W3C home > Mailing lists > Public > public-xg-webid@w3.org > December 2011

RE: broken turtle

From: Peter Williams <home_pw@msn.com>
Date: Thu, 22 Dec 2011 19:40:37 -0800
Message-ID: <SNT143-W40006F12B6F317998CD77F92AB0@phx.gbl>
To: <bergi@axolotlfarm.org>
CC: "public-xg-webid@w3.org" <public-xg-webid@w3.org>


  Im 99% certain there is a significant flaw in your tester. No matter what I do (and Ive done it 3 times, from scratch each time), for a cert with the same URI and same key as an earlier cert (with a different set of [multiple] URIs), your site (even for a new SSL session) uses a cached value (of the old 6-URI cert). It appears to have 6 URIs (when the one actually presented has only 1). Thus I cannot test whether my new https URI works (until I get itchy and go change the name). But, a flaw there probably is - thats worth fixing. In general, the crappy feature of CGI storing a client cert in the SSL sessio buffer is not helping us). If there is one thing to be fixed by browser makers  (moving on a decade), its that. we need real client certs. What is actually presented on a connection/session is what the CGI/servlet gets. > Date: Fri, 23 Dec 2011 01:59:48 +0100
> From: bergi@axolotlfarm.org
> To: home_pw@msn.com
> CC: j.jakobitsch@semantic-web.at; public-xg-webid@w3.org
> Subject: Re: broken turtle
> 
> The mime type for turtle is: text/turtle
> 
> turtle spec:
> http://www.w3.org/TeamSubmission/turtle/#sec-mime
> 
> wikipedia:
> http://en.wikipedia.org/wiki/Turtle_%28syntax%29
> 
> Am 23.12.2011 01:27, schrieb Peter Williams:
> > 
> > 
> > I assumed the same - server/service configuration error. But the
> > semantic web is just a set of document byte streams, tagged with mime
> > types, sitting on endpoints streaming UTF-8 bytes (for turtle).
> > 
> > 
> > So, I checked the external interface for obvious issues. I didnt use
> > curl, but did use an equivalent external interface tester.
> > 
> > 
> > see http://tinyurl.com/6thzu4x This shows the external view (with
> > content type, and byte level information transfer from the wire)
> > 
> > 
> > 
> > More importantly (since its my data format mapper service)
> > morper.talis.com cannot read any triples either.
> > 
> > 
> > 
> > Some magic is required, beyond HTTP wire compliance.
> > 
> > 
> > 
> > When talis puts the #tagged URI on the get, the server objects
> > (correctly). Similarly the server attempts to deliver the object
> > names in part %23 - which doesnt exist).
> > 
> > 
> > 
> > 
> > 
> > Talis does claim to read a document, when teh webid is represented
> > with the tagless path. it just treats the document as a null set of
> > triples.
> > 
> > 
> > 
> > Im guessing there is some wrong semicolon, or something. Perhaps
> > there is a improper space before a dot, or something? Seems an old
> > fashioned language, all super fussy about punctionation.
> > 
> > 
> > 
> > 
> > 
> >> Date: Fri, 23 Dec 2011 01:04:28 +0100 From:
> >> j.jakobitsch@semantic-web.at To: home_pw@msn.com CC:
> >> public-xg-webid@w3.org Subject: Re: broken turtle
> >> 
> >> hi,
> >> 
> >> your server might not be conf'd properly.
> >> 
> >> 1. when i check you'r the webID with tomcat's webIDRealm, openRDF
> >> api tries to parse with the RDFXMLParser, so there might be
> >> something wrong with the mime-type. 2. have you tried the several
> >> responses from your server when using curl 2.1. curl
> >> "http://rapstr1.blob.core.windows.net/ods/user.ttl" 2.2. curl
> >> "http://rapstr1.blob.core.windows.net/ods/user.ttl#me" 2.3. curl
> >> "http://rapstr1.blob.core.windows.net/ods/user.ttl%23me"
> >> 
> >> wkr http://www.turnguard.com/turnguard
> >> 
> >> 
> >> ----- Original Message ----- From: "Peter Williams"
> >> <home_pw@msn.com> To: public-xg-webid@w3.org Sent: Thursday,
> >> December 22, 2011 11:53:15 PM Subject: RE: broken turtle
> >> 
> >> 
> >> @prefix peter: < https://rapstr1.blob.core.windows.net/ods/user.ttl
> >> #> .
> >> 
> >> if I use the explicit https scheme for the prefix of the subject,
> >> and the SAN URI has an http URI, I assume there will no match.
> >> 
> >> Since the RDfa example is relative, perhaps consider make the
> >> turtle sample relative. Just do them the same way, that is, so the
> >> SAN URI controls the http/https issues in the namespace. i think
> >> this means make the subjkect's namespace the default (and move cert
> >> namespace to an explicit tag).
> >> 
> >> but what do I know...
> >> 
> >> 
> >> 
> >> 
> >> From: home_pw@msn.com To: public-xg-webid@w3.org Date: Thu, 22 Dec
> >> 2011 14:31:00 -0800 Subject: broken turtle
> >> 
> >> 
> >> 
> >> Since Ive had two sites able to read a blog post with an bit of
> >> annotated HTML act as a webid profile (tied to a cert), I
> >> professionalized a bit.
> >> 
> >> I built a turtle file, per the spec, and published it on a restful
> >> file server (azure blob service). I configured things so the mime
> >> type is application/turtle.
> >> 
> >> Whats wrong? I tried it on FOAFSSL.ORG, which produced information
> >> document in this report: http://wp.me/p1fcz8-1Kf
> >> 
> >> -- | Jürgen Jakobitsch, | Software Developer | Semantic Web Company
> >> GmbH | Mariahilfer Straße 70 / Neubaugasse 1, Top 8 | A - 1070
> >> Wien, Austria | Mob +43 676 62 12 710 | Fax +43.1.402 12 35 - 22
> >> 
> >> COMPANY INFORMATION | http://www.semantic-web.at/
> >> 
> >> PERSONAL INFORMATION | web : http://www.turnguard.com | foaf :
> >> http://www.turnguard.com/turnguard | skype : jakobitsch-punkt
> >> 
> > 
 		 	   		  
Received on Friday, 23 December 2011 03:41:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 23 December 2011 03:41:18 GMT