See also: IRC log
<jeffs> tnx
Francois: Making progress, with
more problems arising. Difficult to speed things up, trying to
get ready for FPWD. Editor's call next week, depending upon
progress.
... in the current draft there are many red boxes showing
issues that need resolution.
<francois> last CT draft
Francois: there is one big issue
but the little problems take time, e.g. detection of browsers
and non-browsers. These sort of issues need more discussions,
raise concerns/issues, etc.
... from 3.1.2 and beyond need work.On 4/3 we should be ready
to review and make a decision on public draft.
Dan: any other issues?
Francois: TF activity is good, no need to discuss other things here.
Dan: re the Mobile OK Pro TF,
update is requested
... but with Kai not present, we will have to wait.
<francois> ScribeNick: francois
<DKA> Scribe: Francois
<scribe> Scribe: francois
<DKA> ScribeNick: francois
Bryan: small update, I've yet to
find cristallized things to add to the document
... I included some of Sean's Ajax techniques
... 5.11.1: a little bit of update
... In "provide alternatives to Ajax", I have a ref to http://thinkingandmaking.com/entries/63
... It would be really useful if someone could have a look at
it
... and crystallize it
DKA: volunteer to turn that into "How to do it"?
<jeffs> want me to?
Bryan: by default, I'll do that if no one else does.
jeffs: If you want me to take a look at this specific section (5.9.4), I'll do it over the week-end
Bryan: jeffs, I think they have some good ideas, I just don't know how to express it.
<jeffs> nuke?
<scribe> ACTION: jeff to take a look at the ref in 5.9.4 and summarize it [recorded in http://www.w3.org/2008/03/20-bpwg-minutes.html#action01]
<trackbot-ng> Sorry, couldn't find user - jeff
<scribe> ACTION: Sonstein to take a look at the ref in 5.9.4 and summarize it [recorded in http://www.w3.org/2008/03/20-bpwg-minutes.html#action02]
<trackbot-ng> Created ACTION-719 - Take a look at the ref in 5.9.4 and summarize it [on Jeffrey Sonstein - due 2008-03-27].
Bryan: One exchange with francois
which may be good: generalize "Minimize external script files"
to "Minimize external files"
... in 5.5.6, not done in latest draft
... Very small change, but the issue of caching is an important
one.
... In summary, I made a small update based on the discussions
of the mailing-list, if I missed anything, feel free to tell
me
DKA: that's where actions to come up with BPs could be useful. I have one actually.
Bryan: one question: we had two
inputs from ETRI: one about the ADC and the seamless
paper.
... I didn't exactly know what to do with that, and just turned
them into general statements
... It would be useful to expand them
DKA: Yes, I think we need more text in there. The best thing would be to raise this as an issue.
<scribe> ACTION: Dan to raise an issue on extending BPs on ADC and seamless Korean inputs [recorded in http://www.w3.org/2008/03/20-bpwg-minutes.html#action03]
<trackbot-ng> Created ACTION-720 - Raise an issue on extending BPs on ADC and seamless Korean inputs [on Daniel Appelquist - due 2008-03-27].
Bryan: about improving DDC, where does that fit?
DKA: we agreed there would be no
DDC in the document.
... I think we're already addressing the Korean proposal
here.
<jeffs> looking for 5.9.4 reference in draft but not finding it... anyone point me to it?
Bryan: device classification and
appendix on device properties that may be important to request
using the DDR.
... Any help on that would be useful.
<Zakim> Kai, you wanted to ask about ddc
Bryan: was trying to find actionable items about both Appendix A and B (Best Practice Dependent Device Properties and Classification of Devices)
<jeffs> looking for 5.9.4 URI reference in draft but not finding it... anyone point me to it?
<jeffs> thanks
Kai: could someone remind me why we decided against ADC?
DKA: [explaining the decision].
We've gone back and forth on that one. I don't want to do that
once again.
... DDC, the fact that we put it in BP1, the fact that it was
misunderstood by our Korean friends, typically show it probably
wasn't a good idea
Kai: I just have the feeling that ADC is what people need. I think it's a mistake, making the problem worse, rather than solving anything.
DKA: we can open an issue on that
Kai if you want to
... I think we structured the doc in BP2 so that we don't need
an ADC.
Kai: I think the issue is a good idea. I'll raise one not to have the group lose time on this.
Kai: The London F2F was held, not
strongly attended but very good progress made. Tackled more
stringent repeatable test requirements.
... Interesting discussions about Mobile OK Pro and BP2, e.g.
do we want RecTrack for Pro or remain a note; in the end
creating guidelines imply room to comply.
... Good progress was made on interpreting test results.
... an updated document is available, but with 5 or 6 tests
updated, unsure whether to put it on the website
Dan: proposes to put a new draft on the website
Kai: appeal to all for feedback when the new version is posted; on the text differences, and if the type of tests is valuable and we should proceed in this manner
Dan: will you send an email for input?
Kai: Yes
Dan: re the Note vs RecTrack, the BP1 was RecTrack but non-normative; Mobile OK was normative. A note is less potent, so maybe this should be on the RecTrack
Kai: Negative on publishing as a note, but should remain on plan for RecTrack
Dan: any machine testable tests?
Kai: Yes, and some should be supportable by checker
Dan: Jo, anything to add on the Pro F2F?
Jo: not other than the general
discussions
... thought Pro could be made more machine testable by
incorporating BP2
Dan: is there an issue or action to move that forward?
Kai: suggests to resolve subjectivity of the tests, taking that away. Also repeatability. Addressing those first is proposed, then address Pro as a whole e.g. relation to BP2
<jeffs> firmly agree
Kai: no proposed actions other than what is being done
Dan: there were ideas of having a
focus group to address Korean needs, meeting in Korea, able to
report and interact with the group
... trying to get that idea going
<jo> [have exchanged email with Seungyun and he will take it to his colleagues and get back to us]
<francois> thesis summary
Francois: the document is long, related to analysis of AJAX performance, conclusion is that things are slow, and there is wide variance in performance
<jo> ACTION-696?
<trackbot-ng> ACTION-696 -- François Daoust to summarise the U Helsinki masters thesis on Mobile Ajax performance with a view to including some aspects into BP 2 -- due 2008-03-11 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/696
Francois: there seems to be some
interesting material to include or reference; not anything new
necessarily, but some ideas about efficiency for performance
purposes
... we can use it as emphasis for what we have written.
Extracted some info from a Yahoo document, not sure if it is
related to BP or not, e.g. reduce DNS lookups
... not sure if these are BP for mobile, but in the constrained
environment this make more sense
Bryan: this is usually an issue for non-proxied browsers (few) and specialized web applications, but it may be useful to look at
Francois: one issue, re use of HTTP streaming, using persistent HTTP connections, may be valuable
<jo> (keep-alive)
Dan: not sure if these are elaborations, re the Seoul discussion we want clear developer focused details
<jeffs> persistent connections with a device that is mobile may be problematical
Bryan: support the reference to HTTP keep-alive, this is a common optimization that we recommend
Dan: are you suggesting to reference this document?
Francois: other than linking them
to the existing BPs, its one more external source in the same
direction
... don't recommend directly linking each BP to the external
documents, just as a reference
Dan: we will give Bryan an action to review Francois's report to extract as possible
<jo> ACTION-714?
<trackbot-ng> ACTION-714 -- Jo Rabin to draft review of MMA Advertising -- due 2008-03-20 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/714
<jo> jo's proposed response to MMA
Jo: overall we think it is a good
idea to have a global recommendation
... Web should be capitalized, ref MWBP, Mobile OK, Checker,
reducing redirects for Ads
<DKA> PROPOSED RESOLUTION: We agree the text of response to MMA guidelines as drafted by H. Rabin.
<francois> +1
<jo> PROPOSED RESOLUTION: We agree the text of response to MMA guidelines as drafted by H. Rabin under ACTION-714
Bryan: why would we recommend against caching?
Jo: Sean's idea, retrieving an ad, not strongly held, maybe we should be silent on this
<jeffs> sorry, hamfisted today
<francois> Bryan: I think it assumes one particular advertising paradigm, but use of caching for content is typically a practice we would recommend
Bryan: caching is a viable option if statistics are not an issue
<jo> PROPOSED RESOLUTION: We agree the text of response to MMA guidelines as drafted by Jo under ACTION-714 modulo changing text on caching to say "and ensuring that the correct cache options are used"
<Kai> I agree very much with Bryan
Bryan: agrees that cache use is context-specific
+1
<DKA> +1
<Kai> +1
<francois> +1
<jeffs> +1
Jo: looking for Dom to send the response
<jo> RESOLUTION: We agree the text of response to MMA guidelines as drafted by Jo under ACTION-714 modulo changing text on caching to say "and ensuring that the correct cache options are used"
ACTION-710
<jo> ACTION-710?
<trackbot-ng> ACTION-710 -- François Daoust to test the Effect of HEAD Requests on Various Servers -- due 2008-03-13 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/710
Francois: will send a report to the list, it's still open
<jo> [+1 to closing call at this stage]
<francois> ACTION: daoust to prepare a report's form to fill to test Aaron static/dynamic example [recorded in http://www.w3.org/2008/03/20-bpwg-minutes.html#action04]
<trackbot-ng> Created ACTION-721 - Prepare a report's form to fill to test Aaron static/dynamic example [on François Daoust - due 2008-03-27].