Re: major revision of LDP draft charter

>> 1. RDF types supported -- don't we need to support all the RDF types?
> 
> I don't think so.   For one thing, the list of RDF datatypes is
> unbounded -- it's extensible.   So that means we can't mandate they all
> be supported, and does that mean we mandate none of them?  The idea here
> is to pick a reasonable set as a minimum that people can count on being
> supported. 


+1


>> 2. Re. syntax, I think this is orthogonal to our concerns.
> 
> It seems to me, one of the use cases is to be able to write a client
> that you know will work with any server that conforms to this
> specification.    If we don't specify the RDF serialization to use,
> people can't write clients like that.  


+1

Ashok, you might remember the discussions we had in R2RML around the syntax bits? Or, same on the WebID XG, it was again the seemingly so plain syntax issue (RDF/XML vs. RDFa vs Turtle) that caused quite some discussions.

The syntax issue is - in terms of basic I14Y - a MUST, in my understanding. Nothing we need to discuss forever but we need to pin it down, upfront, ASAP. In addition, by embracing JSON we can establish I14Y in this area as well (cf [1]).


Cheers,
	   Michael


[1] http://webofdata.wordpress.com/2012/02/05/json-http-data-links/

--
Dr. Michael Hausenblas, Research Fellow
DERI - Digital Enterprise Research Institute
NUIG - National University of Ireland, Galway
Ireland, Europe
Tel.: +353 91 495730
WebID: http://sw-app.org/mic.xhtml#i

On 18 Mar 2012, at 19:06, Sandro Hawke wrote:

> On Sun, 2012-03-18 at 10:29 -0700, ashok malhotra wrote:
>> That's a good comment, Dan!
>> I had the opposite reaction re. the timing.  Why do we have to wait till
>> June to get start?  Why can't we start next month? :-)
> 
> As Ivan pointed out, it takes about 6 weeks after submitting it to the
> AC for review, so...   we're trying.
> 
>> Re. the technical issues, the charter makes it clear that these are some
>> possible issues and others may come up when the WG starts and some
>> may get dropped.  Nevertheless, I had a couple of comments on the issues:
>> 
>> 1. RDF types supported -- don't we need to support all the RDF types?
> 
> I don't think so.   For one thing, the list of RDF datatypes is
> unbounded -- it's extensible.   So that means we can't mandate they all
> be supported, and does that mean we mandate none of them?  The idea here
> is to pick a reasonable set as a minimum that people can count on being
> supported.   I note that OWL and RIF already did this; we might want to
> start with that same list they used, or whatever ends up in in the
> expected submission.
> 
>> 2. Re. syntax, I think this is orthogonal to our concerns.
> 
> It seems to me, one of the use cases is to be able to write a client
> that you know will work with any server that conforms to this
> specification.    If we don't specify the RDF serialization to use,
> people can't write clients like that.   The server might only do RDF/XML
> or only do RDFa or only do ntriples or only do Turtle or only do
> JSON-LD, or maybe only do some proprietary RDF serialization.  So, I
> think we need to pick one as the required minimum if we want
> out-of-the-box interoperability.
> 
>> 6. Concurrency depends on the type of storage and other concerns
> 
> I'm thinking of the kind of situation addressed by this bit:
> 
>        Because the update process involves getting a resource first,
>        and then modifying it and later putting it back on the server,
>        there is the possibility of a conflict (for example, another
>        client might have updated the resource since the GET action). To
>        mitigate this problem, Basic Profile implementations should use
>        the HTTP If-Match header and HTTP ETags to detect collisions.
> 
>        	-- http://www.ibm.com/developerworks/rational/library/basic-profile-linked-data/index.html
> 
> It seems to me that's an important point that shouldn't be forgotten.
> Perhaps I could explain that concern better.
> 
>> So, I would remove the issue and add it to the para above that
>> discusses authorization and authentication.
>> All the best, Ashok
> 
> Thanks for reviewing this so carefully (and on a Sunday!).  I hope I've
> adequately explained why we included these items....
> 
>   -- Sandro
> 
>> On 3/18/2012 10:18 AM, Dan Brickley wrote:
>>> On 18 Mar 2012, at 13:10, Eric Prud'hommeaux<eric@w3.org>  wrote:
>>> 
>>>> * Dan Brickley<danbri@danbri.org>  [2012-03-18 12:46-0400]
>>>>> Quick iphone reply for now. Basically "what's the hurry?". The whole thing seems to be based on a magazine article that is rumoured to be a potential Submission to W3C. How did we jump from that to a proposed group already?
>>>>> 
>>>>> Apologies if I've missed more context,
>>>> The bulk of the context is the LEDP workshop
>>>>  <http://www.w3.org/2011/09/LinkedData/>
>>>> at which the ~45 participants said they wanted a WG and some wanted to
>>>> help with the Submission which IBM was working on. We promised to
>>>> create a mailing list<mailto:public-ldp@w3.org>  where we would float
>>>> a proposed charter.
>>>> 
>>>> The next natural step is to float that charter by the AC once we have
>>>> a guage for how well it meets the community's needs.
>>> Thanks! In that case role of the workshop, and details on consensus amongst its attendees, should be higher visibility in abstract/intro. Otherwise the motivation feels a bit weak.
>>> 
>>> 'This Group addresses a need identified by ... who agreed ... and asked W3C to ..."? Etc
>>> 
>>> Dan
>>> 
>>> 
>>> 
>>>>> Dan
>>>>> 
>>>>> 
>>>>> 
>>>>> On 18 Mar 2012, at 12:13, Sandro Hawke<sandro@w3.org>  wrote:
>>>>> 
>>>>>> After various discussions, we've rewritten the Linked Data Platform
>>>>>> (LDP) draft charter.  New version is here:
>>>>>> 
>>>>>>       http://www.w3.org/2012/ldp/charter
>>>>>> 
>>>>>> The diff is linked from there, but only the last few paragraphs
>>>>>> (standard charter stuff) are the similar enough for the diff to be
>>>>>> useful.
>>>>>> 
>>>>>> At this point, we're expecting to formally propose this to the W3C
>>>>>> membership within a week or two, so please review it soon.
>>>>>> 
>>>>>>  -- Sandro
>>>>>> 
>>>>>> 
>>>>>> 
>>>> -- 
>>>> -ericP
>> 
> 
> 
> 

Received on Monday, 19 March 2012 05:49:36 UTC