Re: Definition of Collection and LDP Paging

On 04/24/2015 08:20 PM, henry.story@bblfish.net wrote:
> 
>> On 24 Apr 2015, at 19:58, James M Snell <jasnell@gmail.com> wrote:
>>
>> On Fri, Apr 24, 2015 at 10:53 AM, henry.story@bblfish.net
>> <henry.story@bblfish.net> wrote:
>>>
>>>> On 24 Apr 2015, at 18:34, James M Snell <jasnell@gmail.com> wrote:
>>>>
>>>> Please keep in mind that AS2 is not RDF.
>>>
>>> Come on, the examples contain Turtle, RDFa and Json-LD and it has
>>> an ontology at the bottom.
>>> [snip]
>>
>> And only the JSON-LD is normative. The Turtle and RDFa examples are
>> illustrative.
> 
> JSON-LD is an RDF format is it not? Then the minimal test of correctness
> is that  the other examples describe isomorphic graphs as defined by the
> RDF Semantics specification. 
I already added automated tests which compares Turtle examples to
JSON-LD ones and started PR for similar tests for RDFa examples!
https://travis-ci.org/jasnell/w3c-socialwg-activitystreams

(looks like something got broken in last merges :( )


> 
>> If you have a concrete proposal to improve the non-normative RDF
>> alignment, then by all means submit a PR. If the proposal is to remove
>> the existing paging properties ("next", "prev", "first", "last" etc)
>> in favor of a normative dependency on LDP Paging, I'm afraid I'm
>> definitely -1 on that.
> 
> My argument here is about the modelling. I am ok for having the next,
> prev, first and last links also appear in the document in addition to the
> http headers as described in the LDP paging spec, but I think there
> should be alignement there semantically ( especially since both are defined
> in terms of the same IETF spec ). I think that doing this can simplify the model
> too.
> 
> I am not one for making PRs before even discussing the possibilities.
> It requires first getting the model right, then the job of adjusting
> the spec can be made, given that it could have a lot of consequences.
> But it is the job of the group to make sure the specs produced here
> are consistent with other work in the W3C .

Henry, you can start a PR right now with some oncrete changes you
propose e.g. remoivng owl:FunctionalProperty and renaming items to
contains to start aligning better with LDP terminology. If you keep
those changes as small atomic commits James can cherry-pick what he
already agrees to merge and in the PR you can continue conversation with
nice inline images and code snippets :)

Cheers!

Received on Friday, 24 April 2015 18:34:27 UTC