.htaccess a major bottleneck to Semantic Web adoption / Was: Re: RDFa vs RDF/XML and content negotiation

[David Booth-6]

 On Wed 08/07/09  5:08 PM , Olivier Rossel olivier.rossel@... sent:
> Do you mean that all deferencable URIs of a RDF document should have
> their domain name to end with t-d-b.org, so their resolution leads to
> the TDB server which redirects to the final location?

No, I'm not suggesting that *all* deferenceable RDF URIs should use t-d-b.org.  I'm just pointing out that it is an alternative if you cannot configure your own server to do 303 redirects.  Using it does require putting "http://t-d-b.org?" at the beginning of your URI, so if you do not want to do that then you should use a different approach.  To be clear, if you use this approach, then instead of writing a URI such as

          http://example/mydata.rdf

you would write it as

         http://t-d-b.org?http://example/mydata.rdf

and if that URI is dereferenced, the 303-redirect service will automatically return a 303 redirect to

        http://example/mydata.rdf

David Booth

>
> On Wednesday, July 8, 2009, David Booth <david@dbooth
> .org> wrote:> On Wed, 2009-07-08 at 15:50 +0100,
> Pierre-Antoine Champin wrote:> [ . . . ]
> >> ok, the solutions proposed here (by myself
> and others) still involve>> editing the .htaccess.
> >
> > Once again, use of a 303-redirect service such
> as> http://thing-described-by.org/ or
> http://t-d-b.org/> does not require *any* configuration or
> .htaccess editing.  It does not> address the problem of setting the content type
> correctly, but it *does*> provide an easy way to generate 303 redirects,
> in conformance with "Cool> URIs for the Semantic Web":
> > http://www.w3.org/TR/cooluris/#r303gendocument>
> > Hmm, I thought the use of a 303-redirect service
> was mentioned in "Cool> URIs for the Semantic Web", but in looking
> back, I see it was in "Best> Practice Recipes for Publishing RDF
> Vocabularies":> http://www.w3.org/TR/swbp-vocab-pub/#redirect
> > Maybe it should be mentioned in a future version
> of the Cool URIs> document as well.
> >
> >
> > --
> > David Booth, Ph.D.
> > Cleveland Clinic (contractor)
> >
> > Opinions expressed herein are those of the
> author and do not necessarily> reflect those of Cleveland Clinic.
> >
> >
> >
>
>
>
>
>
>

[kidehen]

Martin Hepp (UniBW) wrote:
> Google has just changed the wording of the documentation:
>
> http://knol.google.com/k/google-rich-snippets/google-rich-snippets/32la2chf8l79m/1# 
>
>
> The mentioning of cloaking risk is removed. While this is not final
> clearance,
> it is a nice sign that our concerns are heard.

Great work!

Doing the right thing is what matters, and I do believe that Google (as
they've just demonstrated), Yahoo! etc. all want to make the Web a
better and smarter place etc. :-)

Kingsley

>
> Best
> Martin
>
>
> Martin Hepp (UniBW) wrote:
>> Dear all:
>> Fyi - I am in contact with Google as for the clarification of what
>> kind of empty div/span elements are considered acceptable in the
>> context of RDFa. It may take a few days to get an official statement.
>> Just so that you know it is being taken care of...
>>
>> Martin
>>
>>
>>
>> Mark Birbeck wrote:
>>> Hi Martin,
>>>
>>>  
>>>> b) download RDFa snippet that just represents the RDF/XML content
>>>> (i.e. such
>>>> that it does not have to be consolidated with the "presentation
>>>> level" part
>>>> of the Web page.
>>>>    
>>>
>>> By coincidence, I just read this:
>>>
>>>   Hidden div's -- don't do it!
>>>   It can be tempting to add all the content relevant for a rich snippet
>>>   in one place on the page, mark it up, and then hide the entire block
>>>   of text using CSS or other techniques. Don't do this! Mark up the
>>>   content where it already exists. Google will not show content from
>>>   hidden div's in Rich Snippets, and worse, this can be considered
>>>   cloaking by Google's spam detection systems. [1]
>>>
>>> Regards,
>>>
>>> Mark
>>>
>>> [1]
>>> <http://knol.google.com/k/google-rich-snippets/google-rich-snippets/32la2chf8l79m/1#>
>>>
>>>
>>>  
>>
>

--


Regards,

Kingsley Idehen      Weblog: http://www.openlinksw.com/blog/~kidehen
President & CEO
OpenLink Software     Web: http://www.openlinksw.com

[Roberto García González]

Hi Martin, all,

I would like to point to something that might be useful for RDF data
publishing. The ReDeFer RDF2HTML service
(http://rhizomik.net/redefer/) renders input RDF/XML data as HTML for
user interaction (e.g. as used in http://rhizomik.net/rhizomer/). Now,
it also embeds RDFa that facilitates retrieving the source RDF back.

I've tested it with a pair of GoodRelations examples:
http://rhizomik.net/redefer-services/rdf2html?rdf=http://www.heppnetz.de/projects/goodrelations/minimalExampleGoodRelations.owl
http://rhizomik.net/redefer-services/rdf2html?rdf=http://www.heppnetz.de/projects/goodrelations/goodrelationsExamplesPrimerFinalOWL.owl

I've been able to check that it works for these examples by comparing
the triples generated by RDFa Distiller and RDFa Bookmarklet from the
previous HTML+RDFa pages to those generated by any23 and Triplr from
the original OWL files.

The generated HTML+RDFa can be then used in order to publish RDF just
by Cut&Paste, e.g. using an online editor like FCKEditor. This has
been the procedure followed in order to publish the RDF in
http://rhizomik.net/redefer/rdf2html/minimalExampleGoodRelations/

The HTML+RDFa view might be customised using CSS and made more usable
if the source RDF contains rdfs:labels for the involved resources,
which are used instead of the last part of the URIs if available.

In any case, if it is no to be shown to the user, it is easier to just
model triples using hidden <spans> instead of using this service...

Best regards,


Roberto García
http://rhizomik.net/~roberto

PD: Caution, this is work in progress. Feedback appreciated :-)



On Wed, Jul 8, 2009 at 12:59 PM, Martin Hepp
(UniBW)<martin.hepp@...> wrote:

> Google has just changed the wording of the documentation:
>
> http://knol.google.com/k/google-rich-snippets/google-rich-snippets/32la2chf8l79m/1#
>
> The mentioning of cloaking risk is removed. While this is not final
> clearance,
> it is a nice sign that our concerns are heard.
>
> Best
> Martin
>
>
> Martin Hepp (UniBW) wrote:
>>
>> Dear all:
>> Fyi - I am in contact with Google as for the clarification of what kind of
>> empty div/span elements are considered acceptable in the context of RDFa. It
>> may take a few days to get an official statement. Just so that you know it
>> is being taken care of...
>>
>> Martin
>>
>>
>>
>> Mark Birbeck wrote:
>>>
>>> Hi Martin,
>>>
>>>
>>>>
>>>> b) download RDFa snippet that just represents the RDF/XML content (i.e.
>>>> such
>>>> that it does not have to be consolidated with the "presentation level"
>>>> part
>>>> of the Web page.
>>>>
>>>
>>> By coincidence, I just read this:
>>>
>>>  Hidden div's -- don't do it!
>>>  It can be tempting to add all the content relevant for a rich snippet
>>>  in one place on the page, mark it up, and then hide the entire block
>>>  of text using CSS or other techniques. Don't do this! Mark up the
>>>  content where it already exists. Google will not show content from
>>>  hidden div's in Rich Snippets, and worse, this can be considered
>>>  cloaking by Google's spam detection systems. [1]
>>>
>>> Regards,
>>>
>>> Mark
>>>
>>> [1]
>>> <http://knol.google.com/k/google-rich-snippets/google-rich-snippets/32la2chf8l79m/1#>
>>>
>>>
>>
>
> --
> --------------------------------------------------------------
> martin hepp
> e-business & web science research group
> universitaet der bundeswehr muenchen
>
> e-mail:  mhepp@...
> phone:   [phone]
> fax:     [phone]
> www:     http://www.unibw.de/ebusiness/ (group)
>        http://www.heppnetz.de/ (personal)
> skype:   mfhepp twitter: mfhepp
>
> Check out the GoodRelations vocabulary for E-Commerce on the Web of Data!
> ========================================================================
>
> Webcast:
> http://www.heppnetz.de/projects/goodrelations/webcast/
>
> Talk at the Semantic Technology Conference 2009: "Semantic Web-based
> E-Commerce: The GoodRelations Ontology"
> http://tinyurl.com/semtech-hepp
>
> Tool for registering your business:
> http://www.ebusiness-unibw.org/tools/goodrelations-annotator/
>
> Overview article on Semantic Universe:
> http://tinyurl.com/goodrelations-universe
>
> Project page and resources for developers:
> http://purl.org/goodrelations/
>
> Tutorial materials:
> Tutorial at ESWC 2009: The Web of Data for E-Commerce in One Day: A Hands-on
> Introduction to the GoodRelations Ontology, RDFa, and Yahoo! SearchMonkey
>
> http://www.ebusiness-unibw.org/wiki/GoodRelations_Tutorial_ESWC2009
>
>
>
>
>

[Olivier Rossel]

Externalizing the 303 feature is the good idea, imo.

But such a service should also handle the content negociation feature.
So the 303 may redirect to different URLs depending on the content
negociated. This makes the service more complex internally but
provides a very relevant service for RDF publishers (i.e they just
have to take care of one config on their server : mime-types).
Plus managing the redirect  is as easy as changing the xml:base of
their RDF/XML.



On Wednesday, July 8, 2009, David Booth <david@...> wrote:

>  On Wed 08/07/09  5:08 PM , Olivier Rossel olivier.rossel@... sent:
>> Do you mean that all deferencable URIs of a RDF document should have
>> their domain name to end with t-d-b.org, so their resolution leads to
>> the TDB server which redirects to the final location?
>
> No, I'm not suggesting that *all* deferenceable RDF URIs should use t-d-b.org.  I'm just pointing out that it is an alternative if you cannot configure your own server to do 303 redirects.  Using it does require putting "http://t-d-b.org?" at the beginning of your URI, so if you do not want to do that then you should use a different approach.  To be clear, if you use this approach, then instead of writing a URI such as
>
>           http://example/mydata.rdf
>
> you would write it as
>
>          http://t-d-b.org?http://example/mydata.rdf
>
> and if that URI is dereferenced, the 303-redirect service will automatically return a 303 redirect to
>
>         http://example/mydata.rdf
>
> David Booth
>
>>
>> On Wednesday, July 8, 2009, David Booth <david@dbooth
>> .org> wrote:> On Wed, 2009-07-08 at 15:50 +0100,
>> Pierre-Antoine Champin wrote:> [ . . . ]
>> >> ok, the solutions proposed here (by myself
>> and others) still involve>> editing the .htaccess.
>> >
>> > Once again, use of a 303-redirect service such
>> as> http://thing-described-by.org/ or
>> http://t-d-b.org/> does not require *any* configuration or
>> .htaccess editing.  It does not> address the problem of setting the content type
>> correctly, but it *does*> provide an easy way to generate 303 redirects,
>> in conformance with "Cool> URIs for the Semantic Web":
>> > http://www.w3.org/TR/cooluris/#r303gendocument>
>> > Hmm, I thought the use of a 303-redirect service
>> was mentioned in "Cool> URIs for the Semantic Web", but in looking
>> back, I see it was in "Best> Practice Recipes for Publishing RDF
>> Vocabularies":> http://www.w3.org/TR/swbp-vocab-pub/#redirect
>> > Maybe it should be mentioned in a future version
>> of the Cool URIs> document as well.
>> >
>> >
>> > --
>> > David Booth, Ph.D.
>> > Cleveland Clinic (contractor)
>> >
>> > Opinions expressed herein are those of the
>> author and do not necessarily> reflect those of Cleveland Clinic.
>> >
>> >
>> >
>>
>>
>>
>>
>>
>>
>

[Richard Light]

In message <4A532D33.2070401@...>, Kingsley Idehen
<kidehen@...> writes
>>> PS: any IIS user volunteering to translate those recipies to IIS
>>>configuration?
>>
>> I have implemented the 303 redirection strategy in IIS, but using a
>>custom 404 "page not found" error handler.  Is that relevant to this
>>discussion?
>>
>> Richard
>Absolutely!

OK, here goes:

Attached is an IIS Active Server Page which handles 404 "page not found"
errors.  The logic is that since the abstract URLs/PSIs representing a
concept/subject do not have a corresponding page, they will end up here.

The 404 handler contains rules which pattern-match on the requested URL
and Accept header, and specify an action and a mapped URL for each.
These rules deal with both the initial 303 redirection of the incoming
request, and the delivery of the actual RDF resource when the client
subsequently requests the re-directed URL (which is also a mythical
page!).

1. Rule for simple re-direction
ReDirect_Add "/object/*", "/object/rdf/*", "SEEALSO",
"application/rdf+xml", sRules

SEEALSO implements a "303 See Other" redirect, substituting the second
pattern for the first in the URL when the Accept header specifies
"application/rdf+xml".

2. Rule for direct delivery of RDF content
ReDirect_Add "/object/rdf/*",
"rdf?class=object&target=4&level=3&application=Object&filename=WTcoll&ind
exname=Identity number&indexkey=*", "EXECUTE", "", sRules

EXECUTE actually delivers the requested content.  It does this by
substituting the URL for the second pattern, then actually fetching that
content (which is an XML resource), converting it on the fly to RDF
using an XSLT transform, and returning it as the response to the
(second) HTTP request.

Initially I thought I could do all of this by redirection, so that a
request for the "/object/rdf/*" pattern would simply be handed off yet
again by a second redirect.  However, I found that (for me, at least)
this didn't work.  Hence the need for "grunt" code within the 404
handler, actually delivering RDF content.  Anyone wanting to use this
ASP handler would, of course, have to modify the code which generates
the RDF: this is the routine GetRDF.  (If you're using the same strategy
as me, i.e. on-the-fly conversion of XML, then this may just be a case
of changing the base URL from which the XML comes, and the name of the
XSLT transform.)

Having put all of this in place, it was easy for me to add code to
deliver a Topic Map representation of the same resource, simply adding
another pair of rules.  You could use the same approach to deliver N3,
etc.

Richard



--
Richard Light

[d1fvb]

On Thu, Jul 9, 2009 at 12:27, Richard Light<richard@...> wrote:

>
> Attached is an IIS Active Server Page which handles 404 "page not found"
> errors.  The logic is that since the abstract URLs/PSIs representing a
> concept/subject do not have a corresponding page, they will end up here.

This sounds interesting. Is it possible to prevent IIS to send the 404
status code before continuing processing?

Regards,
Peter

[Richard Light]

In message <7b9ad66d0907090406t37f29aa0qc7271d094914a81@...>,
Peter Krantz <peter.krantz@...> writes
>>
>> Attached is an IIS Active Server Page which handles 404 "page not found"
>> errors.  The logic is that since the abstract URLs/PSIs representing a
>> concept/subject do not have a corresponding page, they will end up here.
>
>This sounds interesting. Is it possible to prevent IIS to send the 404
>status code before continuing processing?

Yes, that's the whole idea.  By the time the 404 handler has finished
its work, the client will receive either a 303 response (for a
redirect), a 200 response (when it actually delivers content), or a 404
(if it can't handle the URL request at all).

Richard
--
Richard Light

[Juan Sequeda]

Hi Richard

This is a great addition to the ConNeg 303 section on linkeddata.org. Awesome!

However, would you be able to host the source code and then I can just link to it from linkeddata.org?

Thanks

Juan Sequeda, Ph.D Student
Dept. of Computer Sciences
The University of Texas at Austin
www.juansequeda.com
www.semanticwebaustin.org

On Thu, Jul 9, 2009 at 1:13 PM, Richard Light <richard@...> wrote:

In message <7b9ad66d0907090406t37f29aa0qc7271d094914a81@...>, Peter Krantz <peter.krantz@...> writes

Attached is an IIS Active Server Page which handles 404 "page not found"
errors.  The logic is that since the abstract URLs/PSIs representing a
concept/subject do not have a corresponding page, they will end up here.

This sounds interesting. Is it possible to prevent IIS to send the 404
status code before continuing processing?

Yes, that's the whole idea.  By the time the 404 handler has finished its work, the client will receive either a 303 response (for a redirect), a 200 response (when it actually delivers content), or a 404 (if it can't handle the URL request at all).

Richard
--
Richard Light

[Richard Light]

In message <f914914c0907090504g4b460511kdc89e99df5f138b@...>,
Juan Sequeda <juanfederico@...> writes
>Hi Richard
>
>This is a great addition to the ConNeg 303 section on linkeddata.org.
>Awesome!
>
>However, would you be able to host the source code and then I can just
>link to it from linkeddata.org?

In due course.  I'll let you know ...

Richard
--
Richard Light