Re: any23 RDFa test suite results

[Stéphane Corlosquet <sc...@gmail.com>]

Hi Lewis and all,

Just checking in on the possibility of using Niklas' RDFa 1.1 compliant
library [1] before the upcoming release of any23. Given that cli-rdfa
already passes all of the official RDFa 1.1 tests, this ought to solve all
the current issues in the custom RDFa parser that any23 currently uses. See
[2] for an example.

Steph.

[1]: http://github.com/niklasl/clj-rdfa-jena
[2]:
https://github.com/niklasl/clj-rdfa-jena/blob/master/examples/rdfa_jena.groovy

On Sat, Mar 31, 2012 at 6:50 AM, Lewis John Mcgibbney <
lewis.mcgibbney@gmail.com> wrote:

> Hi Steph, Greg and others,
>
> This is very very helpful...
>
> We are currently working towards our first incubating release, and the
> comments/results opened up in Gregg's reports will certainly be taken into
> consideration in the next development drive. It would be really great, once
> we get working directly on these issues, to liaise with those familiar with
> the working draft to ensure as close to 100% PASS performance as possible.
> So please keep in touch and feel free to chime in whenever you see fit.
> Your involvement is very much valued.
>
> Thanks again for passing this on.
>
> Lewis
>
> N.B. Please note that the official Apache Any23 homepage and associated
> documentation is now hosted over at incubator.apache.org/any23, here you
> will be able to find the most up-to-date documentation on the project.
>
> On Fri, Mar 30, 2012 at 9:20 PM, Stéphane Corlosquet
> <sc...@gmail.com>wrote:
>
> > Hi Any23 devs,
> >
> > The RDFa 1.1 spec is currently in the Candidate Rec period and a test
> suite
> > is available at http://rdfa.info/test-suite/. Gregg Kellog tested any23
> > with this test suite, and I thought his results should be relevant to
> you.
> >
> > Worth noting is a new RDFa library (cli-rdfa) which passes all RDFa 1.1
> > tests [5].
> >
> > Steph.
> >
> > ---------- Forwarded message ----------
> > From: Gregg Kellogg <gr...@kellogg-assoc.com>
> > Date: Sat, Mar 24, 2012 at 4:50 PM
> > Subject: any23 RDFa test suite results
> > To: public-rdfa-wg WG <pu...@w3.org>
> >
> >
> > I did a recent update and added any23 [1] to the processor list of the
> RDFa
> > Test Suite. As any23 is widely used, and was, in fact, used for the
> recent
> > Web Data Commons structured data extraction [1]. This result is of great
> > benefit to the structured data community. As they report, they used any23
> > for extracting data from the pages. Therefore, I thought it would be
> useful
> > to see how any23 actually performs in extracting RDFa.
> >
> > I ran the RDFa Test Suite using both RDFa XHTML+RDFa 1.0 [3] and
> XHTML+RDFa
> > 1.1 [4] .
> >
> > For XHTML+RDFa 1.0, any23 passes 97 out of 117 (82%) of the tests .
> > For XHTML+RDFa 1.1, any23 passes 89 out of 163 (55%) of the tests.
> >
> > An easier way for the any23 group to go, that would likely also result in
> > improved performance, would be to use Niklas' Clojure implementation [5],
> > which also runs in the Java VM.
> >
> > Gregg
> >
> > [1] http://any23.org/
> > [2] http://webdatacommons.org/
> > [3] http://rdfa.info/earl-reports/any23-rdfa1.0-xhtml1.html
> > [4] http://rdfa.info/earl-reports/any23-rdfa1.1-xhtml1.html
> > [5] https://github.com/niklasl/clj-rdfa
> >
>
>
>
> --
> *Lewis*
>

Re: any23 RDFa test suite results

[Lewis John Mcgibbney <le...@gmail.com>]

Hi Everyone,

Firstly Steph thanks for taking this public. We are aware that our
currently implementation has some areas needing some TLC and some
amount of development to bring it up to scratch with excellent
compliance targets.

On Fri, Jun 1, 2012 at 4:32 PM, Stéphane Corlosquet
<sc...@gmail.com> wrote:
>
> Just checking in on the possibility of using Niklas' RDFa 1.1 compliant
> library [1] before the upcoming release of any23.

We are very close to pushing Apache-any23-0.7.0-incubating and at this
late stage I personally would not pursue the target of integration.
This is not because it is not suitable (this would be naive of me to
suggest) it is merely because for Any23 to progress and improve we
need to push a release (hopefully a couple) within incubation.

It would be great if some other could chime in an provide their
comments as I can only talk on my own behalf. I hope I provide
sufficient justification at least to provoke some more conversation.
Further down the line we may well consider/implement this library
however this would again be subject to review by the Any23 PPMC.

Does this sounds reasonable?

Thank you

Lewis


-- 
Lewis

Re: any23 RDFa test suite results

[Simone Tripodi <si...@apache.org>]

Hi Niklas,

EPL1 is one of the licenses labeled as "Category B" and according to
<http://www.apache.org/legal/resolved.html#category-b>:

"Software under the following licenses may be included in binary form
within an Apache product if the inclusion is appropriately labeled"

> Absolutely! Both the core clj-rdfa library and clj-rdfa-jena are
> licensed under the Eclipse Public License (the same as Clojure). I
> hope that's compatible with the Apache license?

yes, I'd say :)

All the best!
-Simo

http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/


On Thu, Jun 7, 2012 at 7:42 PM, Niklas Lindström <li...@gmail.com> wrote:
> Hello!
>
> On Thu, Jun 7, 2012 at 7:16 PM, Stéphane Corlosquet
> <sc...@gmail.com> wrote:
>> Hi Nick,
>>
>> Yes, I'm touch with the author (Niklas), I've pinged him and asked him to
>> follow up here.
>>
>> Steph.
>>
>> On Thu, Jun 7, 2012 at 5:32 AM, Nick Kew <ni...@apache.org> wrote:
>>>
>>>
>>> On 1 Jun 2012, at 16:32, Stéphane Corlosquet wrote:
>>>
>>> > Hi Lewis and all,
>>> >
>>> > Just checking in on the possibility of using Niklas' RDFa 1.1 compliant
>>> > library [1] before the upcoming release of any23. Given that cli-rdfa
>>> > already passes all of the official RDFa 1.1 tests, this ought to solve
>>> > all
>>> > the current issues in the custom RDFa parser that any23 currently uses.
>>> > See
>>> > [2] for an example.
>>>
>>> Missed that post; just saw Lewis's followup.
>>>
>>> Are you in contact with the author?  I don't see a License that would
>>> permit its inclusion in an Apache project.  Though its presence at
>>> github suggests the author would probably be happy to see it reused!
>
> Absolutely! Both the core clj-rdfa library and clj-rdfa-jena are
> licensed under the Eclipse Public License (the same as Clojure). I
> hope that's compatible with the Apache license? If not I'll gladly
> change it (to e.g. MIT), albeit since it requires Clojure to run I
> suspect its license has to be compatible anyway for this to fly..
>
> (I'm not a lawyer, but AFAIK EPL and APL are compatible at least
> insofar as EPL-licensed code is allowed to be dynamically linked to by
> APL code. I'm less certain about derivative works, there seem be some
> patent-related differences.)
>
> I already had a license statement (in the README.md) for clj-rdfa [1],
> which I just now also added to clj-rdfa-jena [2].
>
> Best regards,
> Niklas
>
> [1]: https://github.com/niklasl/clj-rdfa#license
> [2]: https://github.com/niklasl/clj-rdfa-jena#license

Re: any23 RDFa test suite results

[Niklas Lindström <li...@gmail.com>]

Hello!

On Thu, Jun 7, 2012 at 7:16 PM, Stéphane Corlosquet
<sc...@gmail.com> wrote:
> Hi Nick,
>
> Yes, I'm touch with the author (Niklas), I've pinged him and asked him to
> follow up here.
>
> Steph.
>
> On Thu, Jun 7, 2012 at 5:32 AM, Nick Kew <ni...@apache.org> wrote:
>>
>>
>> On 1 Jun 2012, at 16:32, Stéphane Corlosquet wrote:
>>
>> > Hi Lewis and all,
>> >
>> > Just checking in on the possibility of using Niklas' RDFa 1.1 compliant
>> > library [1] before the upcoming release of any23. Given that cli-rdfa
>> > already passes all of the official RDFa 1.1 tests, this ought to solve
>> > all
>> > the current issues in the custom RDFa parser that any23 currently uses.
>> > See
>> > [2] for an example.
>>
>> Missed that post; just saw Lewis's followup.
>>
>> Are you in contact with the author?  I don't see a License that would
>> permit its inclusion in an Apache project.  Though its presence at
>> github suggests the author would probably be happy to see it reused!

Absolutely! Both the core clj-rdfa library and clj-rdfa-jena are
licensed under the Eclipse Public License (the same as Clojure). I
hope that's compatible with the Apache license? If not I'll gladly
change it (to e.g. MIT), albeit since it requires Clojure to run I
suspect its license has to be compatible anyway for this to fly..

(I'm not a lawyer, but AFAIK EPL and APL are compatible at least
insofar as EPL-licensed code is allowed to be dynamically linked to by
APL code. I'm less certain about derivative works, there seem be some
patent-related differences.)

I already had a license statement (in the README.md) for clj-rdfa [1],
which I just now also added to clj-rdfa-jena [2].

Best regards,
Niklas

[1]: https://github.com/niklasl/clj-rdfa#license
[2]: https://github.com/niklasl/clj-rdfa-jena#license

Re: any23 RDFa test suite results

[Stéphane Corlosquet <sc...@gmail.com>]

Hi Nick,

Yes, I'm touch with the author (Niklas), I've pinged him and asked him to
follow up here.

Steph.

On Thu, Jun 7, 2012 at 5:32 AM, Nick Kew <ni...@apache.org> wrote:

>
> On 1 Jun 2012, at 16:32, Stéphane Corlosquet wrote:
>
> > Hi Lewis and all,
> >
> > Just checking in on the possibility of using Niklas' RDFa 1.1 compliant
> > library [1] before the upcoming release of any23. Given that cli-rdfa
> > already passes all of the official RDFa 1.1 tests, this ought to solve
> all
> > the current issues in the custom RDFa parser that any23 currently uses.
> See
> > [2] for an example.
>
> Missed that post; just saw Lewis's followup.
>
> Are you in contact with the author?  I don't see a License that would
> permit its inclusion in an Apache project.  Though its presence at
> github suggests the author would probably be happy to see it reused!
>
> --
> Nick Kew
>
>
>

Re: any23 RDFa test suite results

[Nick Kew <ni...@apache.org>]

On 1 Jun 2012, at 16:32, Stéphane Corlosquet wrote:

> Hi Lewis and all,
> 
> Just checking in on the possibility of using Niklas' RDFa 1.1 compliant
> library [1] before the upcoming release of any23. Given that cli-rdfa
> already passes all of the official RDFa 1.1 tests, this ought to solve all
> the current issues in the custom RDFa parser that any23 currently uses. See
> [2] for an example.

Missed that post; just saw Lewis's followup.

Are you in contact with the author?  I don't see a License that would
permit its inclusion in an Apache project.  Though its presence at
github suggests the author would probably be happy to see it reused!

-- 
Nick Kew