On 11/29/13 9:00 AM, John F Sowa wrote:
>> I blame sparql.
> At least we can agree on something. (01)
Why is it SPARQL's fault? Is it due to the poor name choice i.e., Named
Graphs? (02)
SPARQL isn't about processing quad patterns at all. It just so happens
that a number of SPARQL compliant engines are able to process TriG [1]
and NQuads [2] documents. The FROM (where IRI denotes a dataset
location) and FROM NAMED (where IRI denotes a local processor partition)
clauses simply set scopes for SPARQL query processors. (03)
Terminology across RDF and SPARQL is utterly problematic, that I will
never defend, but putting that issue aside, I don't see how else SPARQL
can be at fault re. confusion about things like statement reification
and other forms of metadata that relate to provenance. (04)
To conclude, is it poor names or actual semantics at fault here? I
believe its poor naming, poor narratives, and challenges related to
fixing the aforementioned. (05)
SPARQL uses Named Graphs to denote groups of relationships (triples).
The problem is that said grouping can be comprised of many relations
(relationships grouped by predicate) which doesn't align well with
Tables in an SQL RDBMS (where the Table Name denotes a relation because
it denotes an "instanceOf" relation), so we can't leverage familiarity
with Tables in these SPARQL and RDF related narratives. Of course, we
could spend time mapping this out, which is something I am actually
working on right now as part of an effort to address the fact that Data
itself needs to be revisited en route to solving these kinds of issues. (06)
The great thing about this list is that conversations are gradually
unearthing the real problems that have arisen from poor terminology
choices re. SPARQL and RDF. (07)
TriG and NQuads serve a purpose, but said purpose is ultimately rife
with problems; due to the fact that the 4th slot doesn't really engender
coherence, with regards to RDF semantics. (08)
Links: (09)
[1] http://wifo5-03.informatik.uni-mannheim.de/bizer/trig/ -- TriG
[2] http://www.w3.org/TR/2013/NOTE-n-quads-20130409/ -- NQuads (010)
-- (011)
Regards, (012)
Kingsley Idehen
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter Profile: https://twitter.com/kidehen
Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen (013)
smime.p7s
Description: S/MIME Cryptographic Signature
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (01)
|