W3C home > Mailing lists > Public > public-rdf-wg@w3.org > October 2012

Re: Potential Formal Object from DERI over JSON-LD

From: David Wood <david@3roundstones.com>
Date: Thu, 18 Oct 2012 08:26:25 -0400
Message-Id: <7AEAF79E-2BCD-4439-8761-8EA11870A8A5@3roundstones.com>
Cc: Lee Feigenbaum <lee@thefigtrees.net>, RDF WG <public-rdf-wg@w3.org>
To: Michael Hausenblas <michael.hausenblas@deri.org>
Hi Michael,

On Oct 18, 2012, at 8:12, Michael Hausenblas <michael.hausenblas@deri.org> wrote:
> Lee,
> 
>> I've read this thread, the quora question, and the G+ thread and still don't really understand what you'd be objecting to, Michael. Could you clarify? Are you objecting to the use of WG resources on this effort? Are you objecting to the branding of JSON-LD as an RDF serialization? Are you objecting to the fact that it's being produced by the RDF WG and not some other group?
> 
> At the current point in time a FO is certainly not in order as the process doesn't allow it, so let's not go there. As you've read both the G+ and the Quora thread you may have noticed that I personally have never used the term FO. Manu introduced it (maybe intentionally to draw attention to it?) ...

Ah, thanks for clarifying that.  I didn't read the G+ thread because I'm mobile and bandwidth constrained this morning.

> 
> What I want is that either JSON-LD is indeed officially acknowledged as an RDF serialization or, if the JSON-LD proponents think RDF is not suitable, then stop working on it as an RDF WG deliverable.

I agree (completely) that JSON-LD must be acknowledged as an RDF serialization and on record as saying so in May, June and July during telecons.  In fact, I think my insistence on that (when you were not present) led directly to Manu's commitment to include appropriate language to that effect in the spec.  Now, I acknowledge that Manu has been reluctant to include "strong" RDF language in the spec, but we are the RDF WG after all, so we may need to work on the wording.  That's why I asked you to propose some that you and DERI would be happy with.

The cleanest way (to me, chair hat off) would seem to be to include both a weak statement about RDF in the abstract (such as "JSON-LD supports the RDF data model and is thus convertible to RDF serializations") and to include a short section describing that in more detail.  This tie to RDF is the "cost" to the JSON-LD community to having the docs published by the RDF WG and is consistent with my comments from May-July as documented in the minutes.

However, I also think that Manu, Gregg and others have thought through the "marketing" implications of filling the specs with a lot of RDF language.  It is reasonable to take their concerns into account and limit the changes to the minimum we can live with.  A middle ground needs to be found.  Can that approach work for you?

Regards,
Dave


> 
> Cheers,
>       Michael
> 
> --
> Dr. Michael Hausenblas, Research Fellow
> DERI - Digital Enterprise Research Institute
> NUIG - National University of Ireland, Galway
> Ireland, Europe
> Tel.: +353 91 495730
> http://mhausenblas.info/
> 
> On 18 Oct 2012, at 13:02, Lee Feigenbaum wrote:
> 
>> I've read this thread, the quora question, and the G+ thread and still don't really understand what you'd be objecting to, Michael. Could you clarify? Are you objecting to the use of WG resources on this effort? Are you objecting to the branding of JSON-LD as an RDF serialization? Are you objecting to the fact that it's being produced by the RDF WG and not some other group? I really don't quite understand. (Which is odd, since I have no horse in this game and no personal need/use for JSON-LD, but also no problem with its current trajectory.)
>> 
>> Also, for what it's worth, the W3C process strongly encourages technical arguments to back objections:
>> 
>> """
>> An individual who registers a Formal Objection SHOULD cite technical arguments and propose changes that would remove the Formal Objection; these proposals MAY be vague or incomplete. Formal Objections that do not provide substantive arguments or rationale are unlikely to receive serious consideration by the Director.
>> """
>> 
>> Lee
>> 
>> On 10/18/2012 7:30 AM, Michael Hausenblas wrote:
>>>> You may recall that the WG resolved to publish two of the JSON-LD docs (not all four), starting with FPWD at [1].
>>> For the record: unfortunately, at this meeting neither Richard nor myself seems to have been present. Can you confirm this please?
>>> 
>>> Cheers,
>>>       Michael
>>> 
>>> --
>>> Dr. Michael Hausenblas, Research Fellow
>>> DERI - Digital Enterprise Research Institute
>>> NUIG - National University of Ireland, Galway
>>> Ireland, Europe
>>> Tel.: +353 91 495730
>>> http://mhausenblas.info/
>>> 
>>> On 18 Oct 2012, at 12:20, David Wood wrote:
>>> 
>>>> Hi Michael and all,
>>>> 
>>>> You may recall that the WG resolved to publish two of the JSON-LD docs (not all four), starting with FPWD at [1].
>>>> 
>>>> It would seem that your specific concerns regard marketing, not technology.  Manu has already committed to "put a section on RDF in the spec" [2].  I'm sure that he can adjust the wording if needed, but threatening a formal objection on non-technical grounds seems counterproductive.  Instead, can you please suggest some alternative wording for the spec?  Thanks.
>>>> 
>>>> Regards,
>>>> Dave
>>>> 
>>>> [1] http://www.w3.org/2011/rdf-wg/meeting/2012-07-11#resolution_1
>>>> [2] http://www.w3.org/2011/rdf-wg/meeting/2012-06-20#line0268
>>>> 
>>>> 
>>>> On Oct 18, 2012, at 4:57, Michael Hausenblas <michael.hausenblas@deri.org> wrote:
>>>> 
>>>>> Thank you, Manu - you beat me to it ;)
>>>>> 
>>>>> Just to clarify: this is not about the quality or the amount of work that went into JSON-LD. Neither do I want to discuss its usefulness. I acknowledge that there are use cases where JSON-LD certainly serves well.
>>>>> 
>>>>> ## Why, oh why?
>>>>> 
>>>>> We're faced with a situation ATM that the JSON-LD proponents talk with two different groups: on the one hand us here in the WG and on the other hand to potential adopters such as Drupal or WikiData. Towards the former group the  JSON-LD proponents keep maintaining that JSON-LD is in fact an RDF serialization. Towards the latter stake holders, the  JSON-LD proponents claim that JSON-LD has nothing to do with RDF.
>>>>> 
>>>>> You can't have the cake and eat it.
>>>>> 
>>>>> 
>>>>> ## Options
>>>>> 
>>>>> Now, to break it down, I see two options:
>>>>> 
>>>>> 1. JSON-LD is indeed considered as an official RDF serialization by the  JSON-LD proponents. Then, JSON-LD has to follow the RDF model 100% - no more exceptions, no new terms, etc.
>>>>> 2. JSON-LD is not considered as an official RDF serialization by the  JSON-LD proponents, in which case I propose to stop continuing on the REC track in the RDF WG, effective immediately.
>>>>> 
>>>>> Again, it is unfortunate that this surfaces so late in the process but I was observing the JSON-LD development (in RDF WG land and outside) for a while now and was sort of - admittedly naïvely - hoping it would sort out by itself.
>>>>> 
>>>>> 
>>>>> Cheers,
>>>>>     Michael
>>>>> 
>>>>> --
>>>>> Dr. Michael Hausenblas, Research Fellow
>>>>> DERI - Digital Enterprise Research Institute
>>>>> NUIG - National University of Ireland, Galway
>>>>> Ireland, Europe
>>>>> Tel.: +353 91 495730
>>>>> http://mhausenblas.info/
>>>>> 
>>>>> On 17 Oct 2012, at 20:18, Manu Sporny wrote:
>>>>> 
>>>>>> Michael Hausenblas wrote:
>>>>>>> (with my DERI AC rep and RDF WG member hat on) I will strongly
>>>>>>> advise the [RDF] WG to abandon REC track for JSON-LD.
>>>>>> The rest of the conversation is here:
>>>>>> 
>>>>>> https://plus.google.com/u/0/102497386507936526460/posts/KCVJVLNZKNb?cfem=1
>>>>>> 
>>>>>> Bringing it to the groups attention so we're not blind-sided by it
>>>>>> during FTF3, LC or CR.
>>>>>> 
>>>>>> -- manu
>>>>>> 
>>>>>> -- 
>>>>>> Manu Sporny (skype: msporny, twitter: manusporny)
>>>>>> President/CEO - Digital Bazaar, Inc.
>>>>>> blog: HTML5 and RDFa 1.1
>>>>>> http://manu.sporny.org/2012/html5-and-rdfa/
>>>>>> 
>>>>> 
>>> 
>>> 
>> 
> 
Received on Thursday, 18 October 2012 12:27:01 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:52 GMT