See also: IRC log
<scribe> scribe: Gregory_Rosmaita
<scribe> scribeNick: oedipus
<Roland> Agenda: http://lists.w3.org/Archives/Public/public-xhtml2/2008Mar/0015.html
<markbirbeck> zakim code?
<Steven_> -> Previous http://www.w3.org/2008/03/05-xhtml-minutes.html
CSS Namespace review submitted by StevenP
SP: discuss briefly -- traditionally taken
second slot (thurs/fri) at tech plenary -- asking if have preference, i say
no
... need to find out which groups we clash with -- XForms, PF -- any other
groups want to avoid scheduling clash
... don't care if mon/tues or thurs/fri as long as doesn't conflict with
Forms and PF
... tech plenary in nice
RM: send response along SP's lines?
GJR: +1
<scribe> ACTION: Steven to attend to TPAC scheduling [recorded in http://www.w3.org/2008/03/12-xhtml-minutes.html#action01]
SM: nothing from TAG -- pinged them again -- need to procede WITH their input, but if WG wants to can
SP: can go to LC saying dealt with all issues we know of, if TAG doesn't reply, then proceed; if they don't may have to cycle back to LC
SM: what we want to avoid -- outstanding TAG response 2 weeks late -- keep pinging but no response
SP: other people in TAG could be contacted (whisper in ear) -- can approach henry from another angle...
<ShaneM> to be clear - lots of responses. but the summary of the responses is "hang on - we are working on it"
SP: suggestions?
RM: dragging for quite a while -- hanging for more than a month -- give Ben a deadline -- need to procede by end of next week
SP: henry's online now -- should i ping him
SM: i did 10 minutes ago
SP: specifically request earlier in week or put as preference?
SM: don't want to impose, just preference
RM: continue to prod henry
SP: have his attention now
RM: other actions -- upload final CURIEs draft; Shane to add a non-normative schema implementation to CURIE spec
SM: wanted to ask WG what are we trying to solve there
RM: worth talking about
... want schema as well as DTD --
SM: i had added data type to master data types
definition module as part of M12n
... separate data type namespace -- made sense to put there b/c
modularization
... why have 2 modules
... didn't yet check in -- wanted to talk to group; not a "good, tight"
definition -- placeholder right now -- complaints about "tightness" of schema
defs, not sure how to make "tight"
... CURIEs just strings that might have a colon in them -- thoughts?
MB: checking my definition from work ages
ago...
... can't locate, but did have a regular expression for this
RM: how does it compare wity QName built-in type?
MB: not built-in; 2) after colon doesn't have to be NC name
SP: essentially anything that can be in a
URI
... struggled to find more structure than that, but series of any chars that
can be in URI -- don't know what's in prefix
<ShaneM> right now I have this:
<ShaneM> <!-- CURIE -->
<ShaneM> <xs:simpleType name="CURIE">
<ShaneM> <xs:restriction base="xs:string"/>
<ShaneM> </xs:simpleType>
<ShaneM> <xs:simpleType name="CURIEs">
<ShaneM> <xs:list itemType="xh11d:CURIE"/>
<ShaneM> </xs:simpleType>
RM: CURIEs or list of CURIEs or what?
SM: pasted into IRC what i have -- that's what's there right now -- safe CURIE production, might put together 2 -- URI or CURIE data type
MB: right
SM: don't think anything sensible can say in regular expression that anyone would have confidence in
MB: way i did this was copy the productions for
a relative URI -- even though difficult and allows anything, more correct to
have that level of indirection, rather than us saying "only a string"
... do have regular expression that does this, have to locate
SM: why regular expression -- just a URI
SP: lexical space
SM: i believe lexical space same
RM: no, prefix, colon, URI
SM: don't need prefix
RM: optional prefix, colon, URI then
SP: URI not a CURIE and a CURIE that isn't URI
RM: conceptually different -- mistake to call
same thing
... any string is what i'm getting at
SM: any URI?
RM: say "relative URI"
... specifically refer to sub-part of URI expression -- iFragment
SM: right
RM: can be anything but that is conceptual rather than string literal -- can't use in schemas (x followed by y making y fundamental data type -- just doesn't exist)
MB: assign a name
RM: create regular expression that maps to this
MB: URI doesn't contain regular expression
<ShaneM> http://www.w3.org/2001/XMLSchema
SM: have to relook at XML data type schema but son't think regular expression
MB: differences: 1) not built-in type -- built-in types don't need regular expression, but we do
RM: how validate in schema? validate that URI is a URI
MB: discussion just morphed into "do we need to do this" -- yes, can write regular expression, but so broad it isn't completely useful -- making it any string not the correct solution -- need a regular expression, but need to be aware that anything can be in regular expression
RM: restraint for CURIE as whole?
MB: done in past -- have to locate
SM: if MB has regular expression that works, ok
with string -- M12n data types makes most sense
... M12n doesn't actually use it today
... but we have produced modules that use it so M12n is right place (RDFa and
Role need it)
RM: document that schema in m12n spec
SM: yes
<scribe> ACTION: MarkBirbeck - produce regular expression and URI/CURIE type [recorded in http://www.w3.org/2008/03/12-xhtml-minutes.html#action02]
SM: RDFa has use for that -- use case TAG most
afraid of
... mixing CURIEs and URIs -- that's what they are nervous about
<scribe> ACTION: ShaneM - document that "such types exist" in m12n in document; MarkB will produce actual definitions [recorded in http://www.w3.org/2008/03/12-xhtml-minutes.html#action03]
RM: anything else about CURIEs?
... next wednesday, decide if want to procede even if don't hear back from
TAG
SP: 2 other agenda items -- 1) june face 2 face -- created blank face to face page on wiki
SM: contacted hotels and will get info back to WG this week
<Roland> http://www.w3.org/MarkUp/xhtml2/wiki/2008-06-Minneapolis-FtF
SP: 2) getting the message out -- year or 2
ago, some people complained that no one supporting XHTML and publishing
denegrating statistics and so is a "failure"
... this weekend, looked at top web sites -- went to alexa and found list of
100 most popular websites (on basis of bandwith) -- examined to see if use
XHTML -- numbers far greater than that being claimed by XHTML opponents
RM: part of confusion may be mime-type
SP: healthy proportion that did it completely right; also a reasonable proportion that did same for HTML4; then a number of pages where clearly XHTML (had all the attributes had quotes, closing slash and lower case) but not documented correctly -- remainder tag souop
RM: BBC web site -- XHTML Strict
SP: should do analysis properly -- look at top
100, create article about strength of support 30% use correctly, 20-25% don't
declare propertly
... where to publish
RM: can start putting into wiki
SP: if do something like that, good to fanfare it
RM: just make available via wiki as first step
MB: excellent to get numbers, but since XHTML spec says should use XHTML mime-type/DTD, opponents say that is not XHTML
RM: bring all together -- with looser mime-type, we have a lot more XHTML than people are admitting; have to express that spec not so uptight or restrictive
MB: initial claim - XHTML broken, let's get
more mileage out of HTML -- now XML serialization as XHTML5 -- going against
initial claims, but in an end-around manner, letting fox into chicken
house
... no extensibility, but one huge document that has to cater to not being
XHTML; showing XHTML has traction would be useful
RM: need to go more in direction of XHTML being rendered by HTML renderers; nothing in spec about it - XHTML renderers rare, given IE's share, should look at loosening mime type, but also means of XHTML in today's UAs being rendered by HTML engines
SP: thought agreed to change mime-type
SM: discussed it but received massive
objections -- haven't made change publicly
... changed in XHTML 1.1 working draft and people went nuts
RM: also require an XML parser -- inhibits uptake of XHTML
SM: don't require XML parser
RM: XML application, though
SM: big arguements get from WHAT WG people is that XHTML doesn't work in wild -- if but embedded CDATA sections that redo DTD doesn't work in browserX
RM: if XML namespace used, won't work in current browser
<ShaneM> XHTML 1.0 Appendix A was intended to give guidance
RM: confusion -- people don't know where W3C is going on this -- needs more thought and energy
SP: agree
MB: worthwile and ongoing work -- XHTML, XML and XHTML2 -- modularized and extensible benefits
<Steven_> "XHTML Documents which follow the guidelines set forth in Appendix C, "HTML Compatibility Guidelines" may be labeled with the Internet Media Type "text/html" [RFC2854], as they are compatible with most HTML browsers."
SM: core of misunderstanding is Appendix C from XHTML 1.0 -- did a bad job there -- if going to fix, start there -- add similar appendix with XHTML 1.1 as non-normative appendix
[general agreement[
<Steven_> "Those documents, and any other document conforming to this specification, may also be labeled with the Internet Media Type "application/xhtml+xml" as defined in [RFC3236]. "
Yam; last week japanese W3C member meetings; 1 main topic HTML5 -- surprised that more than 75% of HTML5 presentation that IE8 beta2 will support some HTML5 features; some HTML5 features supported by IE8 - no committment to support whole thing in FF;
RM: IE8 beta will improve things generally
SP: do we think analysis worthwile?
YES
SP: put it somewhere?
YES
RM: put into wiki first and then sort and filter
SP: 30 of top 100 (majority) use XHTML
<ShaneM> ACTION: Shane to migrate Appendix C to the draft XHTML 1.1 second edition [recorded in http://www.w3.org/2008/03/12-xhtml-minutes.html#action04]
RM: what happens in diff geographic areas and
diff languages -- browsers differ geogrphically -- content could be used in
diff ways for markets -- know how to ID mobile
... agenda review, redux -- M12n?
SP: collecting dates from w3c people --
communications people don't join call anymore -- hope by end of week will
have range of dates from steve bratt -- then can select suitable date
... Shane, do we have an editor's draft that disposes all of steve b's
comments
SM: don't remember
SP: included pointers in request
SM: have to investigate
SP: please do, then can schedule call
SP: waiting on me -- trying to finish implementation reports this week
RM: skip XHTML 1.1 SE status for now
... actions against RDFa syntax
Steven to reply to TAG that we disagree. (in process)
SM: added note that there is a risk that humans might perceivea CURIE as a URI
RM: Role Module - running bit behind CURIEs --
have the implementation convention and Role Attribute module changes from
WAI-PF -- finally received
... any difficulties
<Steven> But can you spell it?
complimentary
RM: all wai roles into xhtml namespace so don't have to be prefix qualified
<Steven> That was my point
SM: can put PF's in our namespace -- left room for it in Vocabulary document
<Steven> that is not how one spells it
<Steven> I think it is a bad choice for that reason
RM: enough to procede?
<markbirbeck> +1 to Steven
http://lists.w3.org/Archives/Public/public-xhtml2/2008Mar/0014.html
<Steven> I don't like complementary
<scribe> ACTION: ShaneM - incorporate WAI-PF verbiage to Role vocabulary [recorded in http://www.w3.org/2008/03/12-xhtml-minutes.html#action05]
<Steven> 92%+96%! No wonder it was so loud!
<Roland> yes please
no problem
GJR: by the way, PF is giving XHTML2 / Shane the final say on "complimentary" or "supplemental" or something better than "secondary" (deemed pejorative)
This is scribe.perl Revision: 1.133 of Date: 2008/01/18 18:48:51 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/SP: contacted/SM: contacted/ Succeeded: s/Appendix A/Appendix C/ Succeeded: s/MB: what happens in diff geographic/??: what happens in diff geographic/ Succeeded: s/??/RM/ Found Scribe: Gregory_Rosmaita Found ScribeNick: oedipus Default Present: Roland, Gregory_Rosmaita, ShaneM, Steven, yamx, markbirbeck Present: Roland Gregory_Rosmaita ShaneM Steven yamx markbirbeck CSB_on_IRC Regrets: tina allessio Agenda: http://lists.w3.org/Archives/Public/public-xhtml2/2008Mar/0015.html Got date from IRC log name: 12 Mar 2008 Guessing minutes URL: http://www.w3.org/2008/03/12-xhtml-minutes.html People with action items: - incorporate markbirbeck shane shanem steven verbiage wai-pf WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]