Re: sameAs.org & The LOD Cloud Diagram - advice please

On 7/25/14 3:02 PM, Hugh Glaser wrote:
> So sameAs.org never appears in any of this stuff.
> That’s deliberate.
> The whole idea of it is that it doesn’t add to the plethora of URIs by generating new ones.
> But it seems that people do find it useful - I get emails from people about it, especially when it does strange things :-)
>
> sameAs.org is a service that consumes LD URIs from other PLDs and delivers the answer as LD.
> So it is, in fact, a proper Linked Data site.
> So you can do the stuff you expect with a URI like
> http://www.sameas.org/?uri=http%3A%2F%2Fdbpedia.org%2Fresource%2FEdinburgh
>
> As you can see, this is not a Cool URI (I think), and as I said, I really don’t want people to think of this as the ID for a sameAs bundle, although in fact it is!
>
> So what should I do?
> Keep sameAs.org living outside the BTC and LOD Cloud world?
> Or change things so that it becomes a more normal part of the LOD world?

Yes.

>
> In addition, I have quite a lot of other services that work over LD URIs to produce LD about the URI, but as with sameAs.org are also LD URIs because the service interface itself supports the LD model.
> Should these brought into the PR fold, and if so how?

Yes.

> It would be very painful to allow these to be spidered, because they are heavy computations, and running the service over all the URIs that could be run would be many years of computation for my server.

This is where you can put WebID based ACLs to good use. Basically, you 
control how agents work with your data space.

>
> I suspect that others have similar services, so a policy might be useful.

We had to do that is with URIBurner (I note, that's vanished from the 
LOD cloud too!).

>
> In fact, as LD becomes more mature (!), I think we are finding that it is more a communication mechanism between cooperating service than simply delivering RDF in response to an identifying URI - how do we capture this massive LD resource?

By doing better on the communications fronts (human to human), (human to 
machine), and (machine to machine).
>
> I hope that makes some sort of sense.

It does .

>
> Best
> Hugh


-- 
Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog 1: http://kidehen.blogspot.com
Personal Weblog 2: http://www.openlinksw.com/blog/~kidehen
Twitter Profile: https://twitter.com/kidehen
Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen
Personal WebID: http://kingsley.idehen.net/dataspace/person/kidehen#this