W3C home > Mailing lists > Public > public-lod@w3.org > May 2009

Adding Country Specific Data

From: <richard.hancock@3kbo.com>
Date: Mon, 25 May 2009 15:48:32 -0500 (CDT)
Message-ID: <48751.203.97.223.64.1243284512.squirrel@www.3kbo.com>
To: "public-lod@w3.org" <public-lod@w3.org>
Cc: "Kingsley Idehen" <kidehen@openlinksw.com>, "Hugh Glaser" <hg@ecs.soton.ac.uk>
Thanks Kingsley, Hugh and Toby,

for answering my original questions re how to query for country specific
data.

As much as I like Dbpedia I wasn't sure that the URIs that I had found
were the right starting point for focusing on the towns of New Zealand.
Using Kingsley's site http://lod.openlinksw.com/fct/ via its URI Lookup
(by Label) I got to
http://www4.wiwiss.fu-berlin.de/factbook/resource/New_Zealand but that
still doesn't have an easy way to link to towns etc...

As Toby pointed geonames has better linking for towns but doesn't have
Sparql interface.

The rkbexplorer sameAs service also worked well
http://www.rkbexplorer.com/sameAs/?uri=http://dbpedia.org/resource/Wellington
showing that http://dbpedia.org/resource/Wellington was the same
http://www.geonames.org/2179538/

I still have to explore the UK Ordnance Survey and Hugh's cross-references
at http://os.rkbexplorer.com/crs/ in more detail but can see that
http://dbpedia.org/resource/London gets me to
http://os.rkbexplorer.com/id/osr7000000000041428

This leads me back to Hugh's suggestion of reusing the OS ontology for New
Zealand data.

Part of the motivation behind my original query was to get a feel for what
linked data has already been published for New Zealand, what the current
gaps are, and how best to fill them.

Still got a ways to go before making concrete plans re adding data. E.g.
trying to determine what would be good data to add, what data sources are
readily available, and what viable business cases can be made to finance
publication.

For the OS ontology I found the Ordnance Survey Ontologies page at 
http://www.ordnancesurvey.co.uk/oswebsite/ontology/ and can see that
http://os.rkbexplorer.com/id/osr7000000000041428 uses the namespaces
http://www.rkbexplorer.com/ontologies/coref# and
http://www.ordnancesurvey.co.uk/ontology/AdministrativeGeography/v2.0/AdministrativeGeography.rdf#

What would be a good starting point for understanding the Ordnance Survey
Ontologies and the cross-referencing,

Cheers,

Richard

blog: http://blog.3kbo.com


> Thanks Kingsley,
> I think that's enough.
> The only reason I said anything was because you asked me to comment - I
> did.
> If in answering I misrepresented your offering, then I apologise -
> although
> I happen to think that I understand it quite well.
> We clearly need to agree to differ on a number of things.
> Best
> Hugh
>
> On 25/05/2009 13:02, "Kingsley Idehen" <kidehen@openlinksw.com> wrote:
>
>> Hugh Glaser wrote:
>>> Thanks Kingsley.
>>> I'm not sure why you have raised all this again.
>>> I simply suggested to Richard another way of doing what he wanted.
>>>
>> I don't have an issue with you point Richard to alternatives.
>> I do have issues with our offering being misrepresented (albeit
>> unintentionally).
>>> You then asked me whether what you had proposed failed to resolve his
>>> problem.
>>> I can't say whether it does, but perhaps Richard can better answer
>>> that.
>>>
>> Yes.
>>> But it would have been rude of me not to attempt to answer your direct
>>> question to me.
>>> My view is that probably none of this now addresses Richard's
>>> fundamental
>>> problem, I think, (which I was trying to do in my message and which
>>> Toby is
>>> also trying to address). He needs reliable properties that relate
>>> countries
>>> to their geography. It is a problem of ontology and published data, not
>>> how
>>> to access it.
>>>
>> I am very aware of this, I am not an ABox only Linked Data type :-)
>>> Išll trim things a bit to try to get at some essence.
>>>
>>> On 25/05/2009 03:43, "Kingsley Idehen" <kidehen@openlinksw.com> wrote:
>>>
>>>> I am not assuming once source. Of course not. I am assuming a possible
>>>> beachhead :-)
>>>>
>>> And a very nice beachhead.
>>> But your solution only talked about the source at
>>> http://lod.openlinksw.com
>>> It is also interesting to consider how it might interact with other
>>> sources.
>>>
>> If you look closely, we don't take the original URIs out of scope, you
>> always have a route to wherever on the broader Linked Data Web.
>>
>> Nice metaphor: Spaghetti Junction out of B'ham :-)
>>>> The whole point of Linked Data should be to demonstrate how it
>>>> embraces
>>>> and extends the Google full text search realm which is autistic to
>>>> entities, entity types, and entity properties re. disambiguation of
>>>> queries (or as they call them: searches).
>>>>
>>> Ah. I think this is perhaps getting to the nub.
>>> I don't see Linked Data as relating to search - more to lookup, as in a
>>> database record lookup by key.
>>>
>> Linked Data is inextricably linked to search re. the Web, because URIs
>> are inextricably linked to entity identifiers and negotiated
>> representations (documents) that carry their descriptions.
>>> The semantic web is more like one big database then a big file system.
>>>
>> I speak in terms of data spaces, and I see the Web as a federation of
>> Linked Data Spaces.
>>
>> I don't see a Web and a Semantic Web. That thinking and reality died a
>> long time ago (imho).
>>
>> There is just a World Wide Web that have evolved to the point where
>> linkage now occurs at the data -- rather than document-- level.
>>> So you project into the Linked Data world by finding the URI you want,
>>> and
>>> from then on in it is URIs all the way down, until/unless you want to
>>> show a
>>> human something, when you project back into their language.
>>>
>>>>> That is exactly what Richard was trying to do; having found a URI
>>>>> such as
>>>>> http://dbpedia.org/resource/New_Zealand that he is satisfied captures
>>>>> the
>>>>> concept with which he is concerned, he now wants to explore what is
>>>>> known
>>>>> about it in the Linked Data world, without going back to the text
>>>>> world.
>>>>>
>>>>>
>>>> Again, I don't think I am sending him back to the full text pattern
>>>> world.
>>>>
>>>> I am saying:
>>>>
>>>> 1. Enter a patter: New Zealand (as you would re. Google, Yahoo! etc..)
>>>> 2. When presented with hits (which are really Entities with URIs  plus
>>>> excerpts from associated literal object values) filter further by
>>>> Entity
>>>> Type or Entity Property
>>>>
>>> OK, that's how to start.
>>> But he doesn't need to do that - he said he already had the URI he
>>> wanted:
>>> http://dbpedia.org/resource/New_Zealand
>>>
>> Even easier then for him, he just goes to the tab labeled: URI Lookup .
>>
>> And then he can see link constellation associated with this entity.
>>
>>>>
>>>>> And I donšt think he wanted to do any clicking ­ he wanted to just
>>>>> script
>>>>> it
>>>>> all up in a reliable Linked Data sort of way.
>>>>>
>>>>>
>>>> Lets assume he didn't want to click anything, what do you think the
>>>> purpose of the "URI Lookup by Label" and "URI Lookup" tabs are for
>>>> then?
>>>>
>>> For me to put in a URI such as http://dbpedia.org/resource/New_Zealand
>>> and
>>> get the Linked Data back.
>>>
>>>> They are for entering patterns that are associated with Entity Labels
>>>> or
>>>> actual URIs.
>>>>
>>> Have you tried typing in the URIs that Richard specified?
>>> However, looking at it, I think it may just be a bugette, which
>>> confused me.
>>>
>>>> The instance at: <http://lod.openlinksw.com> is but one data space on
>>>> the vast Web of Linked Data. It's a linked data junction box with lots
>>>> of de-referencable URIs that can take you to many places on the Web or
>>>> conduct data via many pathways on the Web.
>>>>
>>> Actually, it is a Linked Data site that has uploaded a lot of data from
>>> other places, and also dynamically gets more. I assume by instance you
>>> mean
>>> it is an instance of the class of Linked Data sites.
>>>
>>>> I don't understand why you find my responses fundamentally incongruent
>>>> with the very essence of Linked Data. We keep on going round the same
>>>> loop in different ways.
>>>>
>>> An interesting question.
>>> I am certainly uncomfortable with responses that never seem to mention
>>> the
>>> idea that Linked Data is a Web of Data, by suggesting the use of data
>>> that
>>> might be accessed on domains other than http://lod.openlinksw.com .
>>>
>> http://lod.openlinksw.com for all intents an purposes in a Linked Data
>> Web lookup service. All URIs are intact meaning, you can dereference the
>> URIs against their sources. Please take a closer look at the @href
>> values in our Web pages.  We are not centralist we are as open as you
>> can get and we tackle real problems based on a wealth of deep experience
>> from both the DBMS (different types) and Middleware realms.
>>
>> Linked Data is a more open vehicle for our inherent passion and
>> expertise. What we showcase is about real issues and practical
>> solutions. The Web isn't about a single company or a single service
>> (I've made this crystal clear numerous times in my comments), it should
>> be a about collections of solutions that adhere to core principles.
>>
>> My general discomfort is that you are not really grasping the essence of
>> our intentions etc..
>>
>>
>> Kingsley
>>> But this would be for another thread, and I don't have the time to do
>>> that.
>>>
>>> Best
>>> Hugh
>>>
>>>
>>>
>>
>>
>> --
>>
>>
>> Regards,
>>
>> Kingsley Idehen       Weblog: http://www.openlinksw.com/blog/~kidehen
>> President & CEO
>> OpenLink Software     Web: http://www.openlinksw.com
>>
>>
>>
>>
>>
>
>
>
Received on Monday, 25 May 2009 20:49:17 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 31 March 2013 14:24:21 UTC