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

[Pat Hayes]

With the sincerest respect, Tom, your attitude here is part of the  
problem. Maybe, along with many other people, I am indeed still stuck  
in the mid-1990s. You have permission to be as condescending as you  
like. But still, here I am, stuck. Thoroughly stuck. So no amount of  
condescending "sooo-20th-century, my dear" chatter is going to  
actually enable me to get to a place where I can do what you think I  
should be doing. I cannot use a rewrite rule to catch incoming  
requests, or do whatever you are talking about here. I live in an  
environment where I simply do not have access at all to the workings  
of my server at a level that close to the metal, because it is already  
woven into a clever maze of PHP machinery which is too fragile to  
allow erks like me to mess with it. Some of the best W3C techies have  
taken a look, and they can't find a way through it, either. Maybe Im  
in a special position, but I bet a whole lot of people, especially in  
the corporate world, are in a similar bind. System level access to a  
server is quite a different beast than being allowed to publish HTML  
on a website somewhere. I can, and do, publish HTML, or indeed just  
about any file I like, but I don't get to insert code. So 6 lines or  
600, it makes no difference.

But in any case, this is ridiculous. RDF is just XML text, for  
goodness sake.  I need to insert lines of code into a server file,  
and write PHP scripts, in order to publish some RDF or HTML?  That is  
insane. It would have been insane in the mid-1990s and its even more  
insane now. IMO, it is you (and Tim and the rest of the W3C) who are  
stuck in the past here.  Most Web users do not, and will not, write  
code. They will be publishing content in a cloud somewhere, even  
further away from the gritty world of scripts and lines of code than  
people - most people - are now. Most actual content providers are  
never going to want to even know that PHP scripts exist, let alone be  
obliged to write or copy one. Martin is exactly right: this is a MAJOR  
bottleneck to SWeb adoption. Its up to the people in the TAG to listen  
to this fact and do something about it, not to keep issuing useless  
'best practice' advice that cannot be followed by 99% of the world.

RDF should be text, in documents. One should be able to use it without  
knowing about anything more than the RDF spec and the XML spec. If it  
requires people to tinker with files with names starting with a dot,  
or write code, or deploy scripts, then the entire SWeb architecture is  
fundamentally broken.

Pat Hayes


On Jun 25, 2009, at 1:30 PM, Tom Heath wrote:

> Hi Martin, all,
>
> 2009/6/25 Martin Hepp (UniBW) <martin.hepp@...>:
>> Hi all:
>>
>> After about two months of helping people generate RDF/XML metadata  
>> for their
>> businesses using the GoodRelations annotator [1],
>> I have quite some evidence that the current best practices of using
>> .htaccess are a MAJOR bottleneck for the adoption of Semantic Web
>> technology.
>
> Are you referring to the best practices at [1]? Unfortunately the
> recipes in that document that use .htaccess and mod_rewrite for conneg
> no longer count as best practices, precisely due to mod_rewrite and
> .htaccess not being adequate for the conneg/303-redirects pattern.
> This has been a known issue since WWW2007 at least, and documented at
> [2] in July 2007. As far as I know, that recipes document hasn't yet
> been updated/deprecated :( (please someone correct me if I'm wrong).
>
> The easiest pattern I've found is to use a RewriteRule to catch all
> incoming requests and pass them through a small PHP script that
> examines the Accept header and sends back 303s (or 200s) as
> appropriate. The code is about 6 lines; I'll publish it somewhere if I
> didn't already.
>
> Admittedly, this doesn't solve the problem of access to .htaccess
> files. This bottleneck sounds to me like someone circa mid-1990s
> saying "my sysadmins won't let me have access to space on the web
> server". I guess we need to use lessons learned from that era to
> address the problems of this one. Anyway fancy doing a Linked Data for
> Sysadmins tutorial at a sysadmin conference?
>
> Cheers,
>
> Tom.
>
> [1] http://www.w3.org/TR/swbp-vocab-pub/
> [2] http://lists.w3.org/Archives/Public/public-swbp-wg/2007Jul/0001.html
>
>
>

------------------------------------------------------------
IHMC                                     [phone] or [phone]
40 South Alcaniz St.           [phone]   office
Pensacola                            [phone]   fax
FL 32502                              [phone]   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes

[Olivier Rossel]

Could you please elaborate a little bit about the requirement
concerning the content-type of RDF being application/rdf+xml.
At the moment, I name my RDF files with .rdf.xml extension, and it
works pretty well (the content-type is text/xml, but it is ok for me).
I have never felt the need to change that to a more specific content-type.



On Thu, Jun 25, 2009 at 9:34 PM, Martin Hepp
(UniBW)<martin.hepp@...> wrote:

> As mostly, recently ;-), I agree with Kingsley - I did not want to say that
> proper usage of http is bad or obsolete. But it turned out unfeasible for
> broad adoption my owners of small Web sites.
>
> For huge data sources and for vocabularies, the current recipes are fine.
> But I want every single business in the world to use GoodRelations for
> publishing at least their opening hours - 19 Million companies in Europe
> alone. I cannot explain to every single one of them how to configure their
> server.
>
> Another thing that might have gone lost in the discussion: Even though we
> knew the recipes, helping the site owners was difficult, because we
> experienced hundreds of different environments - preexisting .htaccess, MS
> IIS, hoster-specific scenarios, etc. So the problem is really that such a
> low-level technique is not feasible if you face so much diversity as far as
> the target system is concerned.
>
> Maybe some day a certain LOD/SW package will be installed by default on most
> servers. But we cannot wait till then.
>
> BTW: We did not even require the full beauty of LOD best practices. We
> simply want them to do as described here:
>
> http://www.ebusiness-unibw.org/wiki/GoodRelations_Recipe_8
>
> Best
> Martin
>
> Kingsley Idehen wrote:
>>
>> Giovanni Tummarello wrote:
>>>>
>>>> That can then be inserted as code snippets via copy-and-paste to any
>>>> XHTML
>>>> document.
>>>>
>>>> Any opinions?
>>>>
>>>
>>> Great, why bother with any other solution.
>>> even talking about any other solution is extraordinarely bad for the
>>> public perception of the semantic web community.
>>>
>>> Giovanni
>>>
>>>
>>>
>>
>> Giovanni,
>>
>> We don't need mutual exclusivity re. Linked Data Deployment.
>>
>> There's nothing wrong with an array of options that cover a broad range of
>> Linked Data deployment circumstances.
>>
>> HTTP is the essence of the Web (what makes it what it is), and Content
>> Negotiation is intrinsic to HTTP.
>>
>> Don't throw the baby out with the bathwater, really.
>>
>>
>
> --
> --------------------------------------------------------------
> 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
>
>
>
>
>

[Jeremy Carroll-2]

Pat Hayes wrote:
>
> RDF should be text, in documents. One should be able to use it without
> knowing about anything more than the RDF spec and the XML spec. If it
> requires people to tinker with files with names starting with a dot,
> or write code, or deploy scripts, then the entire SWeb architecture is
> fundamentally broken.
>

Largely agreeing with you Pat, I think I would want to go a step further
and say that you should be able to use RDF without knowing anything
about the RDF spec or the XML spec, or any other spec. Web users are not
required to read the specs.

Using RDF includes publishing it. The "infrastructure" whatever that is
should achieve the ability to publish my data in an appropriate way.

Jeremy

[Diego Berrueta]

Hi Tom, all,

El 25/06/2009, a las 20:30, Tom Heath escribió:
> Are you referring to the best practices at [1]? Unfortunately the
> recipes in that document that use .htaccess and mod_rewrite for conneg
> no longer count as best practices, precisely due to mod_rewrite and
> .htaccess not being adequate for the conneg/303-redirects pattern.
> This has been a known issue since WWW2007 at least, and documented at
> [2] in July 2007. As far as I know, that recipes document hasn't yet
> been updated/deprecated :( (please someone correct me if I'm wrong).

A revision of the Recipes is coming down the pipe. The new version  
will slightly improve how this issue is tackled, but it won't provide  
a complete solution. I believe the document is fair wrt this point: it  
acknowledges the problem and it provides "best effort" recipes to  
partially implement conneg using exclusively Apache directives (as  
opposed to implementing them using PHP, for instance).

I would like to emphasize that there are two different issues here:

1) the Recipes do not correctly implement conneg in its full extension  
and,

2) the Recipes are difficult to deploy due to the inability to access  
to the .htaccess file.

 From my POV, both issues are serious, and consequently I also think  
that RDFa is a much more (1) correct and (2) user-friendly way to  
publish RDF data.

Best,

Diego.

>
> The easiest pattern I've found is to use a RewriteRule to catch all
> incoming requests and pass them through a small PHP script that
> examines the Accept header and sends back 303s (or 200s) as
> appropriate. The code is about 6 lines; I'll publish it somewhere if I
> didn't already.
>
> Admittedly, this doesn't solve the problem of access to .htaccess
> files. This bottleneck sounds to me like someone circa mid-1990s
> saying "my sysadmins won't let me have access to space on the web
> server". I guess we need to use lessons learned from that era to
> address the problems of this one. Anyway fancy doing a Linked Data for
> Sysadmins tutorial at a sysadmin conference?
>
> Cheers,
>
> Tom.
>
> [1] http://www.w3.org/TR/swbp-vocab-pub/
> [2] http://lists.w3.org/Archives/Public/public-swbp-wg/2007Jul/0001.html
>

--
Diego Berrueta
R&D Department  -  CTIC Foundation
E-mail: diego.berrueta@...
Phone: [phone]
Parque Científico Tecnológico Gijón-Asturias-Spain
www.fundacionctic.org

[Richard Cyganiak-2]

Martin,

On 25 Jun 2009, at 18:44, Martin Hepp (UniBW) wrote:
> As a consequence, we will modify our tool so that it generates  
> "dummy" RDFa code with span/div that *just* represents the meta-data  
> without interfering with the presentation layer.
> That can then be inserted as code snippets via copy-and-paste to any  
> XHTML document.

By "dummy" RDFa markup, do you mean something that is completely  
invisible, so the data wouldn't actually appear on the page, e.g. by  
using style="display:hidden"? That would worry me a bit, because the  
content is invisible without a special RDFa or GoodRelations  
processor, and hence will probably not be kept up-to-date as well as  
information that's plainly visible in a web browser. For evidence,  
have a look at the average SemWeb geek's FOAF profile. Hence, I would  
suggest to keep the information visible. Sorry if I misunderstood and  
that was your intention all along!

Just an idea: How about creating a nicely styled static HTML page that  
sort of looks like an official certificate, with a nice big  
GoodRelations logo etc, and all the data is on the page, annotated  
with RDFa? Might be cleaner than hidden RDFa markup, and a static HTML  
page is hopefully easier to deploy than content-negotiated RDF/XML.

Re .htaccess, you are completely right, it's not an option. For  
average users, the only way they will get content negotiation right is  
if their server (e.g. Virtuoso) or installable web application (e.g.  
Neologism) supports it out of the box.

(On the value of content negotiation in general: I think the key point  
is that any linked data URI intended for re-use, when put into a  
browser by the average person interested in linked data publishing,  
MUST return something human-readable. That's a hard requirement,  
otherwise people will never be confident about what a particular URI  
means and hence they won't re-use. That was the thinking behind the  
Cool URIs note when Leo and I wrote it a few years ago. In the past,  
the only way to get that effect was with content negotiation, so even  
though content negotiation is a pain, it's what we had to do. In the  
present, we have an alternative thanks to RDFa. In the future, maybe  
there'll be a day when the average linked data user will have a  
browser that supports linked data out of the box.)

Best,
Richard

>
> Any opinions?
>
> Best
> Martin
>
> [1]  http://www.ebusiness-unibw.org/tools/goodrelations-annotator/
>
> Danny Ayers wrote:
>> Thank you for the excellent questions, Bill.
>>
>> Right now IMHO the best bet is probably just to pick whichever format
>> you are most comfortable with (yup "it depends") and use that as the
>> single source, transforming perhaps with scripts to generate the
>> alternate representations for conneg.
>>
>> As far as I'm aware we don't yet have an easy templating engine for
>> RDFa, so I suspect having that as the source is probably a good  
>> choice
>> for typical Web applications.
>>
>> As mentioned already GRDDL is available for transforming on the fly,
>> though I'm not sure of the level of client engine support at present.
>> Ditto providing a SPARQL endpoint is another way of maximising the
>> surface area of the data.
>>
>> But the key step has clearly been taken, that decision to publish  
>> data
>> directly without needing the human element to interpret it.
>>
>> I claim *win* for the Semantic Web, even if it'll still be a few  
>> years
>> before we see applications exploiting it in a way that provides real
>> benefit for the end user.
>>
>> my 2 cents.
>>
>> Cheers,
>> Danny.
>>
>>
>>
>
> --
> --------------------------------------------------------------
> 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
>
>
>
>
> <martin_hepp.vcf>

[kidehen]

Martin Hepp (UniBW) wrote:

> As mostly, recently ;-), I agree with Kingsley - I did not want to say
> that proper usage of http is bad or obsolete. But it turned out
> unfeasible for broad adoption my owners of small Web sites.
>
> For huge data sources and for vocabularies, the current recipes are
> fine. But I want every single business in the world to use
> GoodRelations for publishing at least their opening hours - 19 Million
> companies in Europe alone. I cannot explain to every single one of
> them how to configure their server.
>
> Another thing that might have gone lost in the discussion: Even though
> we knew the recipes, helping the site owners was difficult, because we
> experienced hundreds of different environments - preexisting
> .htaccess, MS IIS, hoster-specific scenarios, etc. So the problem is
> really that such a low-level technique is not feasible if you face so
> much diversity as far as the target system is concerned.
>
> Maybe some day a certain LOD/SW package will be installed by default
> on most servers. But we cannot wait till then.
>
> BTW: We did not even require the full beauty of LOD best practices. We
> simply want them to do as described here:
>
> http://www.ebusiness-unibw.org/wiki/GoodRelations_Recipe_8
>
> Best
> Martin

Martin,

So for this particular Linked Data deployment and consumption scenario,
we are going to end up with the following:

Content creators:
Describe your stuff (you, your offerings, your needs etc.) using RDFa
within (X)HTML. Use your current publishing worflow to publish your RDFa
embellished docs.

User Agents (Browsers, Crawlers, etc.):
Get RDFa enabled (i.e., become an RDFa processor) so you can do
something with Linked Data for Web users that enriches their overall
experience courtesy of #1 above.

Re. Virtuoso, you've always been able to SPARQL against any (X)HTML
resource using Virtuoso's SPARQL processor (which leverages the Sponger
Middleware & its RDFa Cartridge); simply use the RDFa embellished
document's URL as the Named Graph IRI in your SPARQL query.

Examples:

1. http://linkeddata.uriburner.com/sparql
2. http://lod2.openlinksw.com/sparql
3. http://bbc.openlinksw.com/sparql
4. Any other Virtuoso instance that enables the Sponging.


Kingsley

>
> Kingsley Idehen wrote:
>> Giovanni Tummarello wrote:
>>>> That can then be inserted as code snippets via copy-and-paste to
>>>> any XHTML
>>>> document.
>>>>
>>>> Any opinions?
>>>>    
>>>
>>> Great, why bother with any other solution.
>>> even talking about any other solution is extraordinarely bad for the
>>> public perception of the semantic web community.
>>>
>>> Giovanni
>>>
>>>
>>>  
>> Giovanni,
>>
>> We don't need mutual exclusivity re. Linked Data Deployment.
>>
>> There's nothing wrong with an array of options that cover a broad
>> range of Linked Data deployment circumstances.
>>
>> HTTP is the essence of the Web (what makes it what it is), and
>> Content Negotiation is intrinsic to HTTP.
>>
>> Don't throw the baby out with the bathwater, really.
>>
>>
>

--


Regards,

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

[Mark Birbeck-4]

Hi Jeremy/Pat,

On Thu, Jun 25, 2009 at 10:16 PM, Jeremy Carroll<jeremy@...> wrote:

> Pat Hayes wrote:
>>
>> RDF should be text, in documents. One should be able to use it without
>> knowing about anything more than the RDF spec and the XML spec. If it
>> requires people to tinker with files with names starting with a dot, or
>> write code, or deploy scripts, then the entire SWeb architecture is
>> fundamentally broken.
>>
>
> Largely agreeing with you Pat, I think I would want to go a step further and
> say that you should be able to use RDF without knowing anything about the
> RDF spec or the XML spec, or any other spec. Web users are not required to
> read the specs.
>
> Using RDF includes publishing it. The "infrastructure" whatever that is
> should achieve the ability to publish my data in an appropriate way.

I guess what you're getting at here is the more general point about
tools hiding RDF, but I'd like to add some comments on HTML more
specifically.

HTML publishing really took off when it became easy for anyone to
publish. Originally that was word processors that converted their
output to HTML, but you still had to deploy your document to a server.

Then it was CMS systems that could be used as easily by small
businesses and schools as large corporations, but you still needed to
have a server.

Then it became blogs, where someone else did the server install, and
you just typed in the content.

Then it was wikis -- ditto.

Now it's Google Docs, Facebook pages, Tweets, and more.

In other words, publishing HTML just gets easier and easier, and
that's the infrastructure that's important -- the HTML publishing
infrastructure.

So to publish RDF, we should simply be leveraging that enormous infrastructure.

This theme was one of the major motivations for the creation of RDFa
(née RDF/XHTML), and I would say that it's an even more important
theme today; so much so that I made it the core of a presentation I
did at SemTech last week, on 'RDFa: The Semantic Web's Missing Link'.
[1].)

Regards,

Mark

[1] <http://webbackplane.com/mark-birbeck/blog/2009/06/slides-for-semtech2009-talk-on-rdfa>

--
Mark Birbeck, webBackplane

mark.birbeck@...

http://webBackplane.com/mark-birbeck

webBackplane is a trading name of Backplane Ltd. (company number
05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
London, EC2A 4RR)

[kidehen]

Jeremy Carroll wrote:

> Pat Hayes wrote:
>>
>> RDF should be text, in documents. One should be able to use it
>> without knowing about anything more than the RDF spec and the XML
>> spec. If it requires people to tinker with files with names starting
>> with a dot, or write code, or deploy scripts, then the entire SWeb
>> architecture is fundamentally broken.
>>
>
> Largely agreeing with you Pat, I think I would want to go a step
> further and say that you should be able to use RDF without knowing
> anything about the RDF spec or the XML spec, or any other spec. Web
> users are not required to read the specs.
>
> Using RDF includes publishing it. The "infrastructure" whatever that
> is should achieve the ability to publish my data in an appropriate way.
>
> Jeremy
>
>
>

Thanks to Twitter (which forces brevity), getting out to Semtech 2009,
and hours of discussion with Martin Hepp and Aldo Bucchi, we summarize
the current phenomena as follows:

If you are comfortable producing (X)HTML documents, then simply use RDFa
and terms from relevant vocabularies to describe yourself, your needs,
your offerings, and other things, clearly. Once you've done that, simply
leave the Web to do the REST :-)

Everything else is a technical detail (imho).

--


Regards,

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

[Mark Birbeck-4]

Hi Kingsley,

> If you are comfortable producing (X)HTML documents, then simply use RDFa and
> terms from relevant vocabularies to describe yourself, your needs, your
> offerings, and other things, clearly. Once you've done that, simply leave
> the Web to do the REST :-)
>
> Everything else is a technical detail (imho).

But that isn't the discussion we're having, IMHO.

We're not talking about how you or I might do it -- people comfortable
with .htaccess files, server configuration, and so on.

My understanding of the discussion that was going on, is that whilst
we all want to see the semantic web succeed (even if we all have a
different view of what the semantic web is), we're asking how exactly
it is that we can achieve it.

And for years, the solutions proposed have been somewhat mysterious;
RDF/XML, SPARQL end-points, N3, content negotiation, 303s, and so on.

You have to ask yourself at some point, do we want the data, or don't
we -- do we want people to publish stuff that we 'semwebbers' can use?
And if we do want it, then let's help them publish it.

I may be biased because I've had my nose pressed up against it for too
many years, but I believe that in this regards, RDFa is a
game-changer.

It's not GRDDL, which says 'publish whatever the hell you like and
we'll convert it'. It's not microformats, which says, 'here are a
handful of centralised vocabularies, for use on a decentralised web'.
And it's not RDF/XML, which requires you to take apart your server and
put it back together again.

It's HTML.

And everyone knows at least one way to publish HTML, don't they?

In the years that I've been involved with the RDFa work, the mental
model I have always had, is of someone using Blogger or Drupal or
something just as simple, to publish RDF. That's now possible with
RDFa, and what's even more exciting, Yahoo! and Google will pick it
up.

I realise I'm sounding like an evangelist (no doubt because I am one). :)

But my suggestion would be that we have a window of opportunity here,
to create a semantic infrastructure that is indistinguishable from the
web itself; the more metadata we can get into HTML-space, the more
likely we are to bring about a more 'semantic' web...before anyone
notices. ;)

Regards,

Mark

--
Mark Birbeck, webBackplane

mark.birbeck@...

http://webBackplane.com/mark-birbeck

webBackplane is a trading name of Backplane Ltd. (company number
05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
London, EC2A 4RR)

[Giovanni Tummarello]

On Fri, Jun 26, 2009 at 12:18 AM, Pat Hayes<phayes@...> wrote:
> With the sincerest respect, Tom, your attitude here is part of the problem.
> But in any case, this is ridiculous. RDF is just XML text, for goodness
> sake.  I need to insert lines of code into a server file,  and write PHP
> scripts, in order to publish some RDF or HTML?  That is insane. It would
> have been insane in the mid-1990s and its even more insane now. IMO, it is
> you (and Tim and the rest of the W3C) who are stuck in the past here.

Cheers for Pat \o/ ,

with all the due respect for all involved and the amount of great work
so far, lets just move ahead.

 If there is a single better simpler way to do things it should be
reccomended at once and the rest be dropped.

"celebrate diversity" and keeping more compelx specs alive while there
is no legacy "use" of it is just  a way to further lose credibilty and
ultimately relevance.

It is not a case that you find a lot of universities talking in here
but no google or yahoo or msn or whoeever else..

just sharing:

I was at last week at SemTech and had a change to have some very
interesting discussions. The big search engine guys i was talking to
were all perfectly aware of all but just decided to stay clear becouse
light years distant from what they know they can (and will) ask people
publish HTML to do in terms of metadata .

"and when we do propose something, any big web site will simply follow
it.. why should they not. ". :-)

a report, not a speculation.

Giovanni

[Bradley Allen-2]

Mark- Beautifully put. +1 on a hopefully accelerating trend towards
simplicity and ease of adoption. - BPA

Bradley P. Allen
http://bradleypallen.org
[phone]



On Thu, Jun 25, 2009 at 3:34 PM, Mark
Birbeck<mark.birbeck@...> wrote:

> Hi Kingsley,
>
>> If you are comfortable producing (X)HTML documents, then simply use RDFa and
>> terms from relevant vocabularies to describe yourself, your needs, your
>> offerings, and other things, clearly. Once you've done that, simply leave
>> the Web to do the REST :-)
>>
>> Everything else is a technical detail (imho).
>
> But that isn't the discussion we're having, IMHO.
>
> We're not talking about how you or I might do it -- people comfortable
> with .htaccess files, server configuration, and so on.
>
> My understanding of the discussion that was going on, is that whilst
> we all want to see the semantic web succeed (even if we all have a
> different view of what the semantic web is), we're asking how exactly
> it is that we can achieve it.
>
> And for years, the solutions proposed have been somewhat mysterious;
> RDF/XML, SPARQL end-points, N3, content negotiation, 303s, and so on.
>
> You have to ask yourself at some point, do we want the data, or don't
> we -- do we want people to publish stuff that we 'semwebbers' can use?
> And if we do want it, then let's help them publish it.
>
> I may be biased because I've had my nose pressed up against it for too
> many years, but I believe that in this regards, RDFa is a
> game-changer.
>
> It's not GRDDL, which says 'publish whatever the hell you like and
> we'll convert it'. It's not microformats, which says, 'here are a
> handful of centralised vocabularies, for use on a decentralised web'.
> And it's not RDF/XML, which requires you to take apart your server and
> put it back together again.
>
> It's HTML.
>
> And everyone knows at least one way to publish HTML, don't they?
>
> In the years that I've been involved with the RDFa work, the mental
> model I have always had, is of someone using Blogger or Drupal or
> something just as simple, to publish RDF. That's now possible with
> RDFa, and what's even more exciting, Yahoo! and Google will pick it
> up.
>
> I realise I'm sounding like an evangelist (no doubt because I am one). :)
>
> But my suggestion would be that we have a window of opportunity here,
> to create a semantic infrastructure that is indistinguishable from the
> web itself; the more metadata we can get into HTML-space, the more
> likely we are to bring about a more 'semantic' web...before anyone
> notices. ;)
>
> Regards,
>
> Mark
>
> --
> Mark Birbeck, webBackplane
>
> mark.birbeck@...
>
> http://webBackplane.com/mark-birbeck
>
> webBackplane is a trading name of Backplane Ltd. (company number
> 05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
> London, EC2A 4RR)
>
>

[kidehen]

Mark Birbeck wrote:

> Hi Kingsley,
>
>  
>> If you are comfortable producing (X)HTML documents, then simply use RDFa and
>> terms from relevant vocabularies to describe yourself, your needs, your
>> offerings, and other things, clearly. Once you've done that, simply leave
>> the Web to do the REST :-)
>>
>> Everything else is a technical detail (imho).
>>    
>
>  

Mark,
> But that isn't the discussion we're having, IMHO.
>
> We're not talking about how you or I might do it -- people comfortable
> with .htaccess files, server configuration, and so on.
>  

I don't see how my comments implies anything to do with people capable
of, or with privileged access to, .htaccess.

I am echoing the sentiment that concluded our conversation last week at
dinner: let people capable of writing HTML (dropped the X) also be
capable of describing things using the RDF data model courtesy of RDFa.

When you get the structured data into the Web it will actually do the
REST i.e. it's architecture kicks in and does all of the other wonderful
things such as:
enabling you to find the nearest *open* BestBuy store that carries an
esoteric MP3 model within 2 km of your current location etc..
> My understanding of the discussion that was going on, is that whilst
> we all want to see the semantic web succeed (even if we all have a
> different view of what the semantic web is), we're asking how exactly
> it is that we can achieve it.
>  
And my point is we should have a simple value proposition that is
aligned with a set of scenarios, and manifestation solutions.

To reiterate: we want people to be able add granularity to the documents
they publish to the Web. A lot of this granularity takes the form of
metadata which can be expressed using the RDF data model by embedding
RDFa into HTML.

> And for years, the solutions proposed have been somewhat mysterious;
> RDF/XML, SPARQL end-points, N3, content negotiation, 303s, and so on.
>  
No, over the years the messaging has been mangled in a myriad of ways.
Again, my goal is to focus on a simple message and from that distill
implementation scenarios. Again, there's no silver bullet for the Linked
Data deployment matter per se. Just scenarios to which deployment
recipes may be aligned.
> You have to ask yourself at some point, do we want the data, or don't
> we -- do we want people to publish stuff that we 'semwebbers' can use?
> And if we do want it, then let's help them publish it.
>  

We just need to politely tell the world: the granularity of markup has
evolved from presentation oriented semantics (HTML) through document
structure semantics (XML) to
metadata semantics (RDF).

> I may be biased because I've had my nose pressed up against it for too
> many years, but I believe that in this regards, RDFa is a
> game-changer.
>  
> It's not GRDDL, which says 'publish whatever the hell you like and
> we'll convert it'. It's not microformats, which says, 'here are a
> handful of centralised vocabularies, for use on a decentralised web'.
> And it's not RDF/XML, which requires you to take apart your server and
> put it back together again.
>
> It's HTML.
>  

Dare I agree ?

:-)
> And everyone knows at least one way to publish HTML, don't they?
>
> In the years that I've been involved with the RDFa work, the mental
> model I have always had, is of someone using Blogger or Drupal or
> something just as simple, to publish RDF. That's now possible with
> RDFa, and what's even more exciting, Yahoo! and Google will pick it
> up.
>  

I've had a simple mental model: the desire to express myself clearly,
via markup :-)
> I realise I'm sounding like an evangelist (no doubt because I am one). :)
>
> But my suggestion would be that we have a window of opportunity here,
> to create a semantic infrastructure that is indistinguishable from the
> web itself; the more metadata we can get into HTML-space, the more
> likely we are to bring about a more 'semantic' web...before anyone
> notices. ;)
>  
Again, dare I agree :-)


Kingsley
> Regards,
>
> Mark
>
>  


--


Regards,

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

[John Graybeal]

Just because it's on your server doesn't mean the visitor to the  
restaurant's web page has to know that.  (Does it?)  Hmm, maybe that  
takes us back to the .htaccess argument....

I agree the shop owner has to feel ownership.  So whatever solution  
you choose, the shop owner has to have access to the tool which  
enables its easy use, in their language and context.   I mention this  
because I don't know if the snippet solution will pass that test. It  
will be cool if it does.  (Please let us know how it turns out, you  
are the cutting edge research here! I find what you are doing very  
exciting.)

John

On Jun 25, 2009, at 12:26 PM, Martin Hepp (UniBW) wrote:

> Hi John:
> We also thought of hosting meta-data for the users, but I don't like  
> that because I want the shop operators to feel ownership for the data:
> If the opening hours expressed in RDF are wrong but on the personal  
> Web page of that restaurant, anybody facing closed doors will blame  
> the restaurant.
> If the outdated opening hours in RDF are on my SW server, the  
> unlucky customer will blame the Semantic Web for having crappy data.
>
> So maybe the snippet solution in RDFa is the best.
>
> Best
> Martin
>
>
> John Graybeal wrote:
>> This is a principal reason MMI decided to offer a vocabulary server  
>> for its community. The idea that 1000 different providers would all  
>> develop a level of web competency (for which there is evidence at  
>> only a minority of providers) for serving their RDF and OWL content  
>> -- let alone the capability to do versioning, adopt best practices,  
>> learn SKOS, and whatever other nuances are called for -- seemed  
>> like a non-starter.
>>
>> This is not exactly the same problem you're facing, but something  
>> to consider (if the model allows it) is creating a way to serve the  
>> annotations from another place than the host institution.  The  
>> institution can refer to those served files from their own sites,  
>> and even update them remotely, but not have to incur all the  
>> management overhead as standards improve, files change, authorship  
>> changes, etc.
>>
>> (Which is not to disagree with your plan either. That sounds fine.)
>>
>> One other delivery model could be for them to give you an existing  
>> HTML, you give them back the modified HTML (saves them cutting and  
>> pasting steps?).
>>
>> I'm a little ignorant on your tools and processes, so apologies if  
>> these are non-starters.
>>
>> John
>>
>>
>> On Jun 25, 2009, at 9:44 AM, Martin Hepp (UniBW) wrote:
>>
>>> Hi all:
>>>
>>> After about two months of helping people generate RDF/XML metadata  
>>> for their businesses using the GoodRelations annotator [1],
>>> I have quite some evidence that the current best practices of  
>>> using .htaccess are a MAJOR bottleneck for the adoption of  
>>> Semantic Web technology.
>>>
>>> Just some data:
>>> - We have several hundred entries in the annotator log - most  
>>> people spend 10 or more minutes to create a reasonable description  
>>> of themselves.
>>> - Even though they all operate some sort of Web sites, less than  
>>> 30 % of them manage to upload/publish a single *.rdf file in their  
>>> root directory.
>>> - Of those 30%, only a fraction manage to set up content  
>>> negotiation properly, even though we provide a step-by-step recipe.
>>>
>>> The effects are
>>> - URIs that are not dereferencable,
>>> - incorrect media types and
>>> and other problems.
>>>
>>> When investigating the causes and trying to help people, we  
>>> encountered a variety of configurations and causes that we did not  
>>> expect. It turned out that helping people just managing this tiny  
>>> step of publishing  Semantic Web data would turn into a full-time  
>>> job for 1 - 2 administrators.
>>>
>>> Typical causes of problems are
>>> - Lack of privileges for .htaccess (many cheap hosting packages  
>>> give limited or no access to .htaccess)
>>> - Users without Unix background had trouble name a file so that it  
>>> begins with a dot
>>> - Microsoft IIS require completely different recipes
>>> - Many users have access just at a CMS level
>>>
>>> Bottomline:
>>> - For researchers in the field, it is a doable task to set up an  
>>> Apache server so that it serves RDF content according to current  
>>> best practices.
>>> - For most people out there in reality, this is regularly a  
>>> prohibitively difficult task, both because of a lack of skills and  
>>> a variety in the technical environments that turns into an  
>>> engineering challenge what is easy on the textbook-level.
>>>
>>> As a consequence, we will modify our tool so that it generates  
>>> "dummy" RDFa code with span/div that *just* represents the meta-
>>> data without interfering with the presentation layer.
>>> That can then be inserted as code snippets via copy-and-paste to  
>>> any XHTML document.
>>>
>>> Any opinions?
>>>
>>> Best
>>> Martin
>>>
>>> [1]  http://www.ebusiness-unibw.org/tools/goodrelations-annotator/
>>>
>>> Danny Ayers wrote:
>>>> Thank you for the excellent questions, Bill.
>>>>
>>>> Right now IMHO the best bet is probably just to pick whichever  
>>>> format
>>>> you are most comfortable with (yup "it depends") and use that as  
>>>> the
>>>> single source, transforming perhaps with scripts to generate the
>>>> alternate representations for conneg.
>>>>
>>>> As far as I'm aware we don't yet have an easy templating engine for
>>>> RDFa, so I suspect having that as the source is probably a good  
>>>> choice
>>>> for typical Web applications.
>>>>
>>>> As mentioned already GRDDL is available for transforming on the  
>>>> fly,
>>>> though I'm not sure of the level of client engine support at  
>>>> present.
>>>> Ditto providing a SPARQL endpoint is another way of maximising the
>>>> surface area of the data.
>>>>
>>>> But the key step has clearly been taken, that decision to publish  
>>>> data
>>>> directly without needing the human element to interpret it.
>>>>
>>>> I claim *win* for the Semantic Web, even if it'll still be a few  
>>>> years
>>>> before we see applications exploiting it in a way that provides  
>>>> real
>>>> benefit for the end user.
>>>>
>>>> my 2 cents.
>>>>
>>>> Cheers,
>>>> Danny.
>>>>
>>>>
>>>>
>>>
>>> --
>>> --------------------------------------------------------------
>>> 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
>>>
>>>
>>>
>>>
>>> <martin_hepp.vcf>
>>
>>
>> John
>>
>> --------------
>> John Graybeal   <mailto:graybeal@...>  -- 831-775-1956
>> Monterey Bay Aquarium Research Institute
>> Marine Metadata Interoperability Project: http://marinemetadata.org
>>
>>
>>
>
> --
> --------------------------------------------------------------
> 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
>
>
>
>
> <martin_hepp.vcf>

John

--------------
John Graybeal   <mailto:graybeal@...>  -- 831-775-1956
Monterey Bay Aquarium Research Institute
Marine Metadata Interoperability Project: http://marinemetadata.org

[Dan Brickley-2]

+cc: Norm Walsh

On 25/6/09 19:39, Juan Sequeda wrote:
> So... then from what I understand.. why bother with content negotiation,
> right?
>
> Just do everything in RDFa, right?
>
> We are planning to deploy soon the linked data version of Turn2Live.com.
> And we are in the discussion of doing the content negotiation (a la
> BBC). But if we can KISS, then all we should do is RDFa, right?

Does every major RDF toolkit have an integrated RDFa parser already?

And yep the conneg idiom isn't mandatory. You can use # URIs, at least
if the .rdf application/rdf+xml mime type is set. I believe that's in
the Apache defaults now. At least checking here, a fresh Ubuntu
installation has "application/rdf+xml                             rdf"
in /etc/mime.types (a file which I think comes via Apache but not 100%
sure).

But yes - this is a major problem and headache. Not just around the
conneg piece, but in general. I've seen similar results to those
reported here with "write yourself a FOAF file" exercises. Even if
people use Leigh Dodd's handy foaf-a-matic webforms to author a file ...
at the end of the session they are left with a piece of RDF/XML in their
hands, and an instruction to "upload it to their sites". Even people
with blogs and facebook profiles and twitter accounts etc. can find this
daunting. And not many people know what FTP is (or was).

My suggestion here is that we look into something like OAuth for
delegating permission tokens for uploading files. OAuth is a protocol
that uses a Web/HTML for site A to request that some user of site B
allow it to perform certain constrained tasks on site B. Canonical
example being "site A (a printing company) wants to see non-public
photos on site B (a photo-sharing site)". I believe this model works
well for writing/publishing, as well as for mediating information access.

If site A is an RDF-generating site, and site B is a generic hosting
site, then the idea is that we write or find a generic OAuth-enabled
utility that B could use, such that the users of site B could give sites
like A permission to publish documents automatically. At a protocol
level, I would expect this to use AtomPub but it could also be WebDAV or
another mechanism.

But how to get all those sites to implement such a thing? Well firstly,
this isn't limited to FOAF. Or to any flavour of RDF. I think there is a
strong story for why this will happen eventually. Strong because there
are clear benefits for many of the actors:

* a data-portability and user control story: I don't want all my music
profile info to be on last.fm; I want last.fm to maintain
http://danbri.org/music for me.
* a benefits-the-data source story: I'm sure the marketing teams of
various startups would be very happy at the ability to directly push
content into 1000s of end-user sites. For the Google/Link karma, traffic
etc.
* benefits the hosts story: rather than having users share their FTP
passwords, they share task-specific tokens that can be managed and
rolled back on finer-grained basis

So a sample flow might be:

1. User Alice is logged into her blog, which is now AtomPub+OAuth enabled.
2. She clicks on a link somewhere for "generate a FOAF file from your
music interests", which takes her to a site that asks some basic
information (name, homepage) and about some music-related sites she uses.
3. That site's FOAF generator site scans her public last.fm profile
(after asking her username), and then does the same for her Myspace and
YouTube profiles.
4. It then says "OK, generated music profile! May we publish this to
your site? It then scans her homesite, blog etc via some auto-discovery
protocol(s), to see which of them have a writable AtomPub + OAuth
endpoint. It finds her wordpress blog supports this.
5. Alice is bounced to an OAuth permissioning page on her blog, which
says something like:
        "The Music Profile site at example.com  would like to
        have read and write permission for an area of your site:
once/always/never or for 6 months?"
6. Alice gives permission for 6 months. Some computer stuff happens in
the background, and the Music site is given a token it can use to post
data to Alice's site.
7. http://alice.example.com/blog/musicprofile then becomes a page (or
mini-blog or activity stream) maintained entirely, or partially, by the
remote site using RDFa markup sent as AtomPub blog entries, or maybe as
AtomPub attachments.

OK I'm glossing over some details here, such as configuration, choice of
URIs etc. I may be over-simplifying some OAuth aspects, and forgetting
detail of what's possible. But I think there is real potential in this
sort of model, and would like a sanity check on that!

Also the detail of whether different sites could/would write to the same
space or feed or not. And how we can use this as a page-publishing model
instead of a blog entry publishing model.

I've written about this before, see
http://markmail.org/message/gplslpe2k2zjuliq

Re prototyping ...

There is some wordpress+atompub code around, see
http://74.125.77.132/search?q=cache:KcriYA9UohcJ:singpolyma.net/2008/05/atompub-oauth-for-wordpress/+wordpress+oauth&cd=3&hl=en&ct=clnk&client=firefox-a

Also I just found http://rollerweblogger.org/roller/entry/oauth_for_roller

And in http://danbri.org/words/2008/10/15/380 there is a point to some
Google announcements from last year.


I hope someone on these lists will find the OAuth/AtomPub combination
interesting enough to explore (and report back on). I've also Cc:'d Norm
Walsh here who knows AtomPub inside out now.

Is this making sense to anyone but me? :)

cheers,

Dan

[Bernhard Schandl]

Mark,

> And everyone knows at least one way to publish HTML, don't they?

I disagree. Most people know how to enter text into a form, but they  
have no clue what HTML is all about, how it comes that one line of  
text is bigger while the other is smaller, how backlinks and  
permalinks are created and resolved, and so on.

The difference between publishing HTML and RDF (in whatever form) is  
exactly the difference between the Web and the Semantic Web: the  
former is "just" human-readable material, while the latter is machine-
interpretable. Just as one needs to know the grammar and vocabulary of  
English in order to publish a proper textual description of goods, one  
needs at least basic understanding of the grammar and vocabularies of  
the Semantic Web to publish proper RDF data about these goods.

Even if blogs, CMS, and whatever other tools there are support the  
user in authoring RDFa with a nice GUI, it will be still up to the  
user to correctly select property URIs and to properly format literal  
values so that they can be used by a generic (!) RDF client.

Alternative: tools that do some sort of NLP / Entity Recognition /  
Information Extraction become so mature that they can be reliably  
deployed into mainstream blogs, CMS, wikis, ...

By the way, the same applies to the client side: in terms of  
applications we must get beyond the tabular rendering of RDF data.  
This is nice, but in the end it provides not much more value to an end  
user than a nicely formatted HTML page. The things that Google and  
Yahoo do are a good step in this direction.

Best,
Bernahrd

[Dan Brickley-2]

On 25/6/09 19:06, Jeff Finkelstein, Customer Paradigm wrote:
> Martin-
>
> I agree that the .htaccess file is a big stumbling block for many people
> with low-cost hosting.  Would a lightweight php-based application that could
> write to the .htaccess  / create the RDF file work to solve this easily?

I'd suggest exploring a lightweight but standards-based PHP app that
could be dropped into a site, given a configuration and top-level
directory, such that external sites could use a OAuth interface to
negotiate write access, and then use AtomPub to post all kinds of data
into that Web space.

(per my previous post)

Actually I forgot another benefit of this model: external sites could
re-post data automatically, with no further user interaction. This
solves a lot of "my foaf file ... i generated it once but didn't keep it
up to date" problems...

Dan

[Toby Inkster-4]

On 25 Jun 2009, at 21:18, Pat Hayes wrote:

> If [RDF] requires people to tinker with files with names starting  
> with a dot [...] then the entire SWeb architecture is fundamentally  
> broken.

RDF doesn't. Apache does.

Many hosts do have front ends for configuring Apache, allowing  
redirects to be set up and content-types configured by filling in  
simple web forms. But there are such a variety of these tools with  
different capabilities and different interfaces that it would be  
difficult to produce advice suitable for them all, so instead  
".htaccess" recipes are provided instead.

That said, there are a couple of steps that Martin could remove from  
his recipe and still be promoting reasonably good practice:

Step 5a - this rewrites <http://example.org/semanticweb> to <http://
example.org/semanticweb.rdf>. Other than aesthetics, there's no real  
reason to do this. Yes, I've read timbl's old Cool URIs document, and  
understand about not wanting to include hints of file format in a  
URI. But realistically, this file is going to always include some RDF  
- perhaps in a non-RDF/XML serialisation, but I don't see anything  
inappropriate about serving other RDF serialisations using a ".rdf"  
URL, provided the correct MIME type is used.

Step 5b - the default Apache mime.types file knows about application/
rdf+xml, so this should be unnecessary. Perhaps instead have a  
GoodRelations "validator" which checks that the content type is  
correct, and only suggests this when it is found to be otherwise.

Steps 3 and 4 could be amalgamated into a single "validate your RDF  
file" step using the aforementioned validator. The validator would be  
written so that, upon a successful validation, it offers single-click  
options to ping semweb search engines, and Yahoo (via a RDF/XML-
 >DataRSS converter).

With those adjustments, the recipe would just be:

        1. Upload your RDF file.
        2. Add a rel="meta" link to it.
        3. Validate using our helpful tool.

--
Toby A Inkster
<mailto:mail@...>
<http://tobyinkster.co.uk>

[Toby Inkster-4]

On Fri, 2009-06-26 at 09:35 +0200, Dan Brickley wrote:

> Does every major RDF toolkit have an integrated RDFa parser already?

No - and even for those that do, it's often rather flaky.

Seseme/Rio doesn't have one in its stable release, though I believe one
is in development for 3.0.

Redland/Raptor often (for me at least) seems to crash on RDFa. It also
complains a lot when named entities are used (e.g.  ) even though
the XHTML+RDFa 1.0 DTD does allow them.

Jena (just testing on sparql.org) doesn't seem to handle RDFa at all.

Not really "toolkits" per se, but cwm and the current release of
Tabulator don't seem to have RDFa support. (Though I think support for
the latter is being worked on.)

For application developers who are specifically trying to support RDFa,
none of this is a major problem - it's pretty easy to include a little
content-type detection and pass the XHTML through an RDFa->XML converter
prior to the rest of your code getting its hands on it - but this does
require specific handling, which must be an obstacle to adoption.

--
Toby A Inkster
<mailto:mail@...>
<http://tobyinkster.co.uk>

[Dan Brickley-2]

On 26/6/09 10:51, Toby Inkster wrote:

> On Fri, 2009-06-26 at 09:35 +0200, Dan Brickley wrote:
>
>> Does every major RDF toolkit have an integrated RDFa parser already?
>
> No - and even for those that do, it's often rather flaky.
>
> Seseme/Rio doesn't have one in its stable release, though I believe one
> is in development for 3.0.
>
> Redland/Raptor often (for me at least) seems to crash on RDFa. It also
> complains a lot when named entities are used (e.g. ) even though
> the XHTML+RDFa 1.0 DTD does allow them.
>
> Jena (just testing on sparql.org) doesn't seem to handle RDFa at all.
>
> Not really "toolkits" per se, but cwm and the current release of
> Tabulator don't seem to have RDFa support. (Though I think support for
> the latter is being worked on.)
>
> For application developers who are specifically trying to support RDFa,
> none of this is a major problem - it's pretty easy to include a little
> content-type detection and pass the XHTML through an RDFa->XML converter
> prior to the rest of your code getting its hands on it - but this does
> require specific handling, which must be an obstacle to adoption.

Yep, pretty much as I feared. Also the Google SGAPI currently only reads
FOAF in RDF/XML form, not yet updated to use the rdfa support in Rapper.

Re app developers, it depends a lot. If your app is built inside some
framework - eg. Protege - RDFa might be quite hard to integrate. Some
apps also store to local disk rather than HTTP space, and so using
content-negotiation is tricky. RDFa files don't have any well known
file-suffix patterns either.

cheers,

Dan

[Damian Steer-3]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Toby Inkster wrote:

> Jena (just testing on sparql.org) doesn't seem to handle RDFa at all.

The jena GRDDL reader handles RDFa. [1]

I've started toying with writing a (direct) RDFa parser, too.

Damian

[1]
<http://jena.sourceforge.net/grddl/javadoc/com/hp/hpl/jena/grddl/GRDDLReader.html#setProperty(java.lang.String,%20java.lang.Object)>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpElVQACgkQAyLCB+mTtylxUACgu6V+4spvcJeGyZGoq+pInOV4
6SgAoKv3/7ivJQNLwnpCUwHgYIkH8V6F
=RDbW
-----END PGP SIGNATURE-----