00:29:36 RRSAgent has joined #css 00:29:36 logging to http://www.w3.org/2009/03/06-css-irc 00:29:38 RRSAgent, make logs member 00:29:38 Zakim has joined #css 00:29:40 Zakim, this will be Style_CSS FP 00:29:40 I do not see a conference matching that name scheduled within the next hour, trackbot 00:29:41 Meeting: Cascading Style Sheets (CSS) Working Group Teleconference 00:29:41 Date: 05 March 2009 00:29:49 rrsagent, make logs public 00:29:56 Meeting; CSS f2f Tokyo 00:30:00 Chair: Chris 00:30:20 zakim, remind me in 8 hours to go home 00:30:20 ok, ChrisL 00:31:01 sylvaing has joined #css 00:31:36 http://people.mozilla.org/~jdaggett/mplustest.html 00:36:16 RRSAgent, make logs public 00:37:28 RRSAgent, spans midnight 00:37:28 I'm logging. I don't understand 'spans midnight', anne. Try /msg RRSAgent help 00:37:40 scribe: anne 00:37:47 Topic: font-weight 00:38:59 sylvaing has joined #css 00:39:04 JD: In Cambridge we discussed bolder and lighter 00:39:06 http://www.w3.org/Style/CSS/Tracker/issues/61 00:39:42 JD: It's a frustrating discussion, because of platform restrictions and lack of [cool] fonts 00:40:33 JD: m+ (Japanese Open Source font) supports seven weights 00:40:45 JD: 300 is equal to 200, 700 is equal to 600 00:45:56 JD: the issue in Cambridge was, given a line of text, given font fallback, multiple fonts can be used, then with nested elements with bolder, bolder, bolder, what should happen? 00:46:52 sylvaing has joined #css 00:46:57 CL: if you have a value already from an inherited context [... scribe missed ...] 00:47:03 JD: where do you make the determination? 00:47:33 JD: am I simply doing incremental steps of boldness or am I picking a font somehow and resolving the calculation at that point 00:47:50 JD: of course the computed value comes in because it becomes funny 00:48:39 DB: computed value can be 1) just a weight, 2) a pair of a weight and a number (number indicates times bolder written), 3) a weight plus a sequence of bolder/lighter steps 00:49:31 SZ: when I say bolder and there are at least two weights in the font, I'm at the lighter of the two, I'm in a span that has a bolder, what should I expect? 00:49:36 JD: euh, the bolder one 00:49:52 SZ: it changes the number 00:50:11 JD: BB proposed that you calculate a number based on the first font 00:50:35 JD: the reason I brought up m+ is that there a number of gotchas with that 00:50:41 JD: [missed] 00:50:42 (Chris demonstrates some tests with ZalamanderCaps, a font with six weights) 00:50:49 JD: if I say my body is 400 and I go to bolder 00:51:34 JD: then you get 500, but then if you have a fallback font that has 400 and 700 you end up with 400 again 00:52:20 JD: in this world, most fonts are 400/700 00:52:43 JD: I suggest we simply iterate through 100/400/700/900 for bolder/lighter 00:53:51 HL: you can still use the absolute values if you really know things 00:53:57 HL: this is just for the relatives 00:54:01 CL: good point 00:54:15 SZ: there are some 400/600 00:54:26 JD: I don't know of those, Japan has some 300/600 00:54:29 not so worried if its just the relatives. As long as higher quality results are not excluded 00:54:30 JD: will still do the right thing 00:54:57 HL: is the rounding defined? 00:54:59 JD: yes 00:55:06 HL: sounds reasonable 00:55:11 JD: you get a consistent computed value 00:55:17 wights in zalamander: 00:55:18 ultrabold 800 00:55:18 bold 700 00:55:18 semibold 600 00:55:18 regular 400 00:55:18 light 300 00:55:20 extralight 250 00:55:39 SZ: IH's distinction goes away, right? 00:55:41 DB: yes 00:55:59 DB: this probably addresses the use case for bolder/lighter better and is much simpler 00:56:03 [humming] 00:56:27 critical difference is that you can do bolder/lighter without knowing the current font family 00:56:39 JD: I will write this out 00:56:54 DB: I think you might want to ping DH directly 00:57:02 DB: I think Opera might do that as well 00:57:20 DB: they basically maintain a sequence of +ses and -ses 00:57:34 [discussion about IH's test and the complexity of the model] 00:58:13 CL: might be useful to make duplicate test with different fonts 00:58:40 JD: In some of the edits made by the other editors they wanted to allow non-multiples of a 100 00:59:11 agree that weights cannot be compared over families 00:59:11 JD: 600 in one font vs 650 in another font is not noticable 00:59:29 s/noticable/consistent/ 01:00:14 SZ: the relatives are not for people who care about typography 01:00:23 [scribe notes they are also for default style of elements] 01:00:51 DB: I don't think authors know the difference between bold and bolder or think bolder is more bold than bold 01:00:55 [joke about adding boldest] 01:01:19 JD: I made a post to www-style 01:01:31 JD: in response to "CSS font selection is broken" (not exact subject) 01:01:44 JD: the way things work on Linux/Windows is unfortunate 01:01:50 http://lists.w3.org/Archives/Public/www-style/2009Mar/0041.html 01:02:28 [people are quoting from the e-mail] 01:02:48 Thomas Phinney's arguments are good but his conclusion does not follow from his arguments 01:02:56 [JD is performing a recap of the e-mail on the whiteboard] 01:04:41 jd: in opentype, font family can have platform-specific variations - localisations, and windows/mac variations so the semantics are platform dependent 01:05:16 ... hence Arial and Arial Black, two families on windows and one family on mac 01:06:09 ... thus MS have a "preferred family" but CSS only allows selecting by slope, width and weight 01:06:42 s/slope/style/ 01:07:02 ... opentype 1.5 has wwsfamily 01:09:43 wpf has a hash table classification of *known* fonts (only) 01:10:25 DB: If I and JD disagree JD is right 01:11:23 JD: there is a possibility in the future that Microsoft might ship fonts with broader families 01:12:01 CL: everyone with a photoshop will have a certain set of fonts and communities around photoshop will start using them, etc. 01:13:36 SZ: [missed] 01:13:47 JD: the first point is that we don't define what a given name maps to on a given platform 01:13:58 JD: on any given platform there will be plethora of different font formats 01:14:03 JD: but we can give guidelines 01:14:10 JD: for OT/TT we suggest that you do certain things 01:14:23 SZ: the important part for me was that it's an abstraction, not a direct 01:14:32 JD: we can suggest a path to goodness 01:14:38 JD: I disagree with your second point 01:15:09 JD: with the exception of optical size I don't see a lot of variations even within Adobe families that somehow would be helped with this 01:15:21 JD: I'm a little sceptical even with optical size 01:15:30 SZ: I understand... in some magical way... 01:15:48 JD: the flipside of this, after the break I want to talk about the specifics of the @font-face mechanism and local faces 01:16:10 JD: via that mechanism you can separate things out... 01:16:31 ... it's going to be so rare; there will be font foundries that define fonts with their own set of axis 01:16:57 ... ... the only thing that we can do is that we put in some kind of string that we can somehow match against the style name 01:17:03 ... which is suicide 01:17:14 ... localization issues, semantics could change, accidental matching 01:17:28 ... using local with the default name is a much more robust mechanism 01:17:42 ... let's take a break and then talk about @font-face 01:18:07 HL: comments from MS? 01:18:35 SG: no, I don't work with the Windows team on fonts, but with IE 01:18:49 SG: and there's a huge back compat 01:50:57 Topic: @font-face 01:51:13 JD: in css3-fonts, descriptors now have default values 01:51:14 JD: two @font-face rules defined; normal and bold 01:51:46 JD: I'm not implying this [see whiteboard photo if someone makes one] is a use case 01:51:49 CL: I think it is 01:52:20 JD: there is no way to have a magic attribute that sucks in a bunch of fonts 01:52:58 JD: MD wants font aliasing 01:53:26 JD: what I'm saying is that essentially having a variable that equals a bunch of fonts 01:53:36 JD: is not a good way to do things 01:53:59 JD: if you want Arial to be just like Helvetica you'd have to enumerate the four faces 01:54:02 _DSC2181 01:54:56 [discussiong about MD's proposal] 01:56:14 JD proposes @font-alias to group font families 01:56:18 _DSC2182 01:58:06 JD asserts that mixing complete families and individual faces gives ambiguity 01:59:01 cl: to point into a zip file with multiple faces you would want a fragment identifier to index into the zipfile 02:01:30 cl: @font-face is designed to point to a face (and give info about it) not to a family 02:02:35 HL: just to try to connect here, can we put @font-alias into a draft 02:02:44 JD: there is a variable proposal, which would do exactly that 02:03:18 HL: not clear whether that will make it and it's far more complex 02:03:55 HL: I think @font-alias is just for browser style sheets 02:04:09 [scribe things the CSSWG should not add features just for browser style sheets] 02:04:12 s/things/thinks/ 02:04:26 JD: I think it is useful for author style sheets too 02:04:37 JD: so you don't have to write everything out 02:05:36 HL: if we see font families expand @font-alias might be useful, but I don't want authors to have to use this 02:05:36 hl: we are seeing font families expand so we need to be able to do this 02:06:09 jd: so @font-alias gives a convenient shorthand 02:06:10 JD: with large corperations that use lots of fonts and if they are doing things cross platform you need to use a lot of fonts 02:06:29 AvK: is font-family the only property it takes? 02:08:02 [simplifying syntax] 02:08:31 @font-alias "my font" "test", "test", "test"; 02:09:35 sylvaing has joined #css 02:11:28 [discussion about parsing rules for font-family] 02:12:07 trackbot, status? 02:12:39 Topic: local() syntax issue 02:12:42 action John to specity @font-alias per 6 March 2009 minutes 02:12:42 Created ACTION-129 - Specity @font-alias per 6 March 2009 minutes [on John Daggett - due 2009-03-13]. 02:12:50 JD: should local require quotes or not? 02:12:55 JD: I put in that quotes are optional 02:13:00 [agreed] 02:13:08 JD: with format() they are required 02:13:14 CL: why? 02:13:50 JD: it's for future safety 02:13:56 to avoid downloading something you know you don't support 02:14:52 JD: we also need it for platform sensitivity 02:15:08 JD: for certain fonts you need to identify what information the font has 02:15:58 JD: so e.g. truetype-aat is skipped over on Windows 02:16:18 _DSC2183 02:17:48 myakura has joined #css 02:19:43 Anne: Spec should say that if the UA doesn't support the format() annotation, the user agent must not download the font. 02:19:53 s/annotation/annotated/ 02:20:01 s/must not download/must not use/ 02:21:09 RESOLVED: format() is authorative 02:21:37 i.e. if the user agent does not support the format listed it will not use the font (and should probably not waste bandwidth either) 02:21:57 jd: src defines load fallback, not character fallback. order is the first font to load succesfully 02:22:51 _DSC2184 02:23:15 JD: in case of multiple listed src() in a single @font-face that are both supported only the first is used 02:23:47 sylvaing has joined #css 02:23:47 [agreed] 02:24:01 RESOLVED: in case of multiple listed src() in a single @font-face that are both supported only the first is used 02:24:15 Topic: unicode-range 02:24:42 JD: the way I have unicode-range is defined right now is that unicode-range has the implicit value of the full Unicode range 02:27:21 unicode-range initial value should be 0-U+10FFFF 02:27:40 JD: In WebKit unicode-range intersects with the supported values of the font 02:27:47 http://www.w3.org/TR/charmod/#C077 02:28:00 JD: within the bounderies of unicode-range 02:29:33 " C077 [S] Specifications MUST NOT allow code points above U+10FFFF. 02:31:10 AvK: e.g. if you have a font with 2,3,5 and unicode-range 1-4 the intersection would be 2,3 02:33:11 CSS 2.1 is correct: "If the number is outside the range allowed by Unicode (e.g., "\110000" is above the maximum 10FFFF allowed in current Unicode), the UA may replace the escape with the "replacement character" (U+FFFD)." 02:34:36 SZ: for big fonts being specific about what is in the font is useful to avoid wasting bandwidth 02:37:49 RESOLVED: unicode-range uses the intersection of the font and unicode-range within the bounderies of unicode-range 02:38:20 s/font/font cmap/ 02:39:45 s/unicode-range uses/the effective unicode range is/ 02:39:45 Topic: type of local() name 02:40:00 JD: there's no ideal name that works on all platforms 02:40:22 JD: on the Mac you'd use the postscript name 02:40:52 ... all fonts on a mac have a postscropt name, either real or synthesized 02:41:11 JD: what I have now is that all platforms allow lookup via the full name 02:41:26 HL: you want to support having style in the name of the family 02:41:41 JD: the name you have here uniquely describes the font within a family 02:41:55 [shouting, minute taker gets lost] 02:42:48 [excited shouting, for those wondering] 02:42:58 fullname is family plus style except for the regular, where style is omitted 02:45:38 JD: postscript name doesn't necessarily match the name in the style 02:45:56 so, you would need to specify it twice, fullname and postscript name? 02:46:03 (cl hears both yes and no) 02:47:06 JD: there are OTF fonts where the full name under Windows is the postscript name 02:47:12 JD: but on the Mac it is the normal full name 02:47:20 so the postscriptname should go first, followed by the fullname (if different to postscript name) 02:50:53 _DSC2185 02:50:55 JD: authors would use local() because they can control individual faces and you cannot do that with font-family 02:51:18 SG: font-family only allows the generic family, not a specific face 02:52:50 JD: with this nomenclature you have the hope of matching cross platform, with file names it fails 02:53:58 SZ: the API is that you give the system a name and use the first it returns 02:54:08 HL: I'm afraid different systems will react differently 02:55:30 JD: with unicode-range and local() you can create interesting combinations that will work across platforms 02:55:51 HL: my experience is different 02:56:01 JD: that is because fonts are not cross platform 02:56:10 JD: this is inherently platform specific 02:56:30 HL: I think that's why we shouldn't do it 02:56:37 AvK: font-family is also platform specific 02:56:44 HL: we should not expand on that 02:59:40 [rehashing of unicode-range and other arguments] 03:00:42 HL: does local() and font-family work the same? 03:00:51 JD: no, but in some cases it will 03:01:01 sz: anything which varies on one of the descriptorts, must be specified by @font face not font-family 03:01:03 CL: we don't want that 03:02:05 s/tthat/people to use font-family: "Helvetica Bold Italic" 03:02:42 s/that/people to use font-family: "Helvetica Bold Italic" 03:07:35 HL: I'm ok with this as long as it is interoperable 03:07:52 I wonder if a font-size descriptor plus src would let us get at optical variants 03:08:01 AvK: the spec can give advice 03:08:03 JD: of course 03:08:27 SZ: I would not like APIs of systems 03:08:39 AvK: if that's the reality it seems better to specify reality 03:09:17 AvK: e.g. ARIA does that too 03:09:26 SZ: as long as it is clear that it is one way of doing it 03:10:21 JD: for the roadmap of this spec I'd like to get what's in the spec now solid 03:10:34 JD: and by the end of this month hopefully get a WD out 03:10:50 JD: and then work on exposing opentype features and some of the other issues regarding better typography 03:10:57 JD: with hopefully a draft for the June F2F 03:12:36 SZ: I did say that I would prefer that the REC not have API based descriptions in the informative note. 03:13:44 SZ: What I really meant was that the mapping of the local name to the font name tables ought to be specified in terms of the font format specifications (without having to invoke that APIs) 03:14:28 SZ: And that the API information should suggest a way to implement that mapping but that not being the only way to implement the mapping 03:21:20 sylvaing has joined #css 04:29:26 scribenick: jdaggett 04:29:41 zakim: hello 04:30:00 zakim, how's the weather? 04:30:00 I don't understand your question, jdaggett. 04:30:24 topic: namespaces 04:30:41 anne: one clarification 04:30:42 Anne: There's just one clarification we want to make to the spec before proceeding 04:31:00 anne: duplicate clarifications are not conformant 04:31:32 anne: after that change and we have two implementations 04:31:43 anne: we can go to pr 04:31:59 discussion of pr criteria 04:32:27 anne: we have a full test suite 04:32:35 chrisl: level of coverage? 04:33:01 discussion of test failurs 04:33:13 s/failurs/failures/ 04:33:43 anne: bugzilla bug exists 04:33:59 discussion of exit criteria 04:35:05 and whether can remove tests 04:35:41 chrisl: no recs got published since i left the group 04:36:34 discussion of specific bug mozilla is currently failing 04:37:11 dbaron: test is wrong 04:38:25 dbaron: has to due with url parsing bug 04:39:00 dbaron: hmmm, maybe the bug does exist 04:39:12 chrisl browbeats dbaron 04:39:48 dbaron resists valiantly 04:40:35 anne: waiting for implementations to pass 04:41:37 http://www.w3.org/Style/CSS/Test/CSS3/Namespace/current/syntax-013.xml 04:41:44 is the test that fails in Gecko 04:45:48 correction: anne states duplication declarations are not conformant 04:46:11 chrisl: agreement? 04:46:32 resolution: duplicate namespace declarations make content nonconforming 04:47:10 topic: june f2f agenda 04:47:38 howcome: multi-col should be into LC 04:47:45 stevel: gcpm? 04:48:10 howcome: yes 04:48:22 stevel: how to reflect font features in css 04:48:31 stevel: test suites? 04:48:58 ee: working on draft of paged media 04:49:18 anne: what's issue with ms test suite? 04:49:29 ee: getting them reviewed is the key 04:50:21 howcome: can we get a demo of the test suite? 04:50:58 http://wiki.csswg.org/test/css2.1/review 04:51:08 ee showing tests suite 04:51:12 http://wiki.csswg.org/test/css2.1/review-checklist 04:51:55 fantasai: managing test checks via mailing list 04:52:26 http://test.csswg.org/source/CSS2.1-test-suite/incoming/microsoft/ 04:52:46 http://test.csswg.org/svn-view/CSS2.1-test-suite/incoming/microsoft/ 04:53:13 send mail to public-css-testsuite 04:53:46 ee: build system will hopefully index metadata for tests 04:53:59 ee: this should give us 04:54:11 ee: a better understanding which tests fail 04:54:54 howcome: who's server is this? 04:54:57 ee: hp 04:55:11 howcome asking about tests 04:56:00 howcome has joined #css 04:56:57 ee: svn view to figure out changes 04:57:18 ee: build script to turn xhtml to html versions 04:57:35 ee: after that i'll be tidying up, cron job to build 04:57:48 sylvain: manpower issue 04:58:18 discussion of format guidelines 04:59:42 EE: Arron has a set of scripts written in C# that do a lot of checks; the CSS group does not have these. 05:00:03 http://test.csswg.org/source/CSS2.1-test-suite/incoming/microsoft/Chapter_5/attribute-value-selector-004.xht not well formed 05:00:03 http://test.csswg.org/source/CSS2.1-test-suite/incoming/microsoft/Chapter_5/attribute-value-selector-004.xht 05:02:05 ee: lots of work to do 05:02:10 it would be easy to pass all files through an xml parser and see which ones are nwf so they can be fixed 05:02:13 ee: some tests not well-formed 05:02:52 howcome: so what's the process here? 05:03:19 anne: not clear when a test is approved 05:03:58 ee: someone needs to review each test and confirm to mailing list 05:04:06 http://lists.w3.org/Archives/Public/public-css-testsuite/2009Mar/ 05:04:30 howcome notes an error in markup 05:05:09 howcome: we follow standards... ;) 05:05:34 anne: reparsing a security issue 05:05:47 http://lists.w3.org/Archives/Public/public-css-testsuite/2009Mar/0002.html 05:06:46 sylvain: not sure when this is going to get completed 05:07:16 ee: so, um.. 05:07:34 ee: can make a wiki page with links to tests 05:07:52 ee: categorized by reviewed, not reviewed 05:08:02 discussion of review process 05:09:53 dbaron proposing a process 05:11:22 chrisl: check out test, confirm, twiddle tasty bits in test file, check back in 05:11:39 ee: need mailing list post 05:13:43 ee, dbaron discussing whether to check-in vs. posting to mailing list 05:14:20 ee: either way (patches or posting mail) is fine 05:14:43 melinda has joined #CSS 05:14:52 hey Melinda, just in time. We're discussing tests 05:15:54 ee: web interface for checking who has done what 05:16:09 heh heh, I was just going to capture the log as I toddled off to bed, but I can hang out for a bit. 05:16:22 ee: aaron wants to know when things change 05:17:07 stevel: automatic checkin notification? 05:17:38 dbaron: can we just do this via email? 05:17:51 stevel: but for others outside the wg 05:18:03 anne: for html5, several lists track changes 05:18:32 dbaron: uh, not just tools, actually doing the work... 05:20:09 ee: need meta element to note review 05:20:16 discussion of meta format 05:20:21 for review comment 05:21:14 stevel making a point 05:21:32 anne commenting 05:21:52 chrisl: i'm taking fonts chapter 05:22:28 more process discussion 05:22:41 ee about to speak 05:23:01 everyone quiet in anticipation 05:23:24 ee fiddling with keyboard 05:24:18 chrisl assigning homework 05:25:56 anne: chapter 4 syntax 05:26:02 dbaron: 5, selectors 05:26:11 chris: 15, fonts 05:26:37 howcome: dbaron, 12 05:26:45 stevel: 14 05:27:01 Uh, what homework is being assigned...? 05:27:06 howcome: 13 05:27:24 MikeSmith getting suckered into more work... 05:27:46 MikeSmith: appendix d 05:28:15 MikeSmith: media types 7 05:28:39 ee: i'm doing bidi tests 05:28:56 anne: hixie, 8-11 05:28:57 ? 05:29:13 howcome: bert, box model? 05:30:03 me: fonts, 15 05:30:10 howcome gives me a hard time 05:31:07 ee: i updated review process 05:31:24 with metatag format 05:32:07 http://wiki.csswg.org/test/css2.1/review 05:32:13 http://wiki.csswg.org/test/css2.1/format#reviewer 05:33:02 ee demoing svn details 05:33:28 howcome: melinda should do? 05:33:40 svn co http://test.csswg.org/svn/ myfavdirectoryname to check out everything 05:33:42 Melinda: we're signing up for CSS 2.1 tests to review 05:33:53 Yeah! 05:33:58 Melinda: each person in the room has picked a chapter 05:34:17 melinda is working on print, 13 05:34:27 ee: ^ 05:34:39 But I need a reviewer (thanks, Hakon!) 05:35:35 http://www.w3.org/Style/CSS/Test/guidelines.html 05:35:38 Recommended reading ^ 05:35:51 http://wiki.csswg.org/test/css2.1/review 05:37:05 i'll do chap. 6 05:37:49 Melinda: is it possible to put the tests on the same server as where the MS tests are? Or, should I look somewhere else? 05:38:23 http://test.csswg.org/ 05:38:26 I've been moving some to the csswg site, and I'll try to get the rest moved over next week. 05:38:55 melinda: Are you putting them in http://test.csswg.org/svn/submitted/css2.1/page/ ? 05:39:04 yes 05:39:35 stevel: asks for pointers on the wiki page 05:40:47 chrisl: agenda item, go over tests in june f2f 05:41:14 chrisl: complete tests by start of april? 05:41:49 chrisl: review test suite at beginning of april 05:43:12 the msft test suite site also allows you to navigate your tests by properties/rules if that helps : http://samples.msdn.microsoft.com/ietestcenter/css.htm 05:46:39 rel="reviewer" is now provisionally registered: http://wiki.whatwg.org/wiki/RelExtensions 05:49:50 sylvaing has joined #css 06:04:19 sylvaing has joined #css 06:05:07 ScribeNick: dbaron 06:06:32 Elika: Do we want to publish a new CR of namespaces with the one change? 06:06:35 Chris: No point holding it back. 06:06:58 ACTION: fantasai publish namespaces CR and Selectors LC 06:06:58