Re: Query and access F2F document template

Olaf,

I will try to answer your questions as to what I believe, but Luc may
wish to correct me as he originally suggested those aims.

> 1.) Does the  I  here really refer to the identity of that resource state
> representation or is  I  an identifier (e.g. a URI) for that resource state
> representation?

The latter, I believe.

> 2.) About what kind of resources do we speak here?
> From reading the email alone I would assume that this can be a state
> representation of an arbitrary kind of resource. However, the corresponding
> wiki page says "A user gains access to an online resource ..."  Does that mean
> we restrict ourself here to state representations of (what I would call) Web
> resources only (i.e. resources that can be made available on the Web)?  It
> would be fine with me; I just want to ensure we all speak about the same
> kind of things.

We suggest we focus on online resources for now, but if any proposals
apply more generally then that's all for the better.

> I guess L does not represent a physical location but a location on the Web.
> Correct?

Yes.

> If so, is L explicitly dedicated to the provenance of the aforementioned
> resource state representation or is it possible to retrieve provenance of
> other things as well from that same L?

I don't think we want to make an assumption either way. Instead the
answer would be specific to a proposal.

> Do we assume here that we know L is the location to retrieve the provenance
> of the aforementioned resource state representation?

Yes, I suggest at this point we do. However, again, if anyone has
proposals which do not require this assumption they are very welcome.

> I guess you mean to obtain a representation (or description) of the
> provenance. Right?

Surely it depends what you include by the term "provenance" as to
whether they're distinct :-). But yes, data is obtained and by reading
its information content, you know the provenance of the
representation.

> Do we assume that, while the browser has the HTML document, it has no
> information about the download of that document? So we want to find I and
> L just from the HTML document?  (Which, again, is fine with me; I only want
> to know that explicitly)

Yes, the second question is, I believe, specifically asking what we
can do starting with just the document.

> What exactly is the  I  here - the identity of the HTML document?

An identifier for the document. Strictly all that is required is that
it is something which would allow you to retrieve the correct thing's
provenance.

Hope that answers your questions adequately.

Thanks,
Simon

> Greetings,
> Olaf
>
>
>> Please see the rest of the document skeleton for details:
>>   http://www.w3.org/2011/prov/wiki/F2F1_Access_and_Query_Proposal
>>
>> We welcome any comments on the skeleton structure proposed, including
>> the scope decided for this document.
>>
>> One specific request to Graham: you suggested Section 4 of the POWDER
>> as providing a solution for the above questions (at least with regard
>> to HTTP, HTML, ATOM). It looks straightforward enough to me what such
>> a solution would look like (the same as described in the POWDER
>> proposal but with provenance specific MIME types?), but it would be
>> very helpful if you could sketch the proposal on the Wiki page as you
>> understand it best.
>>
>> Thanks,
>> Simon
>
>
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>



-- 
Dr Simon Miles
Lecturer, Department of Informatics
Kings College London, WC2R 2LS, UK
+44 (0)20 7848 1166

Received on Tuesday, 7 June 2011 20:19:54 UTC