Re: rql implementation

I'm unsure of the context, so take my comments with a grain of salt...

> Dan was asking about whether any query language implements the stuff  
> I've
> suggested in these messages:
>
> UC&R 4.6 use case and discussion
> http://lists.w3.org/Archives/Public/public-rdf-dawg/2004JulSep/ 
> 0491.html
>
> getting more concrete about 4.6
> http://lists.w3.org/Archives/Public/public-rdf-dawg/2004JulSep/ 
> 0403.html

I've been speaking to our guys, and also Kendall briefly on this, so  
will add my 2c.

iTQL supports almost all of what Kendall is suggesting, with the  
(current) exception of depth [3].

Essentially I think that we need to differentiate between what I'll  
call "explicit" properties (these exist in the base data) and inferred  
properties (inferred from the base data using some set of rules). The  
"distance" or "depth" they are away from the base node defines whether  
they are "direct" (e.g. :a rdfs:subClassOf :b) or "indirect" (e.g. :c  
rdfs:subClassOf :b, :b rdfs:subClassOf :a ==> :c rdfs:subClassOf :a is  
indirect). I think this is similar to what Rob was saying during our  
last telecon.

So for example, in iTQL's trans [1] operation, we differentiate between  
explicit and inferred statements, and if you want both returned, you  
perform a union of the results:

select $subject <rdfs:subClassOf> $object
from <some_model>
where
   trans($subject <rdfs:subClassOf> $object) or
   $subject <rdfs:subClassOf> $object;

The next step would be to allow a depth to be set, so that we could  
stop the generation of inferenced statements at a certain point. iTQL  
does this to a degree (with for example [2]), but we also plan to add a  
depth at some point in the future [3].


[1] http://www.kowari.org/1535.htm
[2] http://www.kowari.org/1532.htm
[3]  
http://sourceforge.net/tracker/index.php? 
func=detail&aid=978074&group_id=89874&atid=591707

-- 
Tom Adams                  | Tucana Technologies, Inc.
Support Engineer           |   Office: +1 703 871 5312
tom@tucanatech.com         |     Cell: +1 571 594 0847
http://www.tucanatech.com  |      Fax: +1 877 290 6687
------------------------------------------------------

Received on Friday, 17 September 2004 14:15:09 UTC