See also: IRC log
<francois> F2F agenda
jo: Idea is to spend Wednesday on
MWABP, Thursday on CT, tidying up on Friday
... including tidying up remaining mobile accessibility issues,
checker library, etc.
yeliz: will be there on Friday
jo: we have mobileOK scheme
francois: not heard from Rigo, hope to have something by the F2F
jo: need to ping Korean folks for a written update
<EdC> three questions about the agenda: (1) are the times indicated local UK time? (2) any detailed schedule? (3) numbers for teleconference?
bryan: I shan't be there in person, conf call bridge would be good.
<jo> ACTION: JO to talk to Adam about getting a conf bridge set up for f2f [recorded in http://www.w3.org/2009/03/17-bpwg-minutes.html#action01]
<trackbot> Created ACTION-915 - Talk to Adam about getting a conf bridge set up for f2f [on Jo Rabin - due 2009-03-24].
francois: will ping Rigo
jo: times of the agenda are in GMT
jo: Adam's issued 2 new drafts for MWABP
francois: the mobileOK checker libraries only apply to HTTP/HTTPS URI schemes. There could be a useful use case where you want to run tests on a file. Most web content is local before it's published.
<yeliz> sorry about the echo, it seems like there is something wrong with my connection today :(
francois: you might want to check
it before you publish it. Some of the tests do not mean
anything outside of HTTP(S), some do and are useful - e.g. page
size
... Yeliz can probably talk about it, I wanted to update the
checker library and add the possibility to check files. We've
discussed a little on the checker mailing lists and have some
ideas for how to add support for file:// URIs. It would mean
that the library contains code that is not defined in the
standard, in particular an additional test outcome
("CANNOTTELL")
... but it's not mobileOK if it's not in the standard, as Jo
has pointed out. The Java library is supposed to be a reference
implementation of the standard, so I don't know if we can
extend it with something outside of the standard.
... We wanted to get the WGs opinion. Do we need to duplicate
the code and work on a separate version of the library? Or can
we incorporate the changes and add file:// scheme support to
the library?
... We might want to issue another document explaining how to
test file://
<EdC> Can you parameterize the library with a switch for "standard mobileOK" and "development mode"? Via a configuration file, for instance?
francois: I think it's useful for us to go ahead with this. Abel proposed to write the WG Note. I'm not sure we need to do this.
jo: I agree
<Zakim> Bryan, you wanted to why not run a local server (http://localhost) instead - this is easy?
bryan: a consistent scheme is a good idea. All you can do with file:// is check a static page, which isn't very common in terms of real live applications
jo: the point about mobileOK is that it tests the operation of your server when serving content - i.e. in the real world.
bryan: that's my point
yeliz: we want to combine the
mobileOK library with another for validating documents. It
would be good to use some of the mobileOK tests with local
files. This could be used for other people (e.g. designers)
wanting to check documents before they upload them. e.g. the
HTML validator lets you upload and validate a doc.
... you can't do all the tests, but a CANNOTTELL would
accommodate this.
<yeliz> sorry about the echo:(
<EdC> question: is the file:// scheme used in some Web applications to access the local storage?
<jo> jo: wondering if there is a way of leaving the reference implementation intact and dealing with file: scheme by subclassing?
<Bryan> fyi I have to go on IRC only for the next hour - will be back asap
francois: I had the same idea - we could do this without altering the ref. implementation. There are a couple of things we can do, but it can't be done completely by subclassing.
<EdC> question: is the file:// scheme used in some Web applications to access the local storage? If yes, shouldn't the scheme be dealt with in the test harness?
francois: I would like to keep the ref. implementation clean
ed: if file:/// is used by web applications, the harness should handle it surely?
jo: mobileOK only tests http(s) URIs
<jo> ACTION: daoust to prepare some material for F2F identifying what changes would be needed to the mobileOK checker to allow subclassing for file: scheme handling [recorded in http://www.w3.org/2009/03/17-bpwg-minutes.html#action02]
<trackbot> Created ACTION-916 - Prepare some material for F2F identifying what changes would be needed to the mobileOK checker to allow subclassing for file: scheme handling [on François Daoust - due 2009-03-24].
jo: if Abel, Miguel and Nacho would like to write a note, we'd be happy for them to do it
francois: maybe we should make sure they're not working on something we might abandon first...
yeliz: what's involved in writing a note about this?
jo: the idea would be to write a WG Note (informative doc, not recommendation) pointing out the differences in the tests
<yeliz> yes, thanks
<yeliz> :)
jo: new version posted on Friday 13th. Francois has noted some typos - thankyou - and there are lots of outstanding issues, which I've yet to post.
<francois> CT announcement by Jo
jo: these will hopefully get out
today/soon
... unless anyone has anything to say about this now, we should
defer til next week
... (at the F2F). Sean, Rob? You may want to comment on HTTPS
link rewriting or link rewriting
rob: happy to keep that for the list
seanP: I need to look it over a bit
jo: hoping to resolve this issue (the main outstanding one, tho there are others we need to go back on, in particular Eduardo's point re changing/replacing headers)
ed: happy to deal with this and the other topics next week
<francois> [I note I'll have a bit to report on X-Device-headers next week based on a discussion with IETF]
francois: are you thinking we misunderstood what a same-document reference is?
jo: either I've misunderstood what they mean or it doesn't quite work
jo: the poll said no-one was
happy for it to advance to a WG Note
... so there's more work to do. Kai's not on the call today.
I've scheduled half a days editorial session on this on Friday
pm
... we need to check Kai will be there for that
... the November questionnaire hasn't been answered much
francois: we only have 1 day left to answer the poll
<Bryan> jo, post the questionairre link please
jo: wondering if more folks can attend the F2F if it's not in San Diego
<Bryan> I prefer San Diego!
jo: can we reopen this questionnaire with the additional answer "I could attend if it's elsewhere"
<EdC> "if it's elsewhere" is really a bit vague. Most answers will be "I do not know"...
jo: if we're to extend the charter we need another F2F. It'll either need to be there, or somewhere else.
<jo> ACTION: daoust to extend the TPAC Noc Questionnaire and add a question to assess whether the meeting would be better attended if it was held somewhere else [recorded in http://www.w3.org/2009/03/17-bpwg-minutes.html#action03]
<trackbot> Created ACTION-917 - Extend the TPAC Noc Questionnaire and add a question to assess whether the meeting would be better attended if it was held somewhere else [on François Daoust - due 2009-03-24].
jo: there won't be a call next Tuesday because of the F2F and we don't usually have calls post-F2F unless someone wants one. So no call on 31.03
<EdC> You mean 7th of April...
jo: We will be back to normal time for everyone on 7 April.
<Bryan> jo, can you post the questionaire link, I can't find it on the BPWG homepage
<jeffs> bye
<jeffs> quit
<Bryan> oh well
<jsmanrique> bye
<jo> bryan - francois will re-post to the list
<francois> and will update the WG home page to link to it, yes.