Hi Heiko,

On 29 Jan 2015, at 09:11, Heiko Paulheim <he...@informatik.uni-mannheim.de> 
wrote:

> thanks for that description, as it shows how to compile such statistics 
> without crashing the public endpoint ;-)

I'm happy that you like it ;)
Still it would be cooler if such simple usage counts were dragged along and 
used for such queries, but i know that SPARQL endpoint optimisation isn't that 
easy.
Maybe providing them as triples themselves would be an alternative, but it has 
obvious problems with consistency :( (do stat triples count as well? ;) )


> Just a remark on your duplicates: they aren't really duplicates. The pattern 
> that occurs are that there are those three lines:
> A dcterms:rights X .
> A foaf:thumbnail B .
> B dcterms: rights X .
> Imho, this makes sense: There is an image A, which has a thumbnail B, and 
> both have a rights statement X, which is the same for both the image and the 
> thumbnail.
> 
> What actually is a bit odd though is that the image and its rights statement 
> are identical.

I think i didn't explain that properly:

>>    4891 
>> <http://commons.wikimedia.org/wiki/Special:FilePath/Flag_of_Slovenia.svg?width=300>
>>  <http://purl.org/dc/elements/1.1/rights> 
>> <http://en.wikipedia.org/wiki/File:Flag_of_Slovenia.svg> .

That line means that the triple occurs 4891 times in the dataset.

So it's less about the pattern, more about certain triples just occurring very 
often.


Cheers,
Jörn


------------------------------------------------------------------------------
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Dbpedia-discussion mailing list
Dbpedia-discussion@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion

Reply via email to