Re: Result format size comparison

Steve, I've been running that case
data:  http://www.ecs.soton.ac.uk/~swh/result-size/source.rdf
query:

[]
q:select { (?uri ?name ?sa) };
q:where  { ?uri a acm:Research-Area.
           ?uri support:has-pretty-name ?name }.

[]
q:select { (?uri ?name ?sa) };
q:where  { ?uri a acm:Research-Area.
           ?uri support:has-pretty-name ?name.
           ?uri <http://www.aktors.org/ontology/portal#sub-area-of> ?sa }.

and got 1486 results (instead of your 1473) and they are in n3 form like
(<http://www.acm.org/class/1998/D.4.1-threads> "Threads" 
<http://www.acm.org/class/1998/D.4.1>) . 
and the total result size is 167KB.

-- 
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/




Steve Harris <S.W.Harris@ecs.soton.ac.uk>
Sent by: public-rdf-dawg-request@w3.org
29/09/2004 13:30

 
        To:     DAWG public list <public-rdf-dawg@w3.org>
        cc:     (bcc: Jos De_Roo/AMDUS/MOR/Agfa-NV/BE/BAYER)
        Subject:        Result format size comparison



I've produced versions of a single, realworld query result in the various
formats I've seen on this list:
http://www.ecs.soton.ac.uk/~swh/result-size/readme.html

Includes aproximate renderings of the different formats (creators can mail
me corrections if thye think they've been misrepresented), I've not
checked any of them for validity, so there might be bugs. I'm only really
interested in how efficient they are. They are all equally easy to
generate.

Also included is some rough calcualtion of the overhead of the different
formats, compared to a baseline UTF8 format.

- Steve

Received on Thursday, 30 September 2004 00:08:43 UTC