See also: IRC log
MarkB: we're talking about CURIE/URI here, we started with delimiters or enclosers, square brackets
<jeremy> Just catching up on the mail: http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2005Nov/0049 is brilliant
<MarkB> If current doc is "http://xyz.com/a/b.html"
<jeremy> the following is a legal rel URI //www.w3.org
<jeremy> also the following //[aaaa::aaaa]
<jeremy> (I think, I can't remember IPv6 IP address format)
someone proposed [cc]license
<Steven> I did
Ben: do we have any further thoughts on this?
MarkB: to be honest, I don't
really care about which syntax we choose.
... our HTML audience isn't very familiar with QNames, so it
might not be a big problem
<jeremy> My only requirement is that the syntax is X [a-zA-Z0-9]+ : .* where X is a non-alphabetic
MarkB: [cc]license is not a legal URI, so it would be an okay syntax
Jeremy: as long as the syntax we agree on is not a legal URI, then I'm happy
Ben: but QNames are legal URIs
Jeremy: dc:title is a legal
URI
... in terms that it matches the URI spec
MarkB: I don't think that's a legal URI anymore
<MarkB> If we are in "http://www.w3.org/index.html"...
<jeremy> consider mailto:foo
<MarkB> ...then it used to be regarded as valid to say "http:index2.html"
MarkB: but that's kind of frowned upon now
<jeremy> path = path-abempty ; begins with "/" or is empty
<jeremy> / path-absolute ; begins with "/" but not "//"
<jeremy> / path-noscheme ; begins with a non-colon segment
<jeremy> / path-rootless ; begins with a segment
<jeremy> / path-empty ; zero characters
<jeremy> path-noscheme = segment-nz-nc *( "/" segment )
<jeremy> oops wrong rules ... try again
Jeremy: 2 sensible choices: a)
CURIEs sensibly different b) CURIEs similar and gives us
features, but requires bracketing (or something like it)
... making CURIEs look like QNAmes is a defensible position
<jeremy> URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
<jeremy> hier-part = "//" authority path-abempty
<jeremy> / path-absolute
<jeremy> / path-rootless
<jeremy> / path-empty
<jeremy> path-rootless = segment-nz *( "/" segment )
MarkB: I feel that we are on
strong ground with this proposal.
... the point that Norm got to, that Best Practices are not
great.
What do we know. The following syntaxes are workable
- [cc:license] as CURIE/URI
- [cc]license as a CURIE
scribe: and a CURIE/URI
<MarkB> [protocol]//mylocation/a/b#x
<MarkB> {$a}abc
<MarkB> {/a/b}a
<MarkB> Steven's other syntax proposal:
<MarkB> :a:b
<MarkB> [[subst]:][[scheme]://]...etc.
MarkB: the advantage of stepping
outside of QNames-proper is that we can use any substitution
mechanism we want, including some that can be Xincluded
... [missed some notes here]
... if we go for a different qname syntax, we should also add
the extra bits for substitution
Jeremy: if we wrote a CURIE note
as a standalone document
... the prefix-binding should be application-specific
... the CURIEs would be the same in SPARQL, XHTML, etc...
... but the mechanism by which prefixes are bound to URIs
*could* be different
MarkB: this would give the IPTC what they want, as well
agendum 2
take up agendumj 2
Jeremy and Ben agree to Christmas deadline for RDF/A XSLT
<scribe> ACTION: Jeremy to have XSLT ready by Christmas [recorded in http://www.w3.org/2005/11/29-swbp-minutes.html#action01]
This is scribe.perl Revision: 1.127 of Date: 2005/08/16 15:12:03 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/stepping outside of xmlns/stepping outside of QNames-proper/ No ScribeNick specified. Guessing ScribeNick: benadida Inferring Scribes: benadida WARNING: No "Topic:" lines found. Default Present: Ben_Adida, jeremy, MarkB Present: Ben_Adida jeremy MarkB WARNING: No meeting title found! You should specify the meeting title like this: <dbooth> Meeting: Weekly Baking Club Meeting WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Got date from IRC log name: 29 Nov 2005 Guessing minutes URL: http://www.w3.org/2005/11/29-swbp-minutes.html People with action items: jeremy WARNING: No "Topic: ..." lines found! Resulting HTML may have an empty (invalid) <ol>...</ol>. Explanation: "Topic: ..." lines are used to indicate the start of new discussion topics or agenda items, such as: <dbooth> Topic: Review of Amy's report[End of scribe.perl diagnostic output]