Re: [foaf-dev] foaf validator

I've come up with a big hiccup in validating the wot namespace: the example
at http://xmlns.com/wot/0.1/ generates an N/A that it doesn't use wot.
*Test 18*: Checking if wot:assurance <http://xmlns.com/wot/0.1/assurance> and
if it is valid:

*N/A*. The document does not make use of the Web of Trust
ontology<http://xmlns.com/wot/0.1/>

I did have to remove a 'wot' from one of the elements, but other than that,
it should properly formed.  I tried the example after mine failed (I built
it to match the example).  The example you provide above generates the same
response.  Any ideas?

Cheers,
Kevin
 ---
Kevin Pickens
A+ | Network + | Security+ | Project+
[email protected] | http://kbpickens.me
+1(540)999-6257 (office) | +1(508)664-8314 (fax)



On Sat, Aug 6, 2011 at 6:08 AM, Mischa Tuffield
<[email protected]>wrote:

> Good morning,
>
> On 6 Aug 2011, at 01:13, Jonas Smedegaard wrote:
>
> On 11-08-06 at 12:18am, Mischa Tuffield wrote:
>
> Hi Jonas,
>
>
> On 5 Aug 2011, at 19:08, Jonas Smedegaard wrote:
>
> I tried validating my own FOAF file:
>
>
> http://foaf.qdos.com/validator/uri?URI=http://dr.jones.dk/me/index.rdf
>
>
> I am puzzled about the results - seems to me that the warnings for
>
> tests 1-3 is bogus as those allegedly missing declarations indeed
>
> exist in my file.  Or am I composing my FOAF data wrongly?
>
>
> when you parse http://dr.jones.dk/me/index.rdf none of the triples in
>
> the document talk about the URI which the document was fetched from.
>
> So yeah your foaf document has an rdf:type
>
> foaf:PersonalProfileDocument triple, but the subject of that tripe is
>
> <http://dr.jones.dk/me/> due to the use of :
>
>
> xml:base="http://dr.jones.dk/me/" in your rdf:xml
>
>
> Ah.  Thanks!
>
> I have fought for a long time to both support content-negotiation and
> serve RDF as nice directory URLs rather than exposing the ugly
> extension.  I use Apache2 but haven't figured out how to make it rank
> RDF higher than GPG sigs.
>
>
> So, yeah the problem the validator is choking is because of its turtle
> parser isn't great. This is a bug in the RAP PHP library, which isn't
> maintained anymore, I have forked it in github, and will look into why n3
> parser is crapping out in the future, but not today...
>
> But to your point above, your apache setup isn't quite right:
>
> When i :
>
> `curl  -H 'Accept: application/rdf+xml, application/rdf+turtle,
> application/x-turtle, text/turtle, application/xml, */*' '
> http://dr.jones.dk/me/'`
>
> I get back turtle ...
>
> RAP's rdfxml parser is way better than its turtle/n3 one, if you could have
> a look at your http://dr.jones.dk/me/ to make it return RDF/XML when a
> client says it prefers it then the validator will work, till i look at the
> n3 parsing.
>
> This is the best description of how to set up your apache instance to
> server up conneg'd RDF:
>
> http://www.w3.org/TR/swbp-vocab-pub/
>
>
>
> FWIW, I believe the validator was only giving off "Warnings" not
>
> errors.
>
>
> Yes.  I strive for perfection, and thus want also warnings fixed. :-)
>
>
> Hehe
>
>
>
> My FOAF file was generated from http://dr.jones.dk/me/index.ttl
>
> using mkfoaf.sh available below http://dr.jones.dk/foaf/ .
>
>
> yay to ttl :) I will have a look at mkfoaf.sh at some point ...
>
>
> There is not much to the script, actually.  The other scripts in same
> dir might be more interesting ;-)
>
> Took me some time to find a method that works both ways: I can generate
> RDF and from the RDF generate the exact same Turtle file.  Added that
> reverse script now as well: rdf2turtle.sh
>
>
> Um, I think there is some confusion when you use the term 'RDF', you seem
> to use it to refer to RDF/XML. This is not the case, 'RDF' is a data model
> as per the 'RDF Concepts/Abstract syntax' W3C note, RDF/XML (ugly) and
> Turtle (pretty) are both serialisation of the RDF data-model... A minor
> point, thought you appreciate it though.
>
> Furthermore, libraptor, and the command line tool which comes with it
> `rapper`, is totally the tool of awesome you want when converting between
> the various RDF serialisations. You must have that installed on your
> machine, I recall you pointing to the ubuntu bug report (which I didn't
> understand) about libraptor1 and libraptor2 dependency stuff.. hrrmm. You
> should totally check it out, it is an awesome bit of RDF conversion magic.
> And finally on this rant, I keep saying this `if libraptor can't parse it,
> it isn't RDF ...` :)
>
>
> Oh, btw: I thought I'd done the WoT setup properly using GPG signing.
> What am I missing there to be "compliant" with your validator?
>
>
> You have the following relevant triples
>
> <http://dr.jones.dk/me/> <http://xmlns.com/wot/0.1/assurance> <
> http://dr.jones.dk/me/index.rdf.asc> .
> <http://dr.jones.dk/me/index.rdf.asc> <
> http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <
> http://xmlns.com/wot/0.1/Endorsement> .
> <http://dr.jones.dk/me/index.rdf.asc> <http://xmlns.com/wot/0.1/endorser>
> <http://dr.jones.dk/me/#pubkey> .
> _:bnode0 <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <
> http://xmlns.com/wot/0.1/PubKey> .
> _:bnode0 <http://xmlns.com/wot/0.1/pubkeyAddress> <
> http://dr.jones.dk/me/index.rdf.asc> .
>
> Which are fine I guess, but they are leaving "a lot to the imagination",
> what my validator is checking for is for more of the details of your GPG key
> to be listed within the RDF, so that one does need to understand signatures
> and stuff at the point of parsing the RDF.
>
> The validator is looking for a thing, which has some details, and which is
> related to a foaf:Person URI, the foaf:Person URI of the foaf:primaryTopic
> of the foaf:Document which has a wot:assurance.
>
> This is what I have in my foaf file, and it is pretty much what the
> validator is looking for (yes, based on my reading of the WoT ontology, but
> I am open to suggestions):
>
> _:bnode1 <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <
> http://xmlns.com/wot/0.1/PubKey> .
> _:bnode1 <http://xmlns.com/wot/0.1/fingerprint>
> "18A2AF280CA59E77AE512BB39ED0B04E51F2F7EF" .
> _:bnode1 <http://xmlns.com/wot/0.1/hex_id> "51F2F7EF" .
> _:bonde1 <http://xmlns.com/wot/0.1/identity> <
> http://mmt.me.uk/foaf.rdf#mischa> .
> _:bnode1 <http://xmlns.com/wot/0.1/length> "4096"^^<
> http://www.w3.org/2001/XMLSchema#integer> .
> _:bnode1 <http://xmlns.com/wot/0.1/pubkeyAddress> <
> http://mmt.me.uk/mischa.pubkey.asc> .
> <http://mmt.me.uk/foaf.rdf>  a foaf:PersonalProfileDocument .
> <http://mmt.me.uk/foaf.rdf> foaf:primaryTopic <
> http://mmt.me.uk/foaf.rdf#mischa> .
> <http://mmt.me.uk/foaf.rdf> <http://xmlns.com/wot/0.1/assurance> <
> http://mmt.me.uk/foaf.rdf.asc> .
>
> I.e. This is a document about a person, it talks about lots of people, but
> the primary focus of the document is Alice. The document has an assurance
> document, which is signed by a PubKey, which is an identity of Alice, our
> protagonist...
>
> I hope that helps!
>
> Mischa
>
>
>
> Am liking your arabic foaf:nick,
>
> يونس
>
> In Farsi we have a sound closer to the english "Je" than the arabic
>
> "You" sounding letter at the start of the foaf:nick, you could also
>
> add a:
>
>
> <foaf:nick>جنس</foaf:nick>
>
>
> Great. Added now!
>
>
> Otherwise it all looks good, /me waves and says thanks in public for
>
> all the debian stuff you have been doing, THANK YOU!
>
>
> Thanks for the kind words.
>
>
> Mischa *goes back to watching "Walking Dead" ...
>
>
>
> Hmm - didn't know that one.  Seems interesting!
>
>
> - Jonas
>
> --
> * Jonas Smedegaard - idealist & Internet-arkitekt
> * Tlf.: +45 40843136  Website: http://dr.jones.dk/
>
> [x] quote me freely  [ ] ask before reusing  [ ] keep private
> _______________________________________________
> foaf-dev mailing list
> foaf-dev-RyYwo1q5J+qsOXdr9/[email protected]
> http://lists.foaf-project.org/mailman/listinfo/foaf-dev
>
>
> ___________________________________
> Mischa Tuffield PhD
> Email: [email protected]
> Homepage - http://mmt.me.uk/
> +44(0)208 439 8200  http://www.garlik.com/
> Registered in England and Wales 535 7233 VAT # 849 0517 11
> Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD
>
>
> _______________________________________________
> foaf-dev mailing list
> foaf-dev-RyYwo1q5J+qsOXdr9/[email protected]
> http://lists.foaf-project.org/mailman/listinfo/foaf-dev
>

_______________________________________________
foaf-dev mailing list
foaf-dev-RyYwo1q5J+qsOXdr9/[email protected]
http://lists.foaf-project.org/mailman/listinfo/foaf-dev