Re: The Great Public Linked Data Use Case Register for Non-Technical End User Applications

On 6/24/13 9:09 AM, Michael Brunnbauer wrote:
> Hello Kingsley,
>
> On Mon, Jun 24, 2013 at 08:31:11AM -0400, Kingsley Idehen wrote:
>>>> Centralization doesn't scale, that's Web 101.
>>> Deactivate lod.openlinksw.com - that's web 101.
>> And you have another example of a live instance comprised of 51 Billion+
>> triples atop which you can perform faceted-style navigation of entity
>> relationship graphs, that's available to any human, program (many
>> end-user, integrator, or developer tools), or crawler?
> No I do not have another example but lod.openlinksw.com is a form of
> centralization - one that is currently needed, I must add.

No it isn't. It is just one of (hopefully) many other services that 
provide you with a "data junction box" en route to discovering data all 
over the Web.
The URIs exposed by <http://lod.openlinksw.com> resolve to entity 
description documents located from wherever they were published.

Linked Data enables us to build a distributed DataDNS system.

>
> You were talking about serendipitous discovery without central repositories.

Yes I am. I've been talking about it for a long time [1].

> How should I guess that you meant search engines and aggregators - which are
> central repositories ?

They are all simply puzzle-pieces in a massive game of jigsaw puzzles.

>
> So how do I get the list Dominic tries to compile from lod.openlinksw.com or
> some other search engine ? Feel free to create a SPARQL query for anyone to
> reuse in the future.

I need to look at what he seeks. That said, even if I constructed a 
SPARQL protocol URL, the net effect is that if I share said URL anyone 
has a starting point from this to drill-down further.

Later this week (or next, latest) we will share a number of detailed 
case studies (live examples) showcasing how Linked Data is used to solve 
a various problems that aid discovery on conventional search engines 
while also laying foundation for the next tier of semantically enhanced 
search engines. We will cover the following areas:

1. product catalogs
2. product offers
3. integration with Google snippets
4. shopping systems (shops and shopcarts)
5. product features and benefits.

Everything will be based on 5-Star Linked Data. All URIs will resolve to 
descriptions of their Referents.

If we dog-food Linked Data we will actually get around many of the 
tiresome issues that have repeatedly tripped up many adopters over the 
years.
>
> Sorry for PGP-signing my last couple of mails. I guess this is not good
> practice on mailing lists.

I sign my emails using an X.509 certificate that includes a person URI 
in its SAN slot :-)


Kingsley
>
> Regards,
>
> Michael Brunnbauer
>


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen