00:00:51 RRSAgent has joined #webapps 00:00:51 logging to http://www.w3.org/2014/05/07-webapps-irc 00:00:58 zakim, this is dom3 00:00:58 ok, Travis; that matches RWC_WAPI(D3E)8:00PM 00:01:09 rrsagent, this meeting spans midnight 00:01:23 RRSAgent, make logs public 00:01:34 Present+ Travis 00:01:39 scribe: Travis 00:01:43 scribeNick: Travis 00:02:33 garykac has joined #webapps 00:02:45 Hello, 00:02:55 hallo 00:03:34 +garykac 00:04:25 Present+ Gary 00:04:41 Present+ Masayuki 00:04:49 Masayuki: I put in a fix for https://www.w3.org/Bugs/Public/show_bug.cgi?id=23906 00:05:01 Zakim, who is on the phone? 00:05:01 On the phone I see [Microsoft], garykac 00:05:10 zakim, [microsoft] is me 00:05:10 +Travis; got it 00:05:24 BAsically just adding a virtual 'Accel' key that maps to 'OS' (Win), 'Meta' (Mac) or Super/Hyper (Linux) 00:05:47 Topic: Trying to close out 23906 00:05:49 https://www.w3.org/Bugs/Public/show_bug.cgi?id=23906 00:06:11 And with the modifier keys, we have attributes for shift, alt, meta and control 00:06:34 We don't want to any more attributes (since we'd probably have to add another 6 or 7, which would be unwieldy) 00:07:17 But if we want people to use getmodifierstate, should these 4 attributes be marked as deprecated? 00:07:48 Also, we have a lot of modifier keys in the modifier key list. I'm not sure that all of them make sense with getmodifierstate(). 00:07:49 garykac: Oh, that's not my intent of bug 23906. I want it maps to Ctrl (Win), Command (Mac), depends on settings (Linux). 00:09:09 masayuki:ah. 'ctrl' for win makes sense. that way 'Accel' + c is "copy" on all 3 OSs. 00:10:36 jcraig has joined #webapps 00:11:55 Um, adding only on getModifierState() is simpler, though... 00:13:51 masayuki: yes, it could have been put there 00:14:52 I think that on most environments, the acceleration key on Linux is Ctrl. E.g., Fedora and Ubuntu. 00:16:08 However, somebody using Mac keyboard may customize acceleration key as Super or Hyper if it's possible. 00:17:34 Is it easy to get the current acceleration key on Linux? 00:18:02 I'm wondering why we are using the term "Accel" -- this confused me, and made me think of ALT key use, such as ALT+F to get the File menu, ALT+T for the tools menu, ALT+H for help... I thought those things were called "accellerators" 00:18:03 If it's hard to determine which key is being used for menu shortcuts, then this might be too complex to be worthwhile. 00:18:21 On Firefox, users can customize it with Firefox's about:config. How about Google Chrome? 00:19:56 masayuki: I don't know 00:21:55 I think that which is an "Accel" modifier should be decided by browser because its purpose is that providing a way to useful web apps own shortcut key implementation. 00:21:58 karl has joined #webapps 00:22:45 If browser's shortcut key is Alt+Foo, when KeyboardEvent.altKey is true, .getModifierState("Accel") should return true. 00:23:07 I just remembered that Web applications can't reliabily work with ALT (on windows) anyway, because the OS steals focus from the browser client to focus the browser frame's accelerator menus. So, yeah, ALT doesn't really work as an Accel key. 00:23:43 Travis: Yes. It's just an example. 00:24:35 My only suggestion at this point is to go with the full name "Accelerator" rather than "Accel" which seems like a weird shortname... 00:24:42 Does FF Mac have the ability to change the accel key? I don't see it in prefs. 00:25:55 garykac: Yes, it's ui.key.accelKey. You need to reboot Firefox, though. 00:26:46 17 for Ctrl, 18 for Alt, 224 for Meta, 91 for Win, 0 for none. 00:27:07 (the values are .keyCode value on Gecko...) 00:29:21 I'm not sure which is better "Accel" or "Accelerator" since it should be natural word for native English engineers. 00:30:26 Actually, "AccelKey" might work for me. 00:30:43 masayuki: on Mac most shortcuts use ⌘, so changing this setting would affect those shortcuts. But the "Home" shortcut is Option + Arrow. What would happen to this shortcut if I change my default shortcut to Option? 00:31:42 garykac: Not affected. And it may conflict with Command + Home. 00:31:52 ok. 00:32:06 so, I'll change the text to map 'Accel''Ctrl' 00:32:16 ...'Accel' to 'Control' for Windows 00:34:04 Another issue: there are 15 modifier keys. I'm not sure it makes sense for getmodifierstate to support them all. 00:34:06 HTML5 is using the term accesskey for the ALT combos in Windows: http://www.w3.org/html/wg/drafts/html/master/single-page.html#using-the-accesskey-attribute-to-define-a-command-on-other-elements 00:34:18 So, basically "access" is out for a name :) 00:34:44 On the Mac, you cannot query for the state of the CapsLock key. So I'm not sure how get modifier state would support that. 00:35:42 And I think that Fn and FnLock are not exposed as well. 00:37:28 Note, we now have "DeadKey" in the key tables as of a few days ago... may need to change that to just "Dead"? 00:38:20 I think that 'Dead' is more consistent - we don't use the 'Key' suffix for any other key value. 00:38:57 garykac: It's included in [NSEvent modifierFlags]. See http://mxr.mozilla.org/mozilla-central/source/widget/cocoa/nsCocoaUtils.mm#550 00:41:30 masayuki: That's great. Being able to query CapsLock state is very useful. I've heard that there were problems getting it on the Mac - glad to see there's a way to do it./ 00:47:38 With beforeinput, FF has a bug: https://bugzilla.mozilla.org/show_bug.cgi?id=970802 00:48:22 jcraig has joined #webapps 00:56:15 We noted that in HTML5, the input event is just a generic Event type, not an InputEvent. There's no technical reason to have .data, but having isComposing is useful. We should be mindful that this is different from HTML. 00:59:45 jungkees has joined #webapps 01:00:43 OK. I'm going to check with folks internally who are interested in working on beforeinput. To check to see if they have any comments on it. 01:00:49 I think we're done for today. 01:00:58 Aside from that, I don't have anything else today to talk about. 01:01:00 I plan to work on some more bugs this week. 01:01:35 I'll make the additional changes to clarify the Accel key text. I'll resolve the bug and ping you two to review the updated text. 01:01:36 I'm working on KeyboardEvent.code on Firefox. 01:01:37 Thansk! 01:01:45 Thanks. 01:01:51 See ya! 01:02:00 masayuki: Cool. Let us know if you encounter anything strange. 01:02:01 -Travis 01:02:07 g'night. 01:02:15 -garykac 01:02:16 RWC_WAPI(D3E)8:00PM has ended 01:02:16 Attendees were garykac, Travis 01:02:33 rrsagent, please make the minutes 01:02:33 I have made the request to generate http://www.w3.org/2014/05/07-webapps-minutes.html Travis 01:02:41 RRSAgent, make logs public 01:48:04 jcraig has joined #webapps 02:04:58 jcraig has joined #webapps 02:19:53 heath has joined #webapps 02:41:28 karl has joined #webapps 02:45:19 lmclister has joined #webapps 03:17:44 karl has joined #webapps 05:00:03 lmclister has joined #webapps 05:02:41 lmclister has joined #webapps 05:07:15 xiaoqian has joined #webapps 06:21:37 chaals has joined #webapps 06:32:38 Zakim has left #webapps 06:55:10 lmclister has joined #webapps 07:35:00 chaals has joined #webapps 08:19:46 darobin has joined #webapps 08:48:37 chaals has joined #webapps 09:54:25 smaug has joined #webapps 10:07:43 abarsto has joined #webapps 10:22:52 RRSAgent, make minutes 10:22:52 I have made the request to generate http://www.w3.org/2014/05/07-webapps-minutes.html ArtB 10:26:50 chaals1 has joined #webapps 10:35:33 rrsagent, bye 10:35:33 I see no action items