ACTION-33: Investigate the use of PURL urls for the SSN ontology
Investigate the use of PURL urls for the SSN ontology
- State:
- closed
- Person:
- Laurent Lefort
- Due on:
- August 5, 2010
- Created on:
- July 15, 2010
- Associated Product:
- W3C Semantic Sensor Network Ontology
- Related emails:
- Re: Question about the deployments, systems and devices model (from Michael.Compton@csiro.au on 2010-07-22)
- Question about the deployments, systems and devices model (from a.nikolov@open.ac.uk on 2010-07-20)
- SSN-XG Meeting Reminder (20 July) (from Laurent.Lefort@csiro.au on 2010-07-19)
Related notes:
Investigate if and how using PURL urls could simplify the transition from the phase where the ontology is developed as part of the XF work to the phase where it is maintained after the end of the XG.
Advantage: it would be easier to move the ontology to a different physical location in the future
Disadvantage: the fact that the ontology is the result of the SSN Incubator Activity will be less apparent
Example of ontology using purl: GoodRelations http://purl.org/goodrelations/
A permanent URL (303) has been defined for the SSN ontology: http://purl.org/NET/ssnx/ssn
It is redirecting to the documentation page http://www.w3.org/2005/Incubator/ssn/wiki/SSN
This purl has been used used for all the versions of the SSN ontology delivered after mid-August 2010 (uploaded on the wiki)
See http://lists.w3.org/Archives/Public/public-xg-ssn/2010Aug/0009.html
Have tried the 302 and 303 solutions. Can have both working at the same time except when it is possible to tweak the Apache config file directly, so currently have the 302 one so that tools like Protege can load everything seamlessly.
Laurent Lefort, 26 Oct 2010, 14:25:02Complement of information:
- the SSN ontology PURL is http://purl.oclc.org/NET/ssnx/ssn and not http://purl.org/NET/ssnx/ssn (and all the PURLs for the other SSN XG starting with the /NET/ssnx/ssn prefix are also using http://purl.oclc.org/NET/ssnx )
- I have not found any explanation of which option is better.
There has been a discussion on the topic of the choice of the cool URI method between Ian Davis (Talis) and David Wood (now also Talis formerly Zeiphera (PURLZ and Callimachus))
David Wood: http://prototypo.blogspot.com/
Ian Davis http://iandavis.com/
Discussion (reverse chronological order)
- http://prototypo.blogspot.com/2011/01/ian-was-right-and-i-was-wrong.html (Jan 6)
- http://blog.iandavis.com/2010/12/06/back-to-basics/ (Dec 6)
- http://prototypo.blogspot.com/2010/11/another-guide-to-publishing-linked-data.html (Nov 9 = before David admitted he was wrong)
- http://iand.posterous.com/a-guide-to-publishing-linked-data-without-red (Nov 7)
- http://iand.posterous.com/is-303-really-necessary (Nov 3)
Summary: don't tweak URIs (don't use 303), it is better to separate URIs to access RDF content from URI defining a resource.
Display change log.