See also: IRC log
DKA: Proposal is to extend the charter for 6 mos. Is that enough time?
Jo: Should we make it 3 mos.? Sense of urgency?
DKA: We're not adding any
deliverables, not changing the charter except for extending
... There have been some suggestions that we add deliverables.
Jo: I think the group should close.
<Jo> [head for the "grey havens"]
DKA: What do we need to do to extend the charter?
Francois: How long is enough? Should we make it 3, 4 or 6?
Jo: What we need to do is
shepherd BPA through CR.
... How many BPs are there?
... Does an implementation need to do all of them?
... just individual ones.
Alan: Sometimes Web sites do it without knowing it.
Francois: You can contact the Web site and ask them about it.
Jo: I wasn't planning on crawling the web. I think we have enough members still around with Web sites that we can ask them.
Francois: We'll look around in the group to see if anyone has implementations of the BPs.
Jo: 6 mos. is sensible.
... Let's do a timetable.
DKA: Should I create the timetable while we are talking?
DKA: We might want to do a
schedule of all of the calls we want to have.
... First of all, do we want to have a call on Dec. 18th to see if should move any documents along?
Francois: I don't think the CTG will be ready by then.
Jo: We also have MWA.
Francois: I need to contact Adam. We need to contact all of the comments of the LC.
Jo: We need to make sure all of the changes get into the new draft.
DKA: Does that mean we need to schedule a special call to see if we do can do something before the moratorium.
DKA: Should we do a call on the 17th or 18th?
Francois: Probably the 17th since other documents will try to slip in before the deadline.
DKA: Just want to know if there is any reason to have a meeting before the end of the year.
Jo: probably not.
Francois: There are some changes we could make to CT before the end of the year.
DKA: What about the 22nd?
Jo: There are some other meetings on that day.
DKA: Shall we have a call on the 21st? We could go over any other changes to CT.
Sean: I won't be able to make the meeting on the 21st.
DKA: Even if we are doing something on the 22nd, I think we could spare some time.
Jo: We could the momo meeting in the morning and do the BPWG meeting at the regular time on the 22nd.
DKA: No call on the 15th.
... What is the timetable for Jan? Should we have a call at the regular time on Jan. 5?
... If so what should the goal of the call be? Advance CT?
Francois: At that time we'll have the comments from the reviews of the MWABP.
DKA: Could request that MWABP is
transitioned to CR.
... For CT we'll possibly have a new draft that we can resolve to publish for a new last call.
[Working on timetable in Google Docs on the screen in the room]
DKA: My view is that we need to
be a reduced schedule for 2010. I don't expect any more F2F
... Are there any opportunities for us to have a F2F where we'll all be around someplace anyway?
... I was thinking of the WWW conference in Bali--Apr. 26-30.
... For the 5th of Jan...
Jo: Doing what you said, would require the documents be done and people had time to review them which might work for MWABP, but not for CT.
DKA: OK, I changed it to the 12th.
Jo: We can do MWABP on the 5th.
Francois: I think that is
possible; we'll need to check with Adam this week.
... There were no strong comments in the last batch.
Jo: We should attempt to do it on the 5th.
DKA: It depends on our call schedule. We should have a reduced call schedule. Shorter calls and not weekly (biweekly and then even less frequently)
Jo: I think shorter calls on a weekly basis would be better to keep attendance up.
DKA: Don't agree. I think
biweekly would be better for me.
... Frequency of calls is what affects my workload.
Jo: I think we should do weekly until CR.
Francois: Biweekly didn't work
for us internally. I think weekly works better.
... I think it is a good idea to have an agenda.
DKA: Have themes for first two
... Let's reduce the time to an hour.
<DKA> PROPOSED RESOLUTION: In 2010 we will reduce the call time to an hour but retain weekly calls for now until both CT and MWABP go to CR.
RESOLUTION: In 2010 we will reduce the call time to an hour but retain weekly calls for now until both CT and MWABP go to CR.
DKA: What else can we add to the timeline?
Jo: We'll get CT into LC on the 12th. We need to give people 4 weeks to review.
Francois: If we decide to publish
the LC on Jan 12th, it will ship on the 14th.
... We could give them to the 9th of Feb to review.
Jo: We review LC comments on the
... We may get some more substantial last call comments.
... on the 16th of Feb. we would reply to the LC comments and then go to CR.
DKA: On Feb. 16th reduce the call schedule.
Jo: We may not need so many calls while CT is in LC.
Francois: We have to reply to the LC comments and work on the test suite.
Jo: When does the test suite have to be done?
Francois: When the CR period is
... We need to validate implementations with the test suite.
Jo: Is there anything I can do to help?
Francois: You can contribute tests.
Jo: What language will be used for the tests?
Francois: I thought maybe we could use Java, but no strong preference.
DKA: Which doc are we talking about.
Francois: CT. There won't be a test suite for MWABP.
DKA: Is there any thought for additional work on the Checker for MWABP?
... It will be worth it in Jan. to keep the group going to review the tests.
... We should write out the tests in pseudo code for people to review.
Francois: We already have the ICS
that has all of the normative statements. I suggest that we use
that to create another document that would define a series of
tests in some human language first, then we try to implement
these in code.
... I'll create the framework to get this started.
Jo: Logically we can't do this until CT is in CR, but in practice it could largely be done before CR.
Francois: The tests could be done while CT is in LC and may generate some more comments.
Jo: In practice a couple of people will be doing it.
Francois: The first step doesn't
require any coding.
... doesn't need to be done on a teleconference; could be done on the mailing list.
Jo: We'll need to find some
additional people besides Francois and Sean to get this work
... It's not an insignificant piece of work. It will require possibly several tests per normative statement with client and server tests.
... And it will all need documenting.
Francois: Why all the documentation.
Jo: Needs to be done in real conditions.
Francois: Isn't this going to be kind of impossible.
Sean: Are there any examples we can look at?
Francois: There are some other documents that have test suites.
Jo: My point is that a laptop using a phone as a modem would rarely be a valid test.
DKA: If you configured your laptop to use the phone as a modem going through a WAP APN, it would make work OK.
Jo: We should establish the viability of ever running the tests.
DKA: Why do these tests need to be done in the wild? Couldn't these tests be done in a controlled environment?
Francois: We don't want to keep people from claiming conformance when they've tested on private network.
Jo: No, the deployment needs to be publicly available on the non-over the air network.
Francois: We're just talking about the connection.
Jo: It is not completely obvious
to me how you would execute these tests and get a meaningful
... Maybe someone could write something up on how this works.
Francois: We don't say that
access needs to be free.
... Section 5 doesn't say that you need to make the deployment available on a mobile network.
... We could build a widget that does all of the tests?
Jo: It would be kind of cool if the test suite was available as a widget for people to download to their phones.
Francois: I'm not sure we can change the UA header using an XHR request.
Jo: It would be a good objective to do as many tests as possible with a widget.
<DKA> PROPOSED RESOLUTION: Regarding the CT test suite work, we will build a cool widget.
Francois: We can try to do as many tests as possible with a client side widget.
Jo: We could deploy using
... we need to a bit more thinking of how we are going to run the tests and who is going to run the tests.
Francois: We need to try to have a client that uses just regular web browsing techniques.
Jo: It would be nice to have
software the automatically runs through all of the tests.
... we can define the tests without worrying about the deployment, but when we write the code we'll need to think about it up front.
<francois> ACTION: francois to think carefully about how tests in the CT test suite could be run from mobile phones and report (Web App? Widget? J2ME?) [recorded in http://www.w3.org/2009/12/11-bpwg-minutes.html#action01]
Francois: the first part is about listing the tests that we want applied.
<trackbot> Created ACTION-1030 - Think carefully about how tests in the CT test suite could be run from mobile phones and report (Web App? Widget? J2ME?) [on François Daoust - due 2009-12-18].
Alan: Presumably phone manufacturers have ways to test apps on phones.
DKA: We have a test network.
[Jo drawing a diagram on the whiteboard]
Jo: Maybe the focus should be on
creating a Java client. It could be used to generate a J2ME
client and generating a GWT app.
... just need to follow some rules to make sure that J2ME and GWT apps can ge generated from J2SE code.
Sean: This looks like it may take several months to do.
Francois: How long did it take to do the mobileOK Checker test suite?
Jo: Took a couple of years.
<francois> ACTION: francois to create the initial document that is to contain human-readable descriptions of the tests to implement. [recorded in http://www.w3.org/2009/12/11-bpwg-minutes.html#action02]
<trackbot> Created ACTION-1031 - Create the initial document that is to contain human-readable descriptions of the tests to implement. [on François Daoust - due 2009-12-18].
DKA: Resource issue here, not
necessarily people currently in the group.
... We may be able to get some VF people to work on this.
Jo: What about Jeff? Maybe he has some students can work on it.
DKA: What are the skills
... HTTP, Java, etc.
Jo: A student would do this would learn a lot about HTTP. It would be work to debug their code, however.
DKA: If you had a number of people collaborating, it could be quite effective.
Jo: Need to be a project plan.
DKA: The risk is that if we don't get it done, the document could stay in CR a long time.
Sean: Do you need actual code, or is pseudo code OK?
DKA: Need actual code.
... We need a test sute, no document, etc.
Jo: We need a separate document
that documents the test sute.
... Could test the document with pseudo code for correctness, then test the actual test suite for correctness using the document.
DKA: We need to get additional people to help out with writing the tests.
Jo: We need to take the rest of this discussion offline. Has possible charter extension consequences.
DKA: We may be making too much
out of this. If we approach it in a systematic way, then we
should be able to get it done in a reasonable amount of
... We need a better idea of what resources we have to bear.
... How it impacts our schedule, after MWC, we will only have this test suite to work on until the end of June.
Jo: The object, then would be to exit CR when the test suite is done.
Francois: When we exit CR, the time between PR and Rec should be short.
Jo: We aim to transistion between
PR and Rec in June, which means that we should go to PR at the
beginning of June for a 3 week transition.
... We have 4 months to create the test suite then.
... which is pushing it.
... Some of what is Mobile OK Checker could be reused, right?
Francois: Some of it, but not the
hard stuff. What makes it hard is that we need to two sides.
None of the tests are hard to write, but it will take a while
to do the work.
... I could check to see if it will be OK just to provide the tests in written form. It might be seen as kind of a failure,but we could say that we have resource constrants.
... For some tests you can't write tests for anyway; for example the requirement of a transparant proxy would require the entire HTTP protocol.
... It will need to be lightweight test suite; it won't be like the MobileOK checker which really has to be detailed since it is a checker.
... I can check to see if the written form would be sufficient.
... Our intention is to cover all of the normative guidelines but not all of the underlying requirements since that is impossible.
... For example, would be hard to test is a proxy is really transparent.
DKA: For the schedule, should we finish by the beginning of June?
<francois> ACTION: francois to investigate on the possibility to restrict ourselves to a pseudo-code written test suite document for CT. [recorded in http://www.w3.org/2009/12/11-bpwg-minutes.html#action03]
<trackbot> Created ACTION-1032 - Investigate on the possibility to restrict ourselves to a pseudo-code written test suite document for CT. [on François Daoust - due 2009-12-18].
Jo: I think charter extension any longer than 6 mos is pushing our luck.
DKA: We don't want to keep
extending the charter.
... That's why I want to get additional people working on the tests so we can finish up the group.
Francois: Most of the tests will
require some human intervention and won't be fully
... We have several guidelines that require human intervention.
DKA: We could say we are going to reduce the call schedule in March.
Jo: Let's wait on that.
DKA: By Jan 19th we need to have an idea of what we need to do for the test suite and resources we have.
Francois: What I can do by then is give an idea what tests can be code and which ones will need human tests.
DKA: Do we still need to write a human readable test plan?
<francois> ACTION: francois to highlight normative statements that can be automatically tested, and those that will be hard or impossible to test on an automated basis by 2010-01-19 [recorded in http://www.w3.org/2009/12/11-bpwg-minutes.html#action04]
<trackbot> Created ACTION-1033 - Highlight normative statements that can be automatically tested, and those that will be hard or impossible to test on an automated basis by 2010-01-19 [on François Daoust - due 2009-12-18].
<francois> ACTION-1033 due 2010-01-19
<trackbot> ACTION-1033 Highlight normative statements that can be automatically tested, and those that will be hard or impossible to test on an automated basis by 2010-01-19 due date now 2010-01-19
Francois: Final thing is to request charter extension.
DKA: We expect the major work of the extension period to be the test suite for CT.
<Jo> PROPOSED RESOLUTION: BPWG requests charter extension to 20 June 2010 to see MWABP through to rec and to see CT Guidelines similarly through to rec with and appropriate level of test suite
<Jo> PROPOSED RESOLUTION: BPWG requests charter extension to 30 June 3010 to see MWABP through to rec and to see CT Guidelines similarly through to rec with and appropriate level of test suite
<francois> [We will also gather implementation reports on Mobile Web Application Best Practices]
<Jo> PROPOSED RESOLUTION: BPWG requests charter extension to 30 June 2010 to see MWABP through to rec and to see CT Guidelines similarly through to rec with and appropriate level of test suite
<Jo> +1 reluctantly
RESOLUTION: BPWG requests charter extension to 30 June 2010 to see MWABP through to rec and to see CT Guidelines similarly through to rec with and appropriate level of test suite
DKA: Can close ACTION-1026
Francois: Can close
... Already have an action to ping internally for CT test resources in VF
DKA: For ACTION-987, draft a blog
post on mobileOK Scheme, whenever I think about this I think we
need awareness of mobileOK and I don't think one blog post is
going to do it.
... We need to engage W3C to help promote awareness of mobileOK.
Francois: I'm not what else I can
do. I talk about mobileOK at every conference I go to. We try
to contact companies that make tools for mobile web
... We already did a press release.
Jo: We talked about an event.
Francois: Everything right now is about HTML 5, etc. Flashy stuff.
DKA: mobileOK still has an important role to play and there communities that aren't aware of it.
<Jo> dka: "one swallow does not a summer make"
<Jo> dka: "let a thousand flowers bloom"
DKA: So we need some sort of event to get people more aware of mobileOK. One blog post is not going to do it; we need some other event to go along with it. Maybe some sort of Twitter campaign.
Phil: It is difficult to inspire people on mobileOK. It's not a sexy thing.
DKA: We had a teen panel; teens
don't necessarily have fancy phones.
... This lines up with Tommy Ahonen who says that not everyone has an iPhone.
Jo: I think the most appropriate channels for pushing mobileOK is the developing nations initiative and academia.
Francois: What do you mean by academia?
Jo: Computer science
... Don't think it is worth pushing the mobileOK message which has to do with validity and so on when the HTML message is the opposite of that.
Alan: Gov't Web sites should be obliged to be mobileOK.
Jo: I think a general awareness campaign would be doomed to failure.
Alan: We could work with the European Commision E-Inclusion Unit.
Francois: I often go to conferences of the European Commision and present mobileOK.
DKA: Maybe this is something you can investigate with Philip to get more awareness with the European Commision.
<francois> ACTION: francois to investigate about opportunities to promote mobileOK with the European Commission [recorded in http://www.w3.org/2009/12/11-bpwg-minutes.html#action05]
<trackbot> Created ACTION-1034 - Investigate about opportunities to promote mobileOK with the European Commission [on François Daoust - due 2009-12-18].
DKA: Let me develop a plan on a mobileOK awareness campaign.
<francois> ACTION: dan to create a plan to raise awareness on mobileOK through some campaign [recorded in http://www.w3.org/2009/12/11-bpwg-minutes.html#action06]
<trackbot> Created ACTION-1035 - Create a plan to raise awareness on mobileOK through some campaign [on Daniel Appelquist - due 2009-12-18].
<francois> close ACTION-987
<trackbot> ACTION-987 Draft a blog post on mobileOK Scheme closed
DKA: Any other actions we can close?
Francois: Can close
... Need to keep Bruce Lawson's action open for new because will need it.
DKA: Do we want to raise an issue around the test suite.
<Jo> PROPOSED RESOLUTION: BPWG thanks Vodafone for its kind hospitality in hosting this F2F
RESOLUTION: BPWG thanks Vodafone for its kind hospitality in hosting this F2F