W3C home > Mailing lists > Public > semantic-web@w3.org > February 2011

Re: {Disarmed} Re: URI in a URI

From: Hugh Glaser <hg@ecs.soton.ac.uk>
Date: Fri, 4 Feb 2011 15:39:59 +0000
To: Andrea Splendiani <andrea.splendiani@bbsrc.ac.uk>
CC: Vincent Huang A <vincent.a.huang@ericsson.com>, "semantic-web@w3.org Web" <semantic-web@w3.org>
Message-ID: <EMEW3|68a43c77fb3166315efeb47be428b910n13Fe902hg|ecs.soton.ac.uk|0CDD84E6-BC81-4180-9977-024EB4AF4481@ecs.soton.ac.uk>
Ah, yes, I had not thought about exactly the meaning of the URI, more the structure.
Sorry.
However, yours and William's messages rightly prompt the view that the property should be the service name and the uri= should identify the location (and I missed a preferred / before the ? in the URI).
More like:
http://example.com/sensors/temperature/?uri=http://sweet.jpl.nasa.gov/2.1/sensor1
if that works for you.

I think one of the things about all this is to try to have a commonish vocabulary for the arguments for the service.
So you could use /?sensor= or /?uri=
I would suggest that where a service only really needs one URI, then making the user remember or find out what the name is makes it a modicum harder, so simply using /?uri= is helpful.

Best

On 4 Feb 2011, at 14:41, Andrea Splendiani wrote:

HI,

why not:
http://example.com/sensors/sensor1/lookup?property=http://sweet.jpl.nasa.gov/2.1/propTemperature<http://x.jmxded133.net/y.z?l=http%3a%2f%2fexample.com%2fsensors%2fsensor1%2flookup%3fproperty%3dhttp%3a%2f%2fsweet.jpl.nasa.gov%2f2.1%2fpropTemperature&r=207926954&d=31602&p=1&t=h>

A minor thing, but may help making things clearer (one could use this approach as standard representation for properties values via URL, where needed).

ciao,
Andrea

<http://x.jmxded133.net/y.z?l=http%3a%2f%2fexample.com%2fsensors%2fsensor1%2flookup%3furi%3dhttp%3a%2f%2fsweet.jpl.nasa.gov%2f2.1%2fpropTemperature&r=207926954&d=31602&p=1&t=h>
Il giorno 04/feb/2011, alle ore 14.08, Hugh Glaser ha scritto:

A very useful question.
It is an important issue for the emerging services accessing the semantic web.

Yes in general using a # is a bad thing.
It is likely the server will never see the fragment after the #.

I don't know if it is written anywhere, but there seems to me a bit of a consensus around this.
And it is folded into the RESTful stuff.
So for example
 http://kmi-web05.open.ac.uk/REST_API.html<http://x.jmxded133.net/y.z?l=http%3a%2f%2fkmi-web05.open.ac.uk%2fREST_API.html&r=207926954&d=31602&p=1&t=h>
describes a typical service invocation with a URI as argument as:
http://watson.kmi.open.ac.uk/API/semanticcontent/metadata/?uri=[docURI]

We do something similar in http://sameas.org/
http://sameas.org/about.php describes in detail, with things like:
http://sameas.org/?uri=http://dbpedia.org/resource/London (the NIR)
and
http://sameas.org/rdf?uri=http://dbpedia.org/resource/London (an IR)
etc.
and also in the rkbexplorer services things like
http://www.rkbexplorer.com/network/?uri=http://southampton.rkbexplorer.com/id/person-00021&type=person-person&format=tsv
as well as with two URIs
http://www.rkbexplorer.com/connections/?source=http://southampton.rkbexplorer.com/id/person-da9c463f8b783083d7d7e9003db8224f-57e2ec2d7aee429c73fef344805033e2&target=http://southampton.rkbexplorer.com/id/person-17e6d4cf4846bd195454a7c1143a20fb-32a6807d38b58d6d56e31d88f5e48de2&type=person-person

So you could use
http://example.com/sensors/sensor1/lookup?uri=http://sweet.jpl.nasa.gov/2.1/propTemperature

Unless someone wants to tell us that is crazy?
This is the sort of thing it is useful to have some best practice emerge on.
Or can anyone point us at where it is written?

Best
Hugh

On 4 Feb 2011, at 12:41, Nathan wrote:

Vincent Huang A wrote:
Can I have a URI in another URI definition?
For example, if I want to give a name to a temperature service coming from a sensor, can I say:
http://example.com/sensors/sensor1/services#http://sweet.jpl.nasa.gov/2.1/propTemperature

well now, that's just one, valid, URI :) so long as it conforms to the syntax restrictions then you're fine.

note: be clear on the fact that it's only one URI though, not two URIs, or a "URI in a URI", and that #fragments have certain restrictions (see [1]), as in you can't have a '#' in a fragment.

[1] http://tools.ietf.org/html/rfc3986#section-3.5

Best,

Nathan


--
Hugh Glaser,
             Intelligence, Agents, Multimedia
             School of Electronics and Computer Science,
             University of Southampton,
             Southampton SO17 1BJ
Work: +44 23 8059 3670, Fax: +44 23 8059 3045
Mobile: +44 78 9422 3822, Home: +44 23 8061 5652
http://www.ecs.soton.ac.uk/~hg/




Andrea Splendiani
Senior Bioinformatics Scientist
Centre for Mathematical and Computational Biology
+44(0)1582 763133 ext 2004
andrea.splendiani@bbsrc.ac.uk<mailto:andrea.splendiani@bbsrc.ac.uk>



[http://x.jmxded133.net/o.z?r=207926954&d=31602]

--
Hugh Glaser,
              Intelligence, Agents, Multimedia
              School of Electronics and Computer Science,
              University of Southampton,
              Southampton SO17 1BJ
Work: +44 23 8059 3670, Fax: +44 23 8059 3045
Mobile: +44 78 9422 3822, Home: +44 23 8061 5652
http://www.ecs.soton.ac.uk/~hg/
Received on Friday, 4 February 2011 15:40:43 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:41 GMT