Seaborne, Andy wrote:
>   
>> -----Original Message-----
>> From: Peter Ansell [mailto:[EMAIL PROTECTED]
>> Sent: 02 November 2008 20:32
>> To: Seaborne, Andy
>> Cc: [email protected]
>> Subject: Re: [Dbpedia-discussion] bif:contains text search treated as a
>> prefix
>>
>>
>> ----- "Andy Seaborne" <[EMAIL PROTECTED]> wrote:
>>
>>     
>>> From: "Andy Seaborne" <[EMAIL PROTECTED]>
>>> To: [email protected]
>>> Sent: Monday, 3 November, 2008 5:03:17 AM GMT +10:00 Brisbane
>>> Subject: Re: [Dbpedia-discussion] bif:contains text search treated as
>>>       
>> a prefix
>>     
>
>
>   
>>> You have used bif: here.  It needs a matching PREFIX .  There are no
>>> default prefixes in SPARQL.
>>>       
>> As far as I know, OpenLink have not provided a prefix definition for
>> bif: (ie, Built-in Functions from SQL).
>>     
>
> Peter - thanks for the information.
>
> OpenLink - any chance of providing a non-proprietary approach within the 
> SPARQL syntax?
>   
Yes, it should be resolved with an OpenLink domain based qname for this 
magic predicate i.e.,
<http://www.openlinksw.com/schema/sparql/extensions#contains> should do, 
such that you simply take the following approach:

prefix bif: <http://www.openlinksw.com/schema/sparql/extensions#>

Kingsley
> It's not very semantic web in three ways.  Not using a URI for something on 
> the web (that function or index); having non-portable syntax (bif: being 
> special and not needing a prefix), and not reusing existing names 
> (fn:contains from XPath/XQuery functions and operators seens to do what this 
> does).
>
> The function/index should have a URI.  That's part of the point, giving 
> global names to things.  It's written as a property in a triple pattern so 
> really it should be a URI, after all, that's the point.  As a triple 
> (pattern) is a relationship between subject and object, the fact it's 
> calculated rather than looked up in some graph is irrelevant but the meaning 
> of the relationship is something to fix with a web-wide name.
>
> The "bif:" special means that an application must reworked to apply the same 
> query to a different source, as might happen walking linked data.
>
> XPath/XQuery has fn:contains which is a function so
>
>     FILTER(fn:contains(?x, "xyz"))
>
> See : http://www.w3.org/TR/xpath-functions/#func-contains
>
> (Or, maybe, the triple pattern equivalent.)
>
> Now at the moment there is no standard for what a processor provides, but at 
> least reusing the F&O function would be good.  Hopefully, the next round of 
> round of SPARQL specification will include a standard set of functions that 
> all implementation provide so helping portability of applications across end 
> points.  If a client application is going to work over multiple endpoints, 
> it's easier if they offer the same core functionality.  At least enable an 
> endpoint to make an RDF description available which says what it offers.
>
> (aside:
> The query processor can easily spot a FILTER like that and use an index so 
> the triple pattern/function difference is merely whether the application 
> writer has to do the work.)
>
> Marvin -
>
> The direct call to QueryEnegineHTTP in ARQ will bypass the client-side syntax 
> checking to allow the non-portable query to be sent.
>
>     Andy
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Dbpedia-discussion mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
>
>   


-- 


Regards,

Kingsley Idehen       Weblog: http://www.openlinksw.com/blog/~kidehen
President & CEO 
OpenLink Software     Web: http://www.openlinksw.com





-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Dbpedia-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion

Reply via email to