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

RE: RE: Re: a totally minimal RDFa doc, please

From: peter williams <home_pw@msn.com>
Date: Mon, 18 Apr 2011 08:52:25 -0700
Message-ID: <SNT143-ds15469C44F2C13E6158007992910@phx.gbl>
To: "'Toeroek, Laszlo \(EXT\)'" <laszlo.toeroek.ext@siemens.com>, "'Akbar Hossain'" <mail@akbarhossain.com>
CC: "'WebID XG'" <public-xg-webid@w3.org>
What matters is not crawlers - what we have to focus on is the average http
client library used by servers. 


If we think about openid, they REQUIRE a resource server to follow redirects
to pick up the XRDS file of the IDP. If I post my XRDS file to opera unite,
note, Ive found openid libraries (at resource servers) can pick up that file
- whereas such as W3C's RDFa extractor cannot (because its not crawling
around my site, but "de-ferencing" more formally). Software is evidently
using the http client library as does a crawler.


It matters to us, as the semantics of webid depend on URIs - and so called
theories of "de-referencing"). If the cert indicates a reference to URI x,
and the site (or an attacking proxy) does a 302 redirect from URI x to URI y
(on a different domain, say), this is material to the webid validation
agent. This spec is rather silent on these topics.





From: public-xg-webid-request@w3.org [mailto:public-xg-webid-request@w3.org]
On Behalf Of Toeroek, Laszlo (EXT)
Sent: Sunday, April 17, 2011 11:41 PM
To: peter williams; 'Akbar Hossain'
Cc: 'WebID XG'
Subject: RE: RE: Re: a totally minimal RDFa doc, please


RDFa is defined in the context of XHTML, but most of the parsers tolerate
(broken) legacy HTML pages. Crawler's of major search engines surely do so,
otherwise they would be forced to ignore 90% of the web.



From: public-xg-webid-request@w3.org [mailto:public-xg-webid-request@w3.org]
On Behalf Of peter williams
Sent: Samstag, 16. April 2011 20:13
To: 'Akbar Hossain'
Cc: 'WebID XG'
Subject: RE: RE: Re: a totally minimal RDFa doc, please

Will the average resource site directed to pick out the RDFa from any old
HTML or XHTML page? Or does it NEED to be carefully crafted with the right
Document type headers, etc. 


If I cannot get the Opera unit stuff to work (and Im about to give up), what
I want is enxt to do to be able to take the content pattern you shoed
(substituted with a 3rd parties user's URI and cert mod), email it to the
3rd party with the counsel that they stick it on a blog post for which only
they have write access. But, obviously, that blog page is not going to be a
dedicated XHTML+RDfa document type; it will be some random bit of HTML,
XHTML - the usual, non-conforming, 80% working, web mess.



From: peter williams [mailto:home_pw@msn.com] 
Sent: Thursday, April 14, 2011 9:59 AM
To: 'Akbar Hossain'
Cc: 'WebID XG'
Subject: RE: RE: Re: a totally minimal RDFa doc, please


That what I did. I copied your page, byte for byte, from a socket spy.
Perhaps I changed a byte too many and upset a (unwebby) ultra fussy syntax? 


The only thing I changed in the graph part was the about ref, and the
cert:id. I only did that after it didn't work when simply hosting your page
with no change (hoping it was an RDF issue.)




Im *guessing* that what the browser sees is NOT what the server http client
sees (be it an extractor client, or uriburner client). There are definitely
redirects and cookies and other stuff going on,  when I spy on the handoffs.
What the betting that the average resource server is not going to be willing
to leverage these "browser" centric features of the endpoint?


If I invoke the extractor, it seems to analyze some intermediate page.


Im trying really hard to use consumer stuff (and not run a webserver). 



From: akkiehossain@gmail.com [mailto:akkiehossain@gmail.com] On Behalf Of
Akbar Hossain
Sent: Thursday, April 14, 2011 8:52 AM
To: peter williams
Cc: mail@akbarhossain.com; WebID XG
Subject: Re: RE: Re: a totally minimal RDFa doc, please



If you run your page thru the rdfa distiller there is no key. Perhaps copy
my page without too many edits first. 


On 14 Apr 2011 15:38, "peter williams" <home_pw@msn.com> wrote:
> Almost perfect - in that there is (i) total minimum of markup, (ii) anyone
can understand that, and (iii) a lovely rdfa extractor proves its viable.
> So, I stuff a variant on
<http://home.homepw2.operaunite.com/webserver/content/>  . and the extractor
doesn't work! I guess uriburner wont work either, for some a related
(unknown) reasons.
> Sigh. 
> But thanks. It's the right idea.
> From: public-xg-webid-request@w3.org
[mailto:public-xg-webid-request@w3.org] On Behalf Of Coralie Mercier
> Sent: Thursday, April 14, 2011 4:53 AM
> To: WebID XG
> Subject: Fwd: Re: a totally minimal RDFa doc, please
> [Messages from "Akbar Hossain" <mail@akbarhossain.com> seem to not end up
on the public-xg-webid@w3.org list and archive. I'm investigating.
Meanwhile, I'm forwarding the message to the list.]
> ------- Forwarded message -------
> From: "Akbar Hossain" <mail@akbarhossain.com>
> To: "peter williams" <home_pw@msn.com>
> Subject: Re: a totally minimal RDFa doc, please
> Date: Thu, 14 Apr 2011 10:23:04 +0200
> Perhaps look at the source of my home page. http://akbarhossain.com/
> You could remove the rdfa validator link, the email verifier link and pem
> Maybe make the cert details non visible elements of the page ...
> Endless possibilities.
> (I generated the page from the second tab of http://x509.me
> The test a certificate option. My cert was generated of the first page
(optional). My cert had a SAN pointing to a blank page to start with. Press
the test button on the second tab. It fails as it was a blank page and spits
out the rdfa required for it to pass. Cut, copy, paste.)
> Hope that helps.
> Thanks.
> On 14 Apr 2011 05:07, "peter williams" <home_pw@msn.com> wrote:
> -- 
> Coralie Mercier - Communications Team - Incubator Activity Lead 
> World Wide Web Consortium - http://www.w3.org
> W3C/ERCIM - N212 - 2004, rte des lucioles - 06410 Biot - FR
> mailto:coralie@w3.org +33492387590 http://www.w3.org/People/CMercier/
Received on Monday, 18 April 2011 15:52:56 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:39:44 UTC