headers
Miles, AJ (Alistair | 1 Nov 2005 18:12
Picon
Picon

schemarama 2


Hi all,

I've written up an idea for using SPARQL to generate reports based on the content of an RDF graph, a pragmatic
way of testing whether an RDF graph says what you want it to say, see:

http://isegserv.itd.rl.ac.uk/schemarama/

I've called this 'schemarama 2' because it's very similar in approach to the 'schemarama' work done by
Libby Miller, Dan Brickley and Leigh Dodds, see [1].

I've done this mainly because I've had a number of requests for 'validation' of SKOS data, but the
'constraints' described for SKOS Core cannot be expressed via OWL, and are not necessarily appropriate
for all uses of SKOS Core anyway.  Hence I thought to define various test cases to support different
applications of SKOS.

Any comments or suggestions most welcome.

Cheers,

Al.

[1] http://swordfish.rdfweb.org/discovery/2001/01/schemarama/

---
Alistair Miles
Research Associate
CCLRC - Rutherford Appleton Laboratory
Building R1 Room 1.60
Fermi Avenue
(Continue reading)

Permalink | Reply | 3 comments |
headers
Karl Dubost | 1 Nov 2005 22:06
Picon
Favicon

Re: Semantic Web gTLDs


Le 05-10-07 à 08:21, Dr. Francis MUGUET a écrit :
> It is not about filtering but about empowering.

The more you create borders, the more you create exclusions.

--

-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager
*** Be Strict To Be Cool ***
Permalink | Reply | 0 comments |
headers
Karl Dubost | 1 Nov 2005 22:11
Picon
Favicon

Re: Semantic Web gTLDs


Le 05-10-07 à 10:38, Dr. Francis MUGUET a écrit :
> that's interesting, but how you are to guarantee that those miscreants
> that are offering porn are going to abide by your rules ?
>  I hope to set up icra.xxx. It

Define porn and the semantics of porn taking into accounts all  
variations through cultures. It's exactly the same when we create  
specific gTLD.

/me is tempted to open a "non porn" Web site under the .xxx domain.  
How do we become a qualified porn content publisher?

--

-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager
*** Be Strict To Be Cool ***
Permalink | Reply | 0 comments |
headers
Kjetil Kjernsmo | 1 Nov 2005 23:54

Specifying units of the RDF geo:alt property


Hi all!

I'm new to both the geowanking list and the semantic-web list, but not 
new to either topic, allthough I cannot claim to be experienced either.

The first thing I'd like to discuss is the alt property of the 
http://www.w3.org/2003/01/geo/wgs84_pos# namespace.

Currently, it is undefined what it is, it could be anything: It could be 
the distance from the global ellipsoid focus, or it could be the 
distance above the local ellipsoid, and everything in between. And it 
could be measured in any unit, whether meter or feet or ancient 
sumerian cubits. 

In a recent chat on #swig on irc.freenode.net, Dan Brickley suggested I 
should take the issue to these list, and he would implement any 
consensus that is formed here.

RDF provides data types and also mechanisms for specifying units, but I 
think the geo:alt property is allready in widespread use. It is also 
suggested in the RDF Primer, 
http://www.w3.org/TR/rdf-primer/#properties
that one may in certain cases specify the unit of a property. 

I would personally tend to favour that the geo:alt property is specified 
as "The altitude above the local WGS 84 ellipsoid in meters", but 
clearly, this is problematic if there is allready a lot of data in the 
wild where it is given in e.g. feet. Then, I guess, the only hope is to 
use rdf:value, as seen in the example
(Continue reading)

Permalink | Reply | 1 comment |
headers
Danny Ayers | 2 Nov 2005 10:45
Picon
Gravatar

Re: schemarama 2


On 11/1/05, Miles, AJ (Alistair) <A.J.Miles <at> rl.ac.uk> wrote:

> http://isegserv.itd.rl.ac.uk/schemarama/

Wonderful.

I'd be interested in hearing your thoughts on how suitable this
approach might be with Atom/OWL [1]. This is basically the Atom
syndication modelled in RDF/OWL (work in progress, expect a
request-for-feedback post to list sometime soon). There is already a
lot of scope with Atom for validation at the syntax level, but I've
been playing a little with validation at the model level using OWL
constraints (some notes at [2]). I anticipate this will be pretty
limited compared to the syntax stuff, but thought it could be useful
for ensuring some level of sanity for Atom data when in the RDF/OWL
world. (A particular variety of app I envisage is an Atom Store built
on a triplestore, so it would have sources/sinks of Atom
format/protocol). If I understand correctly, the Schemarama approach
could take this even further towards the constraints given in the Atom
spec. Does that make sense?

Cheers,
Danny.

[1] http://atomowl.org/
[2] http://dannyayers.com/archives/2005/10/18/atomowl-ontology-testing/

--
(Continue reading)

Permalink | Reply | 2 comments |
headers
Charles McCathieNevile | 2 Nov 2005 11:47
Picon
Favicon

Re: Specifying units of the RDF geo:alt property


Hi Kjetil,

On Tue, 01 Nov 2005 23:54:44 +0100, Kjetil Kjernsmo <kjetil <at> kjernsmo.net>  
wrote:

> Currently, it is undefined what it is...
>
> In a recent chat on #swig on irc.freenode.net, Dan Brickley suggested I
> should take the issue to these list, and he would implement any
> consensus that is formed here.
...
> I would personally tend to favour that the geo:alt property is specified
> as "The altitude above the local WGS 84 ellipsoid in meters", but
> clearly, this is problematic if there is allready a lot of data in the
> wild where it is given in e.g. feet. Then, I guess, the only hope is to
> use rdf:value, as seen in the example

I'd like to see people using datatypes to do this. In the SWAD-E geoInfo  
workshop in Budapest we came up with a couple of simple approaches to  
describing what floor of a building you are on [1] - something that is  
available to many people who don't have a GPS and rely on looking up their  
wgs84 points on some map. (I have no idea what the altitude of my office  
is in any unit except floors).

I know that implementors get upset if there are tags required that are  
long, but it is not actually terribly complex to use a datatype, except  
that with no clear mechanism for defining them in a machine-readable way  
you currently rely on some magic mechanism for dealing with them.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Richard Newman | 2 Nov 2005 14:43
Picon
Picon

Re: schemarama 2


Danny,
   I was going to reimplement schemarama using twinql's matchers a  
month or two ago, and got sidetracked.

   If you can give me a graph pattern that an Atom/OWL node has to  
match (e.g., an entry has to have a title, an author who is a person,  
etc.), then twinql can compile that into a function that says 'yes'  
or 'no'. This graph pattern is a SPARQL SELECT query, basically.

   Sound useful?

-R

On 2 Nov 2005, at 09:45, Danny Ayers wrote:

>
> On 11/1/05, Miles, AJ (Alistair) <A.J.Miles <at> rl.ac.uk> wrote:
>
>
>> http://isegserv.itd.rl.ac.uk/schemarama/
>>
>
> Wonderful.
>
> I'd be interested in hearing your thoughts on how suitable this
> approach might be with Atom/OWL [1]. This is basically the Atom
> syndication modelled in RDF/OWL (work in progress, expect a
> request-for-feedback post to list sometime soon). There is already a
> lot of scope with Atom for validation at the syntax level, but I've
(Continue reading)

Permalink | Reply | 1 comment |
headers
Miles, AJ (Alistair | 2 Nov 2005 16:41
Picon
Picon

RE: schemarama 2


Hey Danny,

I just spent a bit of time reading up on your atomOWL work, nice!

Also nice to see use of skos:editorialNote :)  I'd like to support use of SKOS Core annotation-like
properties (skos:prefLabel, skos:altLabel, skos:definition ...) in OWL ontologies, without
necessarily going outside OWL DL, I wrote this up as an issue with possible solution at [3], any comments welcome.

I don't know atom well enough to know what 'constraints' you want to express, but here's a few examples from
the SKOS Core thesaurus test case [1] you might find handy ...

For SKOS Core there is a need to express 'constraints' that are contingent upon values of language tags in
literals, e.g. 'no concept should have more than one preferred label per language'.  The Schemarama 2 test
looks like:

PREFIX skos: <http://www.w3.org/2004/02/skos/core#>
PREFIX : <http://purl.org/net/schemarama#>
CONSTRUCT
{
	[] a :Error;
		:message 'Resource [1] has more than one preferred lexical label [2][3] in a given language.';
		:implicated ( ?x ?l ?m );
	.
}
WHERE
{
	?x skos:prefLabel ?l; skos:prefLabel ?m.
	FILTER ( str(?l) != str(?m) && lang(?l) = lang(?m) )
}
(Continue reading)

Permalink | Reply | 501 comments |
headers
Danny Ayers | 2 Nov 2005 17:01
Picon
Gravatar

Re: schemarama 2


On 11/2/05, Richard Newman <r.newman <at> reading.ac.uk> wrote:
> Danny,
>    I was going to reimplement schemarama using twinql's matchers a
> month or two ago, and got sidetracked.
>
>    If you can give me a graph pattern that an Atom/OWL node has to
> match (e.g., an entry has to have a title, an author who is a person,
> etc.), then twinql can compile that into a function that says 'yes'
> or 'no'. This graph pattern is a SPARQL SELECT query, basically.
>
>    Sound useful?

Sounds spot on!

It's still pretty much in flux, but to give you an idea of what we've
been looking at, the current version of the Atom/OWL ontology is at
[1], some instance examples towards the top of [2].

One significant difference between the approach here and previous
RSS/Atom-in-RDF efforts is that the versioning of entries is
explicitly modelled. If you did a blog post today and tomorrow
reposted it correcting a spelling mistake, both versions could coexist
in the target store (same id, different updated, different content).
This should fit better with the Semantic Web world view, but does lead
to some complications, not least in keeping consistent with the Atom
spec (and its naming scheme).

Cheers,
Danny.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Danny Ayers | 2 Nov 2005 17:08
Picon
Gravatar

Re: schemarama 2


Thanks Alistair.

I need to take time to digest this, but on first glance it does sound
like one of the main kinds of constraint the Atom spec imposes is
covered - once & once only occurrence of a lot of the literals, pretty
much the same as 'no concept should have more than one preferred label
per language'.

Cheers,
Danny.

On 11/2/05, Miles, AJ (Alistair) <A.J.Miles <at> rl.ac.uk> wrote:
> Hey Danny,
>
> I just spent a bit of time reading up on your atomOWL work, nice!
>
> Also nice to see use of skos:editorialNote :)  I'd like to support use of SKOS Core annotation-like
properties (skos:prefLabel, skos:altLabel, skos:definition ...) in OWL ontologies, without
necessarily going outside OWL DL, I wrote this up as an issue with possible solution at [3], any comments welcome.
>
> I don't know atom well enough to know what 'constraints' you want to express, but here's a few examples from
the SKOS Core thesaurus test case [1] you might find handy ...
>
> For SKOS Core there is a need to express 'constraints' that are contingent upon values of language tags in
literals, e.g. 'no concept should have more than one preferred label per language'.  The Schemarama 2 test
looks like:
>
> PREFIX skos: <http://www.w3.org/2004/02/skos/core#>
> PREFIX : <http://purl.org/net/schemarama#>
(Continue reading)

Permalink | Reply | 501 comments |

Gmane