See also: IRC log
<Schnitz> I still have a bit of my cold left, I might be a bit quiet therefore
<John_Boyer> Scribe: Charlie
Next F2F, June 13-15
please fill out reg form
<Roger> me too
Steven: tweaks done
... XForms title is autogenerated-can't be changed
<scribe> ACTION: Steven to change template [recorded in http://www.w3.org/2007/05/30-forms-minutes.html#action01]
please fill out questionnaire as network access and building access is driven off that too
John: questionnaire on tech
plenary, assume we're shooting for end of the week
... so we have thurs afternoon, fri, sat
Steven: AC meets for 1/2 day, thurs AM so not scheduling WGs for then
so we spill over to Sat AM
and hence need to confirm that WG participants sign up to do this
or we decide to not use sat AM
John: we can work out later, i'll
say we're not flexible on days
... made estimate of 6 people attending wed tech plenary
out of about 17
Charlie: i will attend
<Steven> I will
<unl> i won't attend
<John_Boyer> i will
Steven: good idea to attend so groups can meet each other
topics of interest cross-groups
backplane could be discussed
Charlie: if we get IPR resolved...
<Steven> Looks like there are 4 AC reps on the WG
<Steven> .... Sebastian, Raman, Erik, Kenneth
John: doc is about ready
one outstanding issue is regarding patent policy
pub rules checker on 2nd edition, proposed rec was under 2002 policy
pub rules checker was failing at that time given we're now under 2004 policy
we updated to that, passed pub rules, published the doc
was not right process, pub rules changed by june to understand diff policies
status for 3rd edition has to state relationship to previous version
pub rules has "1.0 still under 2002 policy, governed by transition rules as stated in 2004 policy"
just want to be clear where we are
Steven: 2nd ed claimed 2004, but was actually 2002
3rd ed is still under 2002?
1.0 2nd edition is wrong to say it's under 2004
Steven: Ian issued call for exclusions...indicates falls under 2004 policy
John: Ian indicated we should go under 2002 policy with transition procedures
and I indicated to him we would do this
I clarified to Ian that we had actually published 2nd edition...waiting for confirmation his recommendation is still correct
Steven: let me check now with him on IRC
John: do we need quest. or does it go to HCG
Steven: raised with HCG...need to check their minutes
John: they did discuss, but
waiting for us...
... we'll bring it up in next HCG meeting
<John_Boyer> Question about support for xsi:type
John: is it valid to use xsi:type when there is no schema?
problem with using xsi:type and referring to internal schema is not valid unless server also loads internal schema
so interesting question
just looking at processors, seems like no problem
but on submission those declarations not available
Mark: what are you suggesting?
John: not suggesting either way...
Mark: we discussed a lot of this during xforms basic
I think we should support these types
want the ability to use types without schema for convenience
John: we're not talking about type MIP
Erik: regarding submitted data problem, this is fine with us
no requirement that document needs to be validated with same schema on submission
could have MIP making xsi:types not relevant no submission
stripping those attributes on submisssion
John: anyone believe xsi:type should only be applied if form author as attached a schema?
<Steven> I think the user would expect xsi:type to work
Erik: we did specify xsi:type had semantics of schema, but we didn't specify how this should behave
<John_Boyer> me too
Mark: we refer to schema data types
can be used independently of full schema
Erik: don't have strong
opinion...MIP could do the same thing
... was more concerned that the spec was unclear
John: section 5.2, generally section 5, contains language suggesting that processing of instances is informed by schema of xforms
so there's an implict schema available to the parser
5.2 lists xforms data types
as "built-in xml schema datatypes"
Mark: in basic you can use these independently of full schema, so why inhibit use here?
Erik: clear we want to allow in MIP, but in parsing instance it's a bit different
Mark: but why not for consistency allow both?
Erik: could imagine building an
instance using xsi:type but not having ability to disable those
attributes for validation etc
... but using the bind it's clear whether to valid the node or not
John: this implies xsi:type is not preferred
Mark: xforms full talks about using xml schema, this is available to full processor, basic processors might do something different
Erik: if we want to make xforms schema-language agnostic in the future
Mark: that's future work
Erik: i do think the spec is not very clear on this
need to fix the language to specify that xsi:type attributes are processed even in absence of schema
John: ok...action item???
Erik: do you think it's clear enough?
John: i think you're asking for a statement in section 5 on datatypes
connecting their usage to xsi:type?
Erik: not sure about specific
section, it's mentioned several times
... just need to clarify that processor must deal with xsi:type on instance elements
John: not clear to me where this change should be made...where it's unclear
issue needs some more work
Erik, could you look at the spec and see where to make this change?
<scribe> ACTION: Erik to recommend where the spec should be clarified about xsi:type handling [recorded in http://www.w3.org/2007/05/30-forms-minutes.html#action02]
John: don't mind dropping the word "special"
on the update behavior
since we're just describing normal deferred update behavior, not an exception to it
which is well defined
<scribe> ACTION: John to remove "special" on deferred update behavior [recorded in http://www.w3.org/2007/05/30-forms-minutes.html#action03]
<John_Boyer> Need rigorous definition of "Acceptable XPath Expression"
John: issue we have not defined what's acceptable as an xpath expression
in binding expressions
Erik: also last call comment asking for definition of acceptable xpath expression
"acceptable" is not a good word...we don't say what happens if not acceptable
in bind for example we say a rebuild is required
but in ui binding we don't seem to do the same thing
confusing to me what acceptable means and its consequences
<Steven> My action "10 01Action: Steven to change template" has been done
move away from that term and talk about dynamic bindings and when they can be used
Erik: what we're trying to say is complicated, but we understand how it's supposed to work...wording is just not intuitive
John: sense an action item...
<scribe> ACTION: Erik to propose alternate wording for "acceptable" xpath expression [recorded in http://www.w3.org/2007/05/30-forms-minutes.html#action04]
Steven: [ot] who has responsibility for adding actions to DB? need this for last call?
John: [ot] for Last call in particular?
John: i'd like to clarify this process
Steven: issues we agree to handle, have to forward to email address of the db
when we deal with an issue, need to update DB with solution
then reply to the person asking if they're ok with the decision
John: what did we do for 1.0
Sebastian: we did this for 1.0
Steven: i think the easy way is for some single person to take this on
Sebastian: agree, would be best for someone with interest in the system
Steven: we're using Shane's system so it's easy
issues just need to be forwarded there, with later update after decision
John: on prior telecon we started that process, made progress up to march 14 on the telecon
Steven: and updated db at same time
John: hoped we could continue that process, with someone to handle db updates...volunteers???
Uli: I will take the job
Steven: i'll fill you in on process offline
John: issue is if you replace an instance we don't describe fully in the spec that ui controls might receive value-changed and other MIP events
has led to discussion about other points in the lifecycle where we might have/want events
could be a problem since alerts etc might get fired at initialization
so would suggest we clarify behavior specifically of refresh
after instance replacement, get rrrr sequence
Erik: we have 2 different problems, initialization and instance replacement
can read the spec on refresh and think it works with replacement
John: agree that refresh language is deficient in that it doesn't clarify this
Erik: i was only raising issue of replacement
<markbirbeck> many apologies, but I have to go.
we define refresh based on instance node, with complete replacement it's difficult to define behavior in terms of changes to existing nodes
<Roger> thx & bye
This is scribe.perl Revision: 1.128 of Date: 2007/02/23 21:38:13 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/edition/2nd edition/ Succeeded: s/exlusions/exclusions/ Succeeded: s/belive/believe/ Succeeded: s/imaging/imagine/ Found embedded ScribeOptions: -implicitContinuations *** RESTARTING DUE TO EMBEDDED OPTIONS *** Found Scribe: Charlie Inferring ScribeNick: Charlie Default Present: Charlie, John_Boyer, Schnitz, +1.812.535.aaaa, Blake, +1.919.434.aabb, ebruchez, Steven, roger, Mark_Birbeck, wellsk, David_Landwehr, unl Present: Charlie John_Boyer Schnitz +1.812.535.aaaa Blake +1.919.434.aabb ebruchez Steven roger Mark_Birbeck wellsk David_Landwehr unl Regrets: Susan Nick Leigh Joern Agenda: http://lists.w3.org/Archives/Public/public-forms/2007May/0084.html Got date from IRC log name: 30 May 2007 Guessing minutes URL: http://www.w3.org/2007/05/30-forms-minutes.html People with action items: erik john steven[End of scribe.perl diagnostic output]