On 4 Feb 2013, at 15:42, Pierre-Antoine Champin <pierre-antoine.champin@liris.cnrs.fr> wrote:
> Henry,
>
> On Mon, Feb 4, 2013 at 2:44 PM, Henry Story <henry.story@bblfish.net> wrote:
>
>
> If you want mechanisms to advertise restrictions on graphs for particular purposes
> then you should work on ISSUE-48 "Profile mechanism is Needed", and find some
> good use cases for it, so that we can then open it, with an idea as to what it would
> be to close it.
>
> http://www.w3.org/2012/ldp/track/issues/48
>
> I like the idea of profiles. Whether we should or not design a language to formally describe them, I'm not sure... I like Erik's idea that 'formally describe for [him] usually means "spec language is good enough"' [1].
I think that if you like the idea of profiles it is worth at least adding use cases to that issue
so that we can find out if simple spec language will be good enough or not. If there is to
be a general mechanism by which clients can find out what type of resource they can
POST to a container that has profile restrictions then simple spec language won't do. On the other hand it may not be as difficult as it seems: it may already have been defined somewhere in the W3C or on the web, it may be just a matter of defining a few relations,
it may be that a WG or Community group can do that while this group is progressing. It may be that these restrictions are not needed at all.
Whatever: putting forward use cases for Issue-48 will be the only way to tell.
Henry