Draft document http://www.w3.org/2001/tag/awwsw/issue57/latest/

jar: Still trying to learn what to do on this doc before sending for wider review.
... heart of the doc is sec 5.5. and 5.6

http://www.w3.org/2001/tag/awwsw/issue57/latest/#chimera

michael: In the glossary, the def of "deref". Why URIs with fragIDs are not dereferenceable? You remove the fragid and then deref.

jar: HTTP doesnt' let you put the fragid in the request, so in that sense the URI isn't dereferenceable. Also, look at 3986 and see how they use the term.

dbooth: need to distinguish between direct and indirect dereferencing? Indirect is FYN.

<jar_> The fragment's format and resolution is therefore

<jar_> dependent on the media type [RFC2046] of a potentially retrieved

<jar_> representation, even though such a retrieval is only performed if the

<jar_> URI is dereferenced.

<mhausenblas> [[

<mhausenblas> A URI is dereferenceable if it may be used with a standard access mechanism to retrieve information, or to perform some other action on an associated resource ([rfc-3986] section 1.2.2). URIs possessing fragment identifiers (#) are by definition not dereferenceable. http: URIs without fragment identifiers are dereferenceable if some HTTP method (or equivalent) is successful (2xx response). Some URIs belonging to some other URI schemes are also

<mhausenblas> dereferenceable.

<mhausenblas> ]]

jar: Could clarify def in glossary.

dbooth: Sounds good. Suggest using the term "directly dereferenceable" throughout.

jar: Another possibility is to change "URI" to "fragmentless URI" where appropriate.

<jar_> 'slash URI' or 'fragmentless URI'

"fragid-less"

jar: "hashless"?

<mhausenblas> +1

dbooth, michael: good

jar: I can define "hashless URI" in the glossary.

dbooth: I have reservations about this trying to address protocols other than HTTP.

jar: larry masinter is on the TAG, and he'd want to see other schemes included.

michael: what did you mean by this in 2.2:

[[

[This use case keeps coming up (e.g. tdb:) but I don't think anyone is seriously interested in it. Need text to admit that it's important but not important enough to talk about.]

]]

jar: whether in the LD world, do you ever have a 303 redirect that does not contain the URI being defined.

dbooth: http://thing-described-by.org/ does cover this case.
... I think the topic maps people may do that.

jar: you get different answers whether you assume that the URI refers to the primary topic or not.

dbooth: I think this issue comes up more when the definition is expressed in natural language -- not when it is expressed in RDF.
... If the def is expressed in RDF I don't think there is a reliable way to distinguish between cases 2.1 and 2.2.

jar: the question is whether we need to cover case 2.2 -- whether anyone is using this technique.

Michael: Not sure how the structure of sec 3 relates to the use cases in sec 2.

jar: Section 3 is related to use case 2.1. It doesn't seem to use the word "somehow" any more.

dbooth: would be helpful to make the questions explicit in the use case, e.g., "Where should Alice publish the def?"
... sec 3.1, what does "Put the definition in the document in which the URI occurs. " mean?
... give names to documents that are mentioned, to be clear about which one is meant.
... The doc seems to talk both about the mechanics of how a def is provided and obtained, and about the semantics of what a URI means, as 5.6 talks about IRs.

jar: 5.6 needs to talk about both, to make sense.
... looking at 5.5

dbooth: Statements like "Carol can straighten this out" suggest that there is a problem that *needs* to be straightened out. But if Ch can both have foo:mass and have a dc:creator, then there is no problem to be straightened out.

<jar_> DB and I have been arguing about this for years and have never managed to communicate

And in an *application*, which is the point of doing this, a CH can perfectly fine have both.

wow, still 404 after 12 minutes: http://www.w3.org/2011/03/29-awwsw-minutes.html

This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: dbooth
Inferring Scribes: dbooth
Default Present: jar, DBooth, mhausenblas
Present: jar DBooth mhausenblas
Got date from IRC log name: 29 Mar 2011
Guessing minutes URL: http://www.w3.org/2011/03/29-awwsw-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.