- From: mca <mca@amundsen.com>
- Date: Sat, 10 Aug 2013 19:34:45 -0400
- To: Melvin Carvalho <melvincarvalho@gmail.com>
- Cc: Henry Story <henry.story@bblfish.net>, Read-Write-Web <public-rww@w3.org>, Alexandre Bertails <bertails@w3.org>, Tim Berners-Lee <timbl@w3.org>
- Message-ID: <CAPW_8m6saL1MTjL2dKr5xd2jp1288rehHy4SnGAC-nRTwZtxwg@mail.gmail.com>
NOTE - i was surprised to find that "meta" is not listed in the IANA registry[1] and does not appear in the HTML4[2] or HTML5[3] lists. It *does* appear in the microformats list[6], tho. seems like this is another oppty to standardize this shared understanding; one that is already rather will-used/established. for a while i thought you (timbl) were referring to the <meta> *element* in HTML[4][5] which has the benefit of supporting both name and content attributes. this makes it a nice "catch-all" for extending metadata in messages, but has the downside of a higher degree of indirection (i need to find a meta tag with the name of interest, etc.). IMO, the downside if relying on a rel="meta" is that this is already used as an element w/ the name + content pair. i think that sets an expectation to be dealt with if you start using the same identifier on links via @rel. maybe not... also, if you plan to use this as a generic "here is the metadata about this document" that can turn out to be a very large (and likely unpredictable) collection for machines to trundle when looking for access control information. if "acl" is not semantically appealing, maybe "security" or "access-control" or some other more appropriate identifier can be used to tell machines where to find the access control data for the current representation. Cheers. [1] http://www.iana.org/assignments/link-relations/link-relations.xml [2] http://www.w3.org/TR/REC-html40/types.html#type-links [3] http://www.w3.org/TR/2011/WD-html5-20110525/links.html#linkTypes [4] http://www.w3.org/TR/2011/WD-html5-20110525/semantics.html#the-meta-element [5] http://tools.ietf.org/html/rfc2731#section-3 [6]http://microformats.org/wiki/rel-meta mca +1.859.757.1449 skype: mca.amundsen http://amundsen.com/blog/ http://twitter.com/mamund https://github.com/mamund http://www.linkedin.com/in/mikeamundsen On Sat, Aug 10, 2013 at 4:21 PM, Melvin Carvalho <melvincarvalho@gmail.com>wrote: > > > > On 10 August 2013 10:56, Henry Story <henry.story@bblfish.net> wrote: > >> >> On 10 Aug 2013, at 00:18, Kingsley Idehen <kidehen@openlinksw.com> wrote: >> >> >> When talking about this with Alexandre Bertails he thought that >> rel="meta" was >> >> not the right relation and that rel="acl" would be more correct. >> > >> > Yes. >> > >> > It will be fixed. >> >> We need to get those who have implementations to agree on this first. :-) >> >> And I am not sure what forum is available where we can agree on edits to >> the acl ontolgy or the http://www.w3.org/wiki/WebAccessControl wiki page, >> so I am sending this mail a bit widely around. The WebAccessControl wiki >> page suggests that the RWW Community Group is the place to discuss this. >> >> I suppose for the moment the WebAccessControl wiki page plays the role of >> a >> spec. It says: >> >> [[ >> The client follows, for example, an HTTP header field: >> >> Link: <meta/profile.meta>; rel=meta >> ]] >> >> Alexandre Bertails once argued that meta is too general, and that this >> should >> be an "acl" link. Neither "acl" nor "meta" are registered in the iana >> document >> http://www.iana.org/assignments/link-relations/link-relations.xhtml >> which is I think where this needs to be registered. >> See http://tools.ietf.org/html/rfc5988#section-6.1 >> >> For us to register this we should probably have something a bit more >> spec like than the wiki page. >> >> I also would like to add to the ontology >> - support for regular expressions on urls >> - a acl:include relation to include acls from other documents >> > > FYI: some comments from Timbl on #dig > > <timbl> I think rel=meta is a better choice > <timbl> because we have running code which picks it up > <timbl> and fragmenting the market does less damage. > <timbl> I can see also an ACL file having other stuff (like ownership, > provenance, licence etc) and my gut feeling is that one flexible bag is > better than trying to defined lots of rigid bags and al algorithm for what > triples go in what. > > http://dig.csail.mit.edu/irc/dig/2013-08-10#T20-19-21 > > >> >> >> Social Web Architect >> http://bblfish.net/ >> >> >> >
Received on Saturday, 10 August 2013 23:35:34 UTC