13:42:43 <RRSAgent> RRSAgent has joined #webperf
13:42:43 <RRSAgent> logging to http://www.w3.org/2012/10/31-webperf-irc
13:43:06 <krit> krit has joined #webperf
13:43:17 <cabanier> cabanier has joined #webperf
13:43:31 <noriya_> noriya_ has joined #webperf
13:43:51 <sgodard> sgodard has joined #webperf
13:44:34 <tobie> Dan: missing from WebPerf agenda: end-user and operator visibility in perf optimization.
13:44:37 <jfmoy> Present+ Jean-Francois_Moy
13:44:46 <sgodard> Present+ sgodard
13:45:03 <tobie> Paul: agreed. required info from users of specs
13:45:12 <tobie> ... loook at audio spec
13:45:14 <Arnaud> present+ Arnaud_Braud
13:45:36 <tobie> ... every spec really are missing perf metrics for exit criteria.
13:45:56 <tobie> ... which means that implementations might not be usable by devs.
13:46:19 <tobie> ... I'd like to get perf criterias as a req to go to rec.
13:46:32 <tobie> ... I'd like to get feedback about this.
13:46:36 <plh> q+
13:47:07 <tobie> aaa: skeptical about this because of the wide range of devices that implement specs.
13:47:15 <tobie> ... could define profiles.
13:47:21 <sato> sato has joined #webperf
13:47:28 <tobie> ... much more important is to be able to measure.
13:48:01 <tobie> Daniel: requires a bigger focus on network.
13:48:11 <tobie> aaa: not only.
13:48:21 <tobie> ack phl
13:48:23 <sgodard> q?
13:48:52 <plh> q-
13:48:53 <tobie> phl: difficult to define this because of the different class of devices.
13:48:58 <Zakim> Zakim has joined #webperf
13:49:03 <bjkim> bjkim has joined #webperf
13:49:28 <tobie> Paul Bakaus: this should be done based on use cases that can be easily expressed in perf reqs
13:49:39 <tobie> ... some are hard to express, but some are easy.
13:50:05 <tobie> aaa: seamless audio splicing is an important issue with [?]
13:50:25 <tobie> Dan: perf doesn't necessarily mean numbers.
13:51:03 <tobie> ... defining perf profiles within specs can help devs
13:51:19 <tobie> s/aaa/Henry/
13:51:49 <tobie> Daniel: would like to be able to compare native apps perf to web apps.
13:52:23 <tobie> pbakaus: difficult to do the spec work in the native side.
13:52:35 <tobie> Daniel: its similar at the network level.
13:53:04 <jeff> jeff has joined #webperf
13:53:06 <tobie> ... so you can get at least network perf.
13:53:45 <tobie> phl: network measurement ispart of the discussion at webperf, so adding use cases would be good. The group has little experience in network. so please join if you do.
13:54:03 <tobie> Daniel: paypal native app calls http apis
13:54:28 <tobie> ... backend is similar to what you have for reg web apps.
13:54:41 <tobie> ... one of the thing webperf should do is look into this.
13:55:04 <tobie> pbakaus: other suggestions on how to have perf a criteria for successful specs?
13:55:53 <tobie> ... what bothers me is to see implementors claim support but don't have good enough perf caracteristics.
13:55:57 <tomoyuki> tomoyuki has joined #webperf
13:56:17 <tobie> slightlyoff: phrase your questions in form of benchmarks.
13:56:19 <tomoyuki> tomoyuki has left #webperf
13:56:30 <tomoyuki> tomoyuki has joined #webperf
13:56:34 <tomoyuki> tomoyuki has left #webperf
13:56:35 <tobie> ... eg robohornet effort
13:56:49 <tobie> ... iplementors want clarity too.
13:57:22 <tobie> pbakaus: adding benchmarks to spec criteria seems reasonable.
13:57:40 <tobie> ... related issue:
13:58:17 <tobie> ... problem is not only with slow implementations but also that some specs can't be used because related specs aren't implemented.
13:58:38 <tobie> ... eg every dev runs in the same kind of problem on mobile using gyroscope
13:58:53 <tobie> ... which can't be used because orientation lock is missing.
13:59:14 <tobie> ... real problem that can't be expressed easily.
13:59:21 <tobie> ... what can be done?
13:59:28 <tobie> Daniel: specs list deps
13:59:42 <tobie> ... would you seen a need for something beyond that.
14:00:09 <tobie> pbakaus: in that case, screen orienation doesn't exist yet.
14:01:09 <tobie> slightlyoff: how that's been done is to pile up everything that's required in the same spec.
14:01:12 <kotakagi> kotakagi has joined #webperf
14:01:22 <tobie> Daniel: perf isn't taken seriously.
14:01:35 <tobie> Henry: can't be right.
14:01:54 <tobie> ... perf is crucial.
14:02:52 <tobie> slightlyoff: coremob test suite has perf aspects to it.
14:03:09 <tobie> ... game here is to get vendors interested in your use cases
14:03:15 <tobie> ... benchmarks
14:03:27 <tobie> ... benchmarks are hard to write
14:03:39 <tobie> ... can cause premature optimization
14:03:50 <tobie> ... pushing back on the idea that people don't care.
14:03:53 <tobie> ... they do.
14:05:08 <tobie> bbb: we have people doing games. we found that the best solution was to talk with vendors and ask them why a particular problem is slow.
14:05:51 <tobie> pbakaus: lot of interest in perf, we just need to know how to make it happen.
14:06:02 <tobie> ... works well when benchmarks are not written by vendors.
14:06:29 <tobie> ... used as a way to market browsers.
14:06:50 <tobie> ... any other feedback?
14:07:24 <tobie> ccc: making benchmarks part of an accepted test suite makes sense.
14:07:32 <Arno> Arno has joined #webperf
14:07:47 <tobie> ... better in a test suite for process reasons.
14:08:15 <tobie> Daniel: within w3c we could decide on a vendor agnostic methodology to define how to measure the web.
14:08:28 <tobie> ... this should be pretty straightforward.
14:08:48 <tobie> slightlyoff: you're vastly underestimating the complexity and difficulty of the problem.
14:09:03 <tobie> Daniel: maybe but this would still be a useful doc.
14:09:11 <ht> ht has joined #webperf
14:09:13 <tobie> ccc: we need to write benchmarks first
14:09:15 <plh> q+
14:09:40 <tobie> ... open-source browsers have benchmarks
14:09:42 <ht> q+ to mention IETF work on HTTP nextgen
14:09:50 <tobie> ... why don;t we migrate those back into the W3C?
14:10:02 <tobie> s/;/'/
14:10:18 <tobie> phl: space of webperf is pretty vast.
14:10:31 <tobie> ... it helps when people come with use cases and examples.
14:10:54 <tobie> ... where do we start?
14:11:07 <tobie> pbakaus: is the webperf the right venue?
14:11:19 <tobie> phl: no convinced either way.
14:11:26 <tobie> ... but that could be possible.
14:11:51 <tobie> pbakaus: i like the benchmark approach but that works for single specs only
14:12:11 <tobie> ... devs use tech in ways that aren't considered
14:12:22 <tobie> ... eg property change in IE
14:12:42 <tobie> ... some of those tech, when used together can get really slow.
14:12:54 <tobie> ... not sure how to bring this up to W3C.
14:13:22 <tobie> ... is this a problem? Where should this be brought?
14:13:40 <dbaron> dbaron has joined #webperf
14:13:49 <tobie> ... devs using a combination of tech and discovering it is slow.
14:13:54 <tobie> Daniel: examples?
14:14:04 <tobie> pbakaus: using SVG and canvas is slow.
14:14:15 <tobie> ... technically it works.
14:14:21 <tobie> ... but its slow.
14:14:39 <tobie> phl: use benchmarks is the right approach to raise the issue.
14:14:58 <tobie> ... in the webperf WG, we realized Firefox was creating extra work when using CSS.
14:15:07 <tobie> so we talked to Mozilla.
14:15:39 <tobie> ... vendors need to be told about those issues and help prioretize
14:15:56 <tobie> pbakaus: what is the venue for this?
14:15:58 <Hidetoshi> Hidetoshi has joined #webperf
14:16:26 <tobie> Henry: in terms of end to end benchmarking you have friends at the IETF.
14:16:49 <tobie> ... everyone agrees about that a lot of work will need to be put in benchmarking.
14:17:01 <tobie> ... not the same focus as here, but there's a lot of overlap.
14:17:19 <tobie> Dan: I can also take this to the IETF next week.
14:17:47 <tobie> ... IETF has working groups that are specialized on benchmarking
14:18:20 <tobie> Daniel: early days for that group.
14:18:37 <tobie> Henry: name HTTPBis is getting overloaded.
14:19:39 <tobie> pbakaus: bring pain points to the webperf mailing list.
14:19:52 <tobie> ... so far we've been talking about memory perf.
14:20:02 <tobie> ... garbage collection.
14:20:10 <dan_romascanu> dan_romascanu has joined #webperf
14:20:11 <tobie> and tobie has been talking about scrolling perf.
14:20:22 <tobie> ... we need to hear more from you.
14:20:22 <dan_romascanu> ietf bmwg - https://datatracker.ietf.org/doc/charter-ietf-bmwg/
14:20:38 <tobie> phl: we published a survey
14:20:49 <ht> So, the httpbis IETF WG has been rechartered to put HTTP/2.0 into scope, see http://tools.ietf.org/wg/httpbis/charters for an overview
14:21:07 <tobie> ... we have to decide between now and the end of Nov what our focus is going to be on.
14:21:21 <tobie> ... we are good at implementing stuff.
14:22:26 <tobie> pbakaus: super grateful for the work of the webperf group.
14:22:40 <sgodard> tobie: Great great job for your first time as scribe ;)
14:23:34 <bjkim> bjkim has left #webperf
14:24:13 <krit> krit has joined #webperf
14:24:39 <tobie> rrsagent, [please] create [the] minutes
14:24:39 <RRSAgent> I'm logging. I don't understand '[please] create [the] minutes', tobie.  Try /msg RRSAgent help
14:24:50 <tobie> s/rrsagent, [please] create [the] minutes//
14:25:08 <tobie> RRSAgent: generate minutes
14:25:08 <RRSAgent> I have made the request to generate http://www.w3.org/2012/10/31-webperf-minutes.html tobie
14:25:20 <rotsuya> rotsuya has joined #webperf
14:26:27 <Hidetoshi> Hidetoshi has joined #webperf
14:34:17 <krit> krit has left #webperf
14:36:34 <Arnaud> Arnaud has joined #webperf
14:42:44 <jfmoy> jfmoy has joined #webperf
14:43:22 <cabanier> cabanier has joined #webperf
14:45:22 <jeff> jeff has joined #webperf
14:47:32 <Hidetoshi> Hidetoshi has joined #webperf
14:47:38 <Hidetoshi> Hidetoshi has left #webperf
14:48:55 <morrita> morrita has joined #webperf
14:49:12 <sgodard> sgodard has joined #webperf
14:50:04 <sgodard> sgodard has left #webperf
14:53:12 <rotsuya> rotsuya has joined #webperf
14:54:48 <hiro_away> hiro_away has joined #webperf
14:56:42 <plh> plh has joined #webperf
14:57:10 <cabanier> cabanier has joined #webperf
14:59:36 <cabanier> cabanier has left #webperf
14:59:41 <Arnaud> Arnaud has left #webperf
15:02:49 <jfmoy> jfmoy has left #webperf
15:07:25 <ht> ht has joined #webperf
15:07:31 <ht> ht has left #webperf
15:12:30 <sylvaing> sylvaing has joined #webperf
15:13:48 <kotakagi> kotakagi has joined #webperf
15:17:39 <kotakagi> kotakagi has joined #webperf
15:49:44 <kotakagi> kotakagi has joined #webperf
15:50:47 <jeff> jeff has joined #webperf
16:00:04 <Rossen> Rossen has joined #webperf
16:05:17 <plh> plh has joined #webperf
16:06:19 <trackbot> trackbot has joined #webperf
16:07:42 <trackbot> trackbot has joined #webperf
16:08:29 <rotsuya> rotsuya has joined #webperf
16:08:49 <trackbot> trackbot has joined #webperf
16:12:11 <trackbot> trackbot has joined #webperf
16:13:12 <trackbot> trackbot has joined #webperf
16:15:23 <trackbot> trackbot has joined #webperf
16:18:11 <trackbot> trackbot has joined #webperf
16:22:23 <trackbot> trackbot has joined #webperf
16:25:08 <Zakim> Zakim has left #webperf
16:29:51 <trackbot> trackbot has joined #webperf
16:56:02 <Rossen> Rossen has joined #webperf
16:57:24 <JatinderMann> JatinderMann has joined #webperf
16:57:37 <JatinderMann> trackbot, start telecon
16:57:39 <trackbot> RRSAgent, make logs world
16:57:39 <Zakim> Zakim has joined #webperf
16:57:41 <trackbot> Zakim, this will be WPWG
16:57:41 <Zakim> I do not see a conference matching that name scheduled within the next hour, trackbot
16:57:42 <trackbot> Meeting: Web Performance Working Group Teleconference
16:57:42 <trackbot> Date: 31 October 2012
17:00:02 <plh> plh has joined #webperf
17:00:10 <plh> zakim, this is per
17:00:10 <Zakim> plh, I see RWC_web-per()1:00PM in the schedule but not yet started.  Perhaps you mean "this will be per".
17:00:20 <plh> zakim, this will be per
17:00:20 <Zakim> ok, plh; I see RWC_web-per()1:00PM scheduled to start now
17:00:32 <Zakim> RWC_web-per()1:00PM has now started
17:00:39 <Zakim> +[Microsoft]
17:00:44 <plh> zakim, passcode?
17:00:44 <Zakim> the conference code is 97373 (tel:+1.617.761.6200 sip:zakim@voip.w3.org), plh
17:00:46 <JatinderMann> present+ Jatinder
17:02:19 <Zakim> + +1.503.264.aaaa
17:02:20 <Zakim> +plh
17:02:38 <JatinderMann> present+ Plh
17:02:43 <JatinderMann> prssent+ Ganesh
17:04:13 <Zakim> + +1.650.214.aabb
17:04:41 <tonyg> tonyg has joined #webperf
17:05:03 <GaneshRao> GaneshRao has joined #webperf
17:05:57 <Zakim> + +1.206.496.aacc
17:16:08 <Zakim> - +1.503.264.aaaa
17:16:09 <Zakim> -[Microsoft]
17:16:12 <Zakim> -plh
17:16:14 <Zakim> - +1.650.214.aabb
17:16:25 <Zakim> - +1.206.496.aacc
17:16:27 <Zakim> RWC_web-per()1:00PM has ended
17:16:27 <Zakim> Attendees were [Microsoft], +1.503.264.aaaa, plh, +1.650.214.aabb, +1.206.496.aacc
17:20:27 <trackbot> trackbot has joined #webperf
17:21:32 <trackbot> trackbot has joined #webperf
17:22:21 <trackbot> trackbot has joined #webperf
17:24:21 <trackbot> trackbot has joined #webperf
17:57:14 <Alois> Alois has joined #webperf
17:58:00 <Zakim> RWC_web-per()1:00PM has now started
17:58:08 <Zakim> + +43.664.853.aaaa
18:00:24 <Alois> Hello?
18:00:35 <jfmoy> jfmoy has joined #webperf
18:03:18 <Alois> Nobody on the phone?
18:03:25 <Zakim> - +43.664.853.aaaa
18:03:26 <Zakim> RWC_web-per()1:00PM has ended
18:03:26 <Zakim> Attendees were +43.664.853.aaaa
18:04:36 <Alois> It says the conference is restricted
18:08:46 <Alois> Hello, cannot join the conf call. Anyone else having problems as well?
18:14:46 <Alois> Alois has joined #webperf
18:16:02 <Alois> Hello?
18:54:13 <trackbot> trackbot has joined #webperf
20:18:10 <Zakim> Zakim has left #webperf
22:38:42 <jfmoy> jfmoy has joined #webperf
23:21:19 <rotsuya> rotsuya has joined #webperf
23:37:28 <rotsuya> rotsuya has joined #webperf