Web Events WG Voice Conference

08 Feb 2011


See also: IRC log


Art_Barstow, Josh_Soref, Matt_Brubeck, Doug_Schepers, Laszlo_Gombos, Sangwhan_Moon, Olli_Pettay


<scribe> ScribeNick: ArtB

<scribe> Scribe: Art

Date: 8 February 2011

Tweak the agenda

AB: I posted a draft agenda yesterday ( http://lists.w3.org/Archives/Public/public-webevents/2011JanMar/0045.html ). I intend to merge topics #4 and #6 since they both are about Use Cases. Any change requests?

Issue management:

AB: Working Groups handle issues differently: embed the issues in the spec, use W3C's tracker, Bugzilla, etc. Let's briefly talk about how we want to handle issues.
... anyone have comments or feedback re issue tracking

DS: my preference is to use Tracker (W3C tool)
... it has a nice integration with IRC
... less cluttered with Bugzilla

AB: also has nice integration with e-mail

DS: from the WebEvents page, look for issues

<shepazu> http://www.w3.org/2010/webevents/track/

DS: there is only one action now
... and no issues
... Action-1 is still open
... and I'll take care of it
... Tracker is easy to use

<shepazu> issue-1?

<trackbot> ISSUE-1 does not exist

<shepazu> action-1?

<trackbot> ACTION-1 -- Arthur Barstow to work with Doug on a voice conference time of day that works for most people -- due 2010-12-15 -- CLOSED

<trackbot> http://www.w3.org/2010/webevents/track/actions/1

DS: has some macros that trackbot understands and acts on
... to add issues, may need to copy text from an email to tracker issue
... so a bit of a pain
... to both edit the spec and to do issue tracking
... As such, if someone wants to volunteer to help with issue tracking that would be very welcome
... Requires monitoring the list and then adding issues to Tracker
... One email may have more than one issue

AB: is anyone willing to volunteer to lead the issue tracking

MB: I will tend the issue tracker

DS: this is good and a nice way to ease into the editing
... Think the spec should include credit for people that take on big tasks like Issue tracking and Test suite work, etc.

AB: I think that's a great idea

SM: I just added two issues

DS: that's great
... we can have a 3-way call and go through some of the tasks

<scribe> ACTION: doug set up a conf call with Matt and Sangwhan re issue tracking and editing [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action01]

<trackbot> Created ACTION-6 - Set up a conf call with Matt and Sangwhan re issue tracking and editing [on Doug Schepers - due 2011-02-15].

RESOLUTION: the WG agrees to use Tracker for issue tracking (and action tracking)

AB: thanks guys!

DS: an advantage Bugzilla has is that anyone in the Public create Issues; whereas, only WG members can create issues via Tracker

MB: so if Joe Public wants to create/raise an Issue, they must use the list?

DS: yes

Comments by PPK

<timeless> [of note, bugzilla can be configured to have similar restrictions ]

AB: PPK read Doug's first ED and submitted some comments ( http://www.quirksmode.org/blog/archives/2011/01/w3c_touch_event.html )

DS: does anyone have comments

SM: re his comments about radiusXandY
... physical units create probs for mobile browsers
... pixels make much more sense
... than something like centimeters

DS: I agree
... would be good if we could convert but think screen pixels is reqd
... Someone else raised an issue about radiusxy
... It may disappear

OP: re altKey, etc., what about Pens?

DS: some Pens do have keys
... I talked to at least one engr at Wacom
... they have >=1 buttons on the pens
... and they can be configured

<timeless> [ http://www.wacom.com.au/intuos3/spec/intuos3artpen.html ]

<timeless> [ Wacom ]


SM: we should add Meta key

MB: PPK links to Apple Safari doc

DS: I created a blog ( http://schepers.cc/getintouch )
... after reading PPK's blog, I went thru point-by-point and cleaned up the spec to address most of his points
... Sangwhan, perhaps you could add the Meta key
... I think I addressed ontouch move
... I still have questions about time stamp
... Units for radiusXY will need some work
... Unit of force: defined it as 0 to 1 as relative due to device

<Sangwhan_Moon> ACTION: Add meta key for the TouchPoint interface [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action02]

<trackbot> Sorry, couldn't find user - Add

DS: touch cancel event - I'm not sure what this could be needs some work

<mbrubeck> hg log of shepazu's spec changes: http://dvcs.w3.org/hg/webevents/

<timeless> [ http://twitter.com/w3cwebevents ]

<Sangwhan_Moon> http://www.w3.org/2010/webevents/track/actions/7

DS: fyi, I created a twitter account for webevents

<shepazu> twitter feed @w3cwebevents

DS: I intend to push spec changes to @w3cwebevents

SM: is it branch aware?

DS: not sure
... we want to be able to tweet directly rather than going thru a 3-rd party service as I do now

AB: excellent
... anything else on PPK's comments?

Comments by Andrew Grieve - Use Cases and Reqs

AB: Google's Andrew Grieve submitted some comment regarding use cases ( http://lists.w3.org/Archives/Public/public-webevents/2011JanMar/0043.html )

<mbrubeck> http://dvcs.w3.org/hg/webevents/rss-log/default might work

AB: so far, no one has responded

DS: I will respond to Andrew
... it's encouraging that we already have comments so early in the spec process

AB: yes, I agree

DS: I prefer to take other topics and come back to Andrew's email if we have time

Use Case and Requirements

AB: Use Case and Reqs wiki was created ( http://www.w3.org/2010/webevents/wiki/UCsandReqs ) and it needs some work. I don't think we need to be overly prescriptive on how the UCs and Reqs are documented and there is significant variability in the way WGs have documented them.

DS: I would like to work on these but I don't have the time
... pointing to some examples would be good

<scribe> ACTION: barstow send to the list some examples of WGs' documenting UCs and Reqs [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action03]

<trackbot> Created ACTION-8 - Send to the list some examples of WGs' documenting UCs and Reqs [on Arthur Barstow - due 2011-02-15].

<Sangwhan_Moon> It should be possible to start off from previous use cases that Cathy wrote up

AB: is anyone interested in leading or contributing to that process of UCs and Reqs?
... Cathy did some work

<timeless> i'd suggest tweeting asking for people to add to the wiki :)

AB: anything else on UCs and Requs for today?

SM: do we have any Web developers in the WG?
... it would be good to get their input

DS: good point
... and Timeless has a good point too
... asking web devs and tweeting could help; that's kinda' what I did with PPK
... So there is an Action for Everyone to tweet for UCs and Reqs

SM: I've found that browser developers aren't particularly good at writing UCs

<timeless> s/particuarly/particularly/


AB: strictly speaking, a WG is not required to create test cases until a spec enters the Candidate Recommendation phase. In some cases, WGs try to work on test cases much earlier. Ideally, a comprehensive test suite would exist before a spec enters Last Call WD but in practice, I think they are rarely created that early.
... there is also a significant amount of variability in test suites from WG to WG and in some cases e.g. WebApps WG, there is variability from spec to spec.

DS: if tests aren't developed early ...
... when the spec gets to CR and test case are then started, some people will have already left the WG or become inactive
... and when test cases are written, realize the spec has a hole but the CR means people will start implementing
... that ends up tying the hands of the WG
... would be good to stay in the ED state until we have some test cases
... that is, don't publish a WD until test cases exist
... this would be more stringent then what is required
... but this restriction could help us advance the spec more quickly

<timeless> shepazu: so you're holding the WD ransom against implementers providing testcases?

DS: Naturally, we need volunteers to write the test cases
... and intend to include a section in the spec that points to the test suite and give an attribution to whomever leads the test suite work

MB: PPK has a lot of experience with test suites

DS: I asked him but he can't make the commitment
... Does anyone here have experience writing tests? Or no experience writing tests?

OP: how does adding TCs to the TS acutally work?
... because they all need to be reviewed

<timeless> of note, some WGs have published test suites with poor quality tests

OP: are TCs added to bugzilla

<timeless> which lead to problems

<timeless> i think that's SVG/HTML/CSS (?)

OP: it is too easy to write Invalid tests

DS: agree; but a hard problem to solve

SM: we could create a mirror e.g. bitbucket for prelim work

<timeless> the general suggestion is a reviewer.

DS: would prefer to use W3C services

<timeless> at the risk of volunteering for work, i could probably help out here

SM: must have a W3C account to fork

DS: what about just to pull?

<timeless> cloning is possible w/ public repositories

<timeless> but the issue is informing the owner of the w3 repo that your changes are available

SM: need to differentiate merging and just making a clone

<timeless> in theory one could again use twitter :)

DS: think we should have a call about sysadmin type stuff

<timeless> (please include me for this call/topic)

<timeless> the general suggestion was http://bitbucket.org since it provides for pull requests

DS: if a large number of people aren't interested in infra type discussions we could use this call

<timeless> there aren't to my knowledge many hg hosts which do it

DS: how about we dedicate next week's call to infra/sysadmin?



<Sangwhan_Moon> http://www.w3.org/2010/webevents/track/actions/9

AB: coming back to Doug's proposasl, I'm a little uneasy with making test case availability be a block for FPWD
... would make more sense as a block for LCWD

<timeless> art: i think that there might be enough web dev demand that we can remind the web that they won't be able to play with this stuff unless they help provide test cases

<mbrubeck> As vendors like Mozilla implement the spec, we will be writing test cases for our own use...

<timeless> personally, i think we do need to block some *WD on having thoroughly reviewed the test suite

DS: I think people need to speak up, even if you don't care

AB: if anyone has comments about Doug's test case proposal, please speak up now

MB: I don't think I know enough about W3C process to make an informed opinion
... we are already getting comments

DS: we aren't lacking any visibility
... it's too easy to slip into a mode where we are not creating TCs
... will help us narrow down the scope of the spec

<mbrubeck> Is there a standard test harness / library / tool for W3C specs, or should we just create any old web page that runs scripts and displays the results?

<timeless> mbrubeck: there are a couple of harnesses used by different WGs

<timeless> each WG can choose from them or use a new one

AB: I think some people won't review the spec until there is a FPWD

<timeless> iirc some of the recent WG / Specs have tended toward certain harnesses, but i can't recall which

DS: the spec is relatively small
... and don't expect a huge number of TCs

AB: I'd like a little time to think about this; see some clear advantages to the proposal

JS: I think implementors at Moz will write test cases as they implement
... so, if they can easily contribute their TCs, it will help
... I don't think most implementors will distinguish between ED and WD
... they are going to follow the EDs anyway
... Do implementors have a way to namespace-protect their early implementations?

<mbrubeck> The global namespace is rather polluted in this space already. :(

DS: there is a way in D3E
... eg -moz-touchstart

<mbrubeck> Mozilla currently implements MozTouchDown, MozTouchMove, MozTouchUp - https://developer.mozilla.org/en/DOM/Touch_events

DS: I will add something to the spec about adding prefixes

<mbrubeck> If the spec continues to follow WebKit in a mostly backward-compatible way, then I'm not sure prefixes are useful

LG: this will be tricky with WebKit i.e. to state pre-spec and post-spec state
... WK already has some TCS

<mbrubeck> We never prefixed things like <canvas> that followed existing implementations

LG: WG may be able to use them
... We need to think about how to automate the testing
... The WK tests cases have some platform specific parts

DS: automation is not a requirement
... I can work with you Laszlo re if we can re-use WK tests
... Do, you know Art if we can use WK tests?

<scribe> ACTION: barstow work with Doug and Laszlo re if we can re-use Webkit tests [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action04]

<trackbot> Created ACTION-10 - Work with Doug and Laszlo re if we can re-use Webkit tests [on Arthur Barstow - due 2011-02-15].

<timeless> lgombos noted that the webkit tests tend to be platform specific

AB: does anyone object to Doug's proposal that test cases be required before we publish a FPWD?

[ None ]

<mbrubeck> no objection, but I'll read more about process and see if I have any further comments

AB: so, let's give everyone a week to think about Doug's proposal

Plan for high-level intentional events spec:

<timeless> I guess w3 process doesn't require test cases not be limited to a specific platform/device/impl?

AB: Doug, do you have an Estimated Time of Arrival (ETA) for an Editor's Draft of the high-level intentional spec?

DS: I think it is weeks away
... I need to work with other people on that spec

<Sangwhan_Moon> http://www.w3.org/2010/webevents/track/actions/4

DS: I hope to get it out before March 1

<mbrubeck> "Uncontacted tribe in Amazon rain forest announces W3C membership, plans to support HTML5"


AB: we will have a call next week; primary topics are infrastructure and sysadmin
... meeting adjourned

Summary of Action Items

[NEW] ACTION: Add meta key for the TouchPoint interface [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action02]
[NEW] ACTION: barstow send to the list some examples of WGs' documenting UCs and Reqs [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action03]
[NEW] ACTION: barstow work with Doug and Laszlo re if we can re-use Webkit tests [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action04]
[NEW] ACTION: doug set up a conf call with Matt and Sangwhan re issue tracking and editing [recorded in http://www.w3.org/2011/02/08-webevents-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2011/02/08 17:20:04 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/thru/through/
Succeeded: s/Qualcomm/Wacom/
FAILED: s/Qualcomm/Wacom/
Succeeded: s/Qualcomm/Wacom/
Succeeded: s/use case that/use cases that/
Succeeded: s/deverlopers/developers/
Succeeded: s/particuarly/particularly/
FAILED: s/particuarly/particularly/
Succeeded: s/Call WG/Call WD/
Succeeded: s/left the WG/left the WG or become inactive/
Succeeded: s/cases/suite/
Succeeded: s/commentns/comments/
Succeeded: s/intensional/intentional/
Found ScribeNick: ArtB
Found Scribe: Art
Default Present: +1.781.534.aaaa, +1.206.792.aabb, Shepazu, Matt_Brubeck, Josh_Soref, Art_Barstow, Laszlo_Gombos, Olli_Pettay, Sangwhan_Moon
Present: Art_Barstow Josh_Soref Matt_Brubeck Doug_Schepers Laszlo_Gombos Sangwhan_Moon Olli_Pettay
Agenda: http://lists.w3.org/Archives/Public/public-webevents/2011JanMar/0045.html
Found Date: 08 Feb 2011
Guessing minutes URL: http://www.w3.org/2011/02/08-webevents-minutes.html
People with action items: add barstow doug send

[End of scribe.perl diagnostic output]