possible to use any  
> reasonable
> tool to explore (given the ontology);

Well, this line of reasoning is a bit one-sided. Yes, the data is  
public. Yes, this means I can use any reasonable tool. But on the  
other hand, it is a good idea to keep in mind the actual capabilities  
of currently existing tools. That would be Tabulator, OpenLink  
browser, Sindice, and so on. It is only through these tools that users  
can currently interact with the Linked Data interfaces. (And yes, we  
all know that these tools have to get a lot better to be truly useful.)

> 2) We provide an application of our own (the RKB Explorer,
> http://www.rkbexplorer.com/ ) which is a human interface.

Yes, although you certainly know that this is not a very good human  
interface. It allows RDF geeks like me to poke around in the datasets,  
but I don't think it is appropriate (or intended) for end users.

> I think this issue relates very neatly to the useability discussion:  
> I see
> the Semantic Web as essentially a universe of machines/agents  
> communicating,
> with humans being blissfully unaware that the interfaces they are  
> using are
> part of the Semantic Web.
> This is our intention for RKB Explorer.

Of course it must be a goal to make the user experience as effective  
as possible. But I wonder why users should be unaware that they are  
using the Semantic Web? The traditional Web is also a universe of  
communicating machines and agents (HTTP servers and Web browsers), and  
still the surfer is aware that he is using the WWW, and there's  
nothing wrong with that.

Best,
Richard


>
>>
>> 6. You should consider adding a few domain-specific search functions,
>> such as the simple “Find Yourself” function provided at
>> http://dblp.l3s.de/d2r/
> I suspect that the Search facility in the RKB Explorer will be enough.
>>  .
>>
>> I'm a bit frustrated because this looks like an amazingly great
>> resource, but I can't actually get any clear feeling for its scope or
>> quality or contents. This feels like exploring a pitch black room
>> while wearing boxing gloves.
> I do take your point - sorry.
>>
>> I'm very hopeful that you can greatly improve this experience with
>> little effort.
>>
>> Thanks a lot,
>> Richard
>>
>>
>> On 6 Nov 2007, at 16:00, Hugh Glaser wrote:
>>
>>>
>>> Hi,
>>>
>>> We have constructed some rdf resources and an application that uses
>>> them,
>>> that can be found at:
>>> http://www.rkbexplorer.com/
>>>
>>> Each resource has a different domain, to make it easy to use, and
>>> emphasising the fact that these are in fact completely separate
>>> knowledge
>>> bases.
>>>
>>> For each resource, you can access it in the following ways:
>>>
>>> Browse it, looking at the RDF triples;
>>> Query it, using SPARQL;
>>> Resolve each of the unique URIs for resources: depending on content
>>> negotiation, RDF or HTML will be returned;
>>> Query the associated CRS (Consistent Reference Service), to find out
>>> which
>>> other URIs are considered to be the same non-information resource.
>>>
>>> We currently have the following:
>>>
>>> budapest.rkbexplorer.com
>>> citeseer.rkbexplorer.com
>>> cordis.rkbexplorer.com
>>> darmstadt.rkbexplorer.com
>>> dblp.rkbexplorer.com
>>> deepblue.rkbexplorer.com
>>> eurecom.rkbexplorer.com
>>> ft.rkbexplorer.com
>>> ibm.rkbexplorer.com
>>> ieee.rkbexplorer.com
>>> irit.rkbexplorer.com
>>> italy.rkbexplorer.com
>>> kaunas.rkbexplorer.com
>>> laas.rkbexplorer.com
>>> lisbon.rkbexplorer.com
>>> newcastle.rkbexplorer.com
>>> nsf.rkbexplorer.com
>>> pisa.rkbexplorer.com
>>> roma.rkbexplorer.com
>>> southampton.rkbexplorer.com
>>> ulm.rkbexplorer.com
>>> wiki.rkbexplorer.com
>>>
>>> There are some quite big resources here: citeseer, cordis, dblp and
>>> nsf are
>>> a good size.
>>>
>>> They all use the same ontology (AKT Reference Ontology, with some
>>> small
>>> additions), which hopefully makes them a useful resource for
>>> applications
>>> that use ontology-mediated separate knowledge bases.
>>>
>>> There is also such an application:
>>> http://www.rkbexplorer.com/explorer/
>>> Since we have rebuilt the knowledge bases to conform to linked data
>>> standards, the search and CRS information is still being analysed,
>>> and thus
>>> this version of the RKB explorer does not yet present as unified a
>>> view as
>>> we would like.
>>> For a better unified view, please see the old version at
>>> http://resist.ecs.soton.ac.uk/explorer/
>>>
>>> Best
>>> Hugh
>>>
>>> -- 
>>> Hugh Glaser,  Reader
>>>              Dependable Systems & Software Engineering
>>>              School of Electronics and Computer Science,
>>>              University of Southampton,
>>>              Southampton SO17 1BJ
>>> Work: +44 (0)23 8059 3670, Fax: +44 (0)23 8059 3045
>>> Mobile: +44 (0)78 9422 3822, Home: +44 (0)23 8061 5652
>>> http://www.ecs.soton.ac.uk/~hg/
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>
>

Received on Thursday, 8 November 2007 13:05:09 UTC