IRC log of css on 2011-07-25
Timestamps are in UTC.
- 16:06:00 [RRSAgent]
- RRSAgent has joined #css
- 16:06:00 [RRSAgent]
- logging to http://www.w3.org/2011/07/25-css-irc
- 16:06:07 [Zakim]
- Zakim has joined #css
- 16:06:41 [plinss_]
- rrsagent, make logs public
- 16:06:47 [bradk]
- Is it david.baron on skype then?
- 16:06:57 [dino]
- dino has joined #css
- 16:07:04 [smfr]
- smfr has joined #css
- 16:07:04 [nimbupani]
- ScribeNick: nimbu
- 16:08:02 [dbaron]
- bradk, having trouble getting the thing to work
- 16:08:10 [florian]
- florian has joined #css
- 16:08:16 [nimbupani]
- Topic: CSSOM
- 16:08:32 [bradk]
- dbaron, OK, will wait for further instructions.
- 16:08:41 [nimbupani]
- glazou: report whats going on with CSSOM, what are the stuff we need to add, change.
- 16:08:59 [shans]
- shans has joined #css
- 16:09:21 [nimbupani]
- glazou: anne, tell us where you are right now, what you need to add, change, etc
- 16:09:37 [dbaron]
- bradk, see /query
- 16:09:43 [kimberlyblessing]
- kimberlyblessing has joined #css
- 16:09:47 [nimbupani]
- anne: the OM covers how to serialize mediaq, selectors, stylesheets in general
- 16:09:59 [sylvaing]
- http://dev.w3.org/csswg/cssom/
- 16:10:06 [nimbupani]
- anne: it has alt stylesheet api so people can control stylesheet sets from scripts.
- 16:10:17 [JohnJansen]
- JohnJansen has joined #css
- 16:10:23 [nimbupani]
- anne: it def processing requirements for the link header so you can associate stylesheets with documents via http
- 16:10:35 [nimbupani]
- anne: it then defines the "real" OM all the stylesheet rules and their apis
- 16:10:43 [nimbupani]
- anne: e.g. @import, @media, font-face, etc
- 16:10:56 [nimbupani]
- anne: there is a new thing called the values api, so far OM has been string based.
- 16:11:20 [dbaron]
- does somebody with a Mac want to try running skype on their machine for brad?
- 16:11:29 [nimbupani]
- anne: in 2003 we obsoleted parts of DOM level2 style that were object based but never wrote a replacement, section 6.6 in CSSOM draft sketches out how Object based API would look
- 16:11:47 [nimbupani]
- anne: lacking: implementors need to look at it to poke holes in idea before we formally define it.
- 16:12:00 [nimbupani]
- anne: this was the case a year ago, so people have not gotten around to looking at it yet
- 16:12:08 [arronei]
- http://dev.w3.org/csswg/cssom/
- 16:12:30 [nimbu]
- anne: there is a priority of list of features, e.g. to get to results styles of elements.
- 16:12:50 [smfr]
- s/results/resolved
- 16:13:25 [konaya]
- konaya has joined #css
- 16:13:36 [nimbu]
- anne: I would like to not add too much features before current features are interoperably implemented.
- 16:13:57 [nimbu]
- anne: we need to find out a plan on how to, since css is fairly modularized, how we do the same for the OM
- 16:14:13 [nimbu]
- anne: if we get interfaces for components like callers, it would be best if they are defined in their respective drafts.
- 16:14:22 [nimbu]
- anne: it would mean those drafts would ahve some dependency on the OM
- 16:14:35 [nimbu]
- fantasai: could we split out the serialization into its own module
- 16:14:54 [nimbu]
- smfr: for css transforms we need to expose an api for matrix transformations.
- 16:15:07 [stearns]
- s/callers/colors
- 16:15:11 [nimbu]
- anne: the color module needs to expose an api.
- 16:15:23 [nimbu]
- anne: its not just serialisation but the whole…
- 16:15:40 [hober]
- good morning
- 16:15:46 [nimbu]
- anne: each draft that defines a "bracketed new thing", needs to define an interface to it as well.
- 16:15:52 [nimbu]
- fantasai: two ways to do that
- 16:16:02 [nimbu]
- fantasai: have it paired with the module to define OM stuff
- 16:16:08 [vhardy]
- vhardy has joined #css
- 16:16:20 [nimbu]
- fantasai: for drafts that are advanced publish a .1 which includes OM stuff
- 16:16:43 [nimbu]
- fantasai: e.g. css color 3.1 or css color object model.
- 16:17:17 [nimbu]
- fantasai: in some cases the editor of the module can do the OM stuff but in other cases the editor might not have the expertise, so we need two editors or more.
- 16:17:29 [nimbu]
- florian: if we have 3 3.1 and 4 some people can get confused
- 16:17:35 [nimbu]
- anne: maybe we have it into 4.
- 16:17:53 [nimbu]
- anne: i think its best if its in same module as def and such will be tightly coupled
- 16:18:04 [nimbu]
- fantasai: going forward we can do that, but for some specs we cant do that
- 16:18:17 [nimbu]
- anne: maybe put them into 4 for current specs that you cant change.
- 16:18:29 [nimbu]
- florian: it might be quite a while before it becomes stable
- 16:18:49 [arno]
- arno has joined #css
- 16:18:58 [nimbu]
- fantasai: anther problem with OM is that it might hold back development of rest of the spec.
- 16:19:11 [nimbu]
- anne: we might have to separate them
- 16:19:17 [nimbu]
- fantasai: we can cross-link the documents
- 16:19:32 [nimbu]
- arno: we might have conformance problem too, as OM is not necessary for implementations
- 16:19:36 [szilles]
- szilles has joined #css
- 16:19:59 [nimbu]
- s/arno/ aaronei
- 16:20:28 [nimbu]
- fantasai: suggestion to go with separate modules but paired, if we want to merge we can merge them later.
- 16:20:35 [nimbu]
- mollydotcom: what does it look like.
- 16:21:37 [nimbu]
- vhardy: for specs like css3 regions which have a little bit of OM in them, keep them in the spec rather than splitting
- 16:22:13 [nimbu]
- glazou: apart from css values, this doc is specification of current implementation, what are the plans for the future?
- 16:22:26 [nimbu]
- glazou: i am a heavy user of OM it is not practical at this point
- 16:22:35 [nimbu]
- anne: it needs improvements we need more tests for existing stuff first.
- 16:22:57 [nimbu]
- anne: it has better def for DOM level 2 styles, which lacked a lot of details
- 16:23:12 [florian]
- s/improvements we/improvements but we/
- 16:23:37 [nimbu]
- glazou: everybody knows that OM is resolved, its probably time to push for new things.
- 16:23:56 [nimbu]
- anne: making it more interoperably and finding more stuff …
- 16:24:33 [nimbu]
- glazou: if we do stabilization for the next 3 years, it requires commitment from editors, implementors. it would be more than 2 years before we implement the new stuff
- 16:24:48 [nimbu]
- glazou: we are not able to serialise stylesheets, css text.
- 16:24:58 [nimbu]
- anne: those are minor features its not smthing developers really want
- 16:25:02 [nimbu]
- glazou: they are not minor
- 16:25:11 [dbaron]
- s/css text/cssText/
- 16:25:22 [nimbu]
- arronei: they maybe minor for devs, but not for web browsers
- 16:25:43 [nimbu]
- glazou: people who are writing editors need to play with these things every day. we need to improve the OM for new services on the web
- 16:25:58 [szilles]
- szilles has joined #css
- 16:26:04 [nimbu]
- anne: we have not defined serialization for most of the things. Just adding new things does not seem to be a good way forward.
- 16:26:21 [nimbu]
- glazou: i know, we should start collecting new things we are doing to the OM, and what are the expectations from web authors.
- 16:26:29 [nimbu]
- anne: sure getting feedback from web authors would be great
- 16:26:57 [shepazu]
- (another interesting thing might be font/text metrics: http://lists.w3.org/Archives/Public/public-svg-wg/2010OctDec/0004.html)
- 16:27:04 [nimbu]
- anne: better value api, once that is there, they want to have access to resolved vaules, there are various features browsers have added for their debugging tools, figure out which property calls current styles.
- 16:27:11 [nimbu]
- glazou: what are the rules that are being called at a given time
- 16:27:42 [nimbu]
- shepazu: in svg wg we have thought about adding font metric api, i thought it might be interesting
- 16:27:45 [nimbu]
- anne: like in canvas?
- 16:27:58 [nimbu]
- shepazu: the thing in canvas might be suitable why not generalize for genereal users
- 16:28:13 [nimbu]
- dino: you do not exactly know thats the font being used.
- 16:28:33 [nimbu]
- dino: all the canvas does is how long the piece of text was, does not tell you the ascenders,
- 16:28:41 [nimbu]
- szilles: or where the baselines are
- 16:28:48 [nimbu]
- shepazu: the things the canvas one does is useful
- 16:29:03 [nimbu]
- shepazu: i suspect chris saying web fonts working group will be interested in
- 16:29:23 [nimbu]
- anne: there have been requests for font loading api, but we have not figured out right way to do it
- 16:29:28 [tantek]
- tantek has joined #css
- 16:29:33 [nimbu]
- vhardy: how do you capture requirements? do u have a place for that?
- 16:29:48 [nimbu]
- anne: there is a page in whatwg wiki. we should probably create a place in csswg
- 16:30:15 [nimbu]
- glazou: long ago we had a list of suggestions for css3 we can have the list of suggestions for cssom
- 16:30:17 [anne]
- http://wiki.csswg.org/spec/cssom
- 16:30:31 [nimbu]
- sylvaing: do we have use cases?
- 16:30:43 [nimbu]
- anne: it does not have much yet.
- 16:30:57 [nimbu]
- florian: anne says stabilize, glazou says new features
- 16:31:04 [nimbu]
- glazou: not new features now, but soon
- 16:31:17 [nimbu]
- anne: i am also curious what implementors think, last year, we discussed this
- 16:31:30 [nimbu]
- anne: TabAtkins said google was planning on experimenting on values api
- 16:31:35 [nimbu]
- TabAtkins: we are still planning on it
- 16:31:46 [nimbu]
- anne: it kinda depends on implementors as well how fast it will move forward
- 16:32:07 [nimbu]
- anne: if there is not sufficient interest there is no point in adding a bunch of things and spend time on this if there are other things more interesting
- 16:32:23 [szilles]
- szilles has joined #css
- 16:32:32 [kojiishi]
- kojiishi has joined #css
- 16:32:45 [nimbu]
- vhardy: would it make sense to get a list of features and get feedback from community. we have some input and prioritize which are important
- 16:33:19 [nimbu]
- glazou: microsoft has been doing online applications using formats that are not based on w3c standards.
- 16:33:35 [ChrisL]
- ChrisL has joined #css
- 16:33:40 [nimbu]
- glazou: the cost for that [feedback] is low
- 16:33:46 [nimbu]
- glazou: anything else to decide o OM
- 16:34:16 [ChrisL]
- rrsagent, here
- 16:34:16 [RRSAgent]
- See http://www.w3.org/2011/07/25-css-irc#T16-34-16
- 16:34:28 [nimbu]
- vhardy: first set of extensions and ask for feedback?
- 16:34:33 [nimbu]
- glazou: no just ask for feedback
- 16:34:45 [nimbu]
- shepazu: i would like people to see action to review it
- 16:34:53 [nimbu]
- anne: that is more problematic than community input
- 16:35:26 [nimbu]
- anne: we have got a few feedback like features like resolved value, authors want access to browser's proprietary extensions for their debugging tools
- 16:35:33 [nimbu]
- anne: problem is getting implementors attention.
- 16:35:44 [nimbu]
- anne: i work for an implementor but I dont think opera can really drive this.
- 16:36:04 [nimbu]
- glazou: user feedback can help. the feedback for css3 created traction.
- 16:37:22 [tantek]
- anne, are there particular web applications that would be helped (e.g. made faster, or new higher fidelity features) by more CSSOM features?
- 16:37:43 [nimbu]
- arronei: you have unique knowledge in the room, so your feedback would be useful
- 16:38:10 [nimbu]
- glazou: cost of a single missing feature of OM can be about 2000 likes of code. as you start duplicating the cost becomes prohibitive. 2 or 3 lines in c++ for 1000 lines of js
- 16:38:39 [nimbu]
- ACTION: glazou collect feedback from webdev community for CSSOM and put it on the wiki http://wiki.csswg.org/spec/cssom
- 16:38:39 [trackbot]
- Created ACTION-349 - Collect feedback from webdev community for CSSOM and put it on the wiki http://wiki.csswg.org/spec/cssom [on Daniel Glazman - due 2011-08-01].
- 16:38:58 [alexmog]
- alexmog has joined #css
- 16:39:13 [nimbu]
- anne: can we do one more item
- 16:39:19 [arno]
- let's get input from framework vendors as well (jQuery, Sencha, etc…)
- 16:39:31 [nimbu]
- anne: there is also the view module? can we publish it once more, so there is a link to the editor's draft
- 16:39:36 [nimbu]
- fantasai: its already in the module template
- 16:39:53 [nimbu]
- anne: since it is not published already, so it does not have that link
- 16:40:06 [nimbu]
- szilles: he says the view module does not have the new template
- 16:40:12 [nimbu]
- fantasai: we can publish then
- 16:40:27 [nimbu]
- RESOLVED: Publish the CSSOM View module
- 16:41:20 [nimbu]
- Topic: CSS Regions and CSS Exclusions Floats
- 16:41:25 [tantek]
- possible classes of use-cases to drive CSSOM evolution: web app games, where greater responsiveness (speed) and higher fidelity experience are both desired.
- 16:41:34 [nimbu]
- vhardy: last meeting we decided to name it CSS Floats not Exclusions
- 16:41:49 [nimbu]
- vhardy: there is a ED that has resolution from last f2f there is a wd that was published after f2f as well
- 16:42:03 [nimbu]
- vhardy: wiki has a bunch of issues and usecases that we will go over
- 16:42:14 [nimbu]
- alan hooked up the test harness and the spec references it now
- 16:42:40 [nimbu]
- vhardy: next step is to go over the bunch of issues we have and pub together a new draft
- 16:42:53 [nimbu]
- vhardy: for css exclusions we have the original fraft same as kyoto
- 16:43:06 [nimbu]
- vhardy: ms sent a proposal and glazou sent one over email
- 16:43:32 [nimbu]
- vhardy: we tried to converge on the proposals and the next step is to resolve issue sand produce new ED to propose as WD
- 16:43:46 [ChrisL]
- rrsagent, this meeting spans midnight
- 16:43:51 [szilles]
- s/sand/and/
- 16:44:35 [smfr]
- http://wiki.csswg.org/spec/css3-regions
- 16:44:57 [nimbu]
- vhardy: we discussed there was no formal resolution in previous meeting. there was a q on the amiling list if css regions should allow copying from element into flow instead of moving content into flow. the editors agree this is something we should move to future release might be interesting but more complicated, so proposal is not to do it now and see if there is consensus in not adding this feature at this time
- 16:45:04 [nimbu]
- smfr: def agree on that
- 16:45:20 [nimbu]
- RESOLVED: Issue 4: Copying Flow Content is not something we are looking at right now.
- 16:45:25 [lhnz]
- lhnz has joined #css
- 16:45:55 [nimbu]
- vhardy: next issue is CSSOM issue. we dont have absility to find out which element is in a named flow
- 16:46:01 [nimbu]
- vhardy: there is no way to find that out right now.
- 16:46:23 [nimbu]
- vhardy: our proposal is to get an action item for alexmog and me to add to the next draft unless someone thinks this feature should not be in the draft
- 16:46:34 [nimbu]
- florian: are u going to do this programmatically or declaratively
- 16:46:42 [nimbu]
- smfr: that could be regions plural right
- 16:46:53 [nimbu]
- vhardy: programatically, plural
- 16:47:38 [nimbu]
- glazou: this will be useful for debugging tools
- 16:47:57 [nimbu]
- ACTION: vhardy alexmog to add an api to find out which element is in a named flow
- 16:47:57 [trackbot]
- Created ACTION-350 - Alexmog to add an api to find out which element is in a named flow [on Vincent Hardy - due 2011-08-01].
- 16:48:33 [nimbu]
- vhardy: in css should you be able to set display on grid cell?
- 16:48:43 [nimbu]
- vhardy: how do we integrate with css grid
- 16:48:54 [nimbu]
- vhardy: one is to have a way to adress a gridcell to make it a region
- 16:48:56 [nimbu]
- Hixie:
- 16:49:17 [nimbu]
- vhardy: second, add an element as a child to the grid and declare it a region
- 16:49:46 [nimbu]
- alexmog: it would be interesting to have pseud elements for other virtual things like grid regions for really advanced scenarios.
- 16:50:54 [nimbu]
- alexmog: in pseudo elements you cant get a OM you need an entirely diff mechanism
- 16:51:14 [nimbu]
- alexmog: it has to be a parallel model.
- 16:51:15 [bradk]
- So one HTML element to create a grid slot, and another inside it to hold a flow?
- 16:51:26 [nimbu]
- szilles: there needs to be a CSSOM module that gives us back the faciilities
- 16:51:29 [JohnJansen]
- correct Brad
- 16:52:16 [nimbu]
- RESOLVED: Change the grid to integrate with css regions through regular elements
- 16:52:56 [nimbu]
- vhardy: if you want to grab content from flow, you use content property with from-flow.
- 16:52:56 [JohnJansen]
- (in other words, not do anything special for grid)
- 16:53:22 [nimbu]
- vhardy: there was an alt proposal you should make two properties parallel and it should be separate from content use float from and use the flow name to get the content from the flow
- 16:53:36 [nimbu]
- vhardy: should we make it through two parallel properties
- 16:54:00 [nimbu]
- vhardy: after we have had more discussions we think this is more clear use flow-form: <flow name>
- 16:54:09 [nimbu]
- smfr: would flow-from work on pseudo elemtn before / after
- 16:54:13 [nimbu]
- alexmog: vhardy yes
- 16:54:36 [nimbu]
- dbaron: i am uncomfortable making an additional property here. as it seems to do same thing as content.
- 16:55:08 [nimbu]
- alexmog: there are couple of considerations, content has a lot of baggage, there is content in css3 content, which is not where the spec is going.
- 16:55:23 [nimbu]
- alexmog: the flow-from applies to anything into a region.
- 16:55:38 [nimbu]
- alexmog: the content is somewhat different as it cant seem like it changes the nature of what it is applied to
- 16:56:08 [nimbu]
- fantasai: does that mean i cant have a table that accepts content from the region
- 16:56:26 [nimbu]
- vhardy: that is one of the issues, conceptually region is smthing that redirects content from somewhere else into this element.
- 16:56:33 [nimbu]
- s/vhardy/alexmog
- 16:57:03 [nimbu]
- fantasai: lets say i ahve a region and I want it to be a table row and I grab the contents of the table row and put it into regions what happens
- 16:57:09 [nimbu]
- alexmog: there is nothing in regions that wont work.
- 16:57:30 [nimbu]
- fantasai: so thats not display-inside region you are not overriding display inside just putting contents into it.
- 16:57:41 [nimbu]
- vhardy: we have a proposal to limit regions to display-inside block.
- 16:57:52 [nimbu]
- vhardy: otherwise it will become too complicated.
- 16:58:09 [nimbu]
- vhardy: we can resume discussion once we reach that.
- 16:58:25 [nimbu]
- alexmog: i dont feel very strongly about flow-from w.r.t content
- 16:58:38 [nimbu]
- alexmog: content has a history that does not make it designed for this.
- 16:58:51 [nimbu]
- arno: how would flow-from property how would it interact with content
- 16:59:00 [nimbu]
- alexmog: content would be overriden
- 16:59:30 [nimbu]
- vhardy: content is grabbing content from an element but flow-from is grabbing a segment not whole element. so its different
- 16:59:48 [nimbu]
- plinss: thats what you get from from-flow so i dont see why thats a problem, you just need to adapt that notion to the content.
- 17:00:08 [nimbu]
- alexmog: from the content property you can figure out what content is, from region that is not the same.
- 17:00:21 [nimbu]
- fantasai: having one property that overrides another causes problems with cascade
- 17:00:38 [nimbu]
- fantasai: which one wins? we should use the cascade as thats what it is for.
- 17:00:53 [nimbu]
- vhardy: other people felt it was good to add a property.
- 17:01:34 [nimbu]
- alexmog: content applies to inline flow-from doesn't. different compatibility. I am not sure it would always be that way. I am not sure if flow-from would always override.
- 17:01:46 [nimbu]
- vhardy: content would only apply to before and after in 2.1
- 17:01:52 [nimbu]
- fantasai: in css3 it would apply to all.
- 17:02:02 [nimbu]
- plinss: that has been the plan for 10/11 years at least
- 17:02:11 [nimbu]
- vhardy: the draft has the way you [fantasai] like it
- 17:02:21 [nimbu]
- vhardy: if majority feel we are good with content, I am okay with that.
- 17:02:27 [nimbu]
- alexmog: i dont have really strong feelings.
- 17:02:57 [nimbu]
- alexmog: having seperat property makes writing spec easier. Understanding easier. Not redefine content
- 17:03:30 [nimbu]
- RESOLVED: Stick to content property and content: from-flow(<flow-name>)
- 17:03:57 [nimbu]
- vhardy: currently to put smthing in flow you use flow: <flow-name>
- 17:04:02 [ChrisL]
- does tha mean content can have internal structure, now?
- 17:04:28 [nimbu]
- vhardy: alexmog pointed out flow could be short hand, so should we use flow-into and reserve flow for shorthand
- 17:04:29 [ChrisL]
- or is it flattened text?
- 17:04:49 [dbaron]
- "rename flow into flow-into" needs to be written down to be comprehensible
- 17:05:07 [nimbu]
- smfr: flow-into does not work for me, how about flow-name
- 17:05:44 [nimbu]
- szilles it is doing two things naming and adding content to flow
- 17:06:00 [nimbu]
- smfr: which property is canonical in terms of naming of the flow
- 17:06:13 [nimbu]
- vhardy: the flows get their name from the first piece of content that is moved into the flow
- 17:06:21 [nimbu]
- smfr: it is a bit too magic for me
- 17:06:32 [nimbu]
- vhardy: we iterated over that and we ended ups ettling on this way to do it.
- 17:06:37 [glazou]
- CSS _is_ magic anyway :-)
- 17:06:40 [nimbu]
- smfr: this is more like it is belonging to a flow.
- 17:06:45 [nimbu]
- smfr: make this part of a flow.
- 17:07:08 [bradk]
- I prefer just 'flow:'
- 17:07:12 [nimbu]
- alexmog: we tried things like flow-source and flow-target and those were very confusing.
- 17:07:32 [nimbu]
- glazou: we need ar esolution to move from flow to smthing
- 17:07:50 [nimbu]
- szilleswhat it actually is flow-out-to rather than into as it is taking it out of normal flow
- 17:08:01 [nimbu]
- RESOLVED: change flow from flow-smthing
- 17:08:37 [nimbu]
- vhardy: alexmog asked for renaming from-flow to flow-from
- 17:08:57 [nimbu]
- jRESOLVED: rename content: from-flow to content: flow-from
- 17:09:07 [nimbu]
- RESOLVED: rename content: from-flow to content: flow-from
- 17:09:22 [nimbu]
- dbaron: it makes more sense to me the other way
- 17:09:42 [nimbu]
- alexmog: we have border-top-right
- 17:10:25 [nimbu]
- vhardy: auto sizing of regions. this has been called as intrinsic size. i am not sure if all the history we have we are using intrinsic as a term in our discussions
- 17:10:33 [nimbu]
- vhardy: intrinsic has only been used for replaced content
- 17:10:48 [nimbu]
- fantasai: there is terminology for that in writing modes draft.
- 17:11:02 [fantasai]
- http://www.w3.org/TR/css3-writing-modes/#intrinsic-sizing
- 17:11:15 [nimbu]
- vhardy: i tried to stay clear of intrinsic size coz of possible confusion
- 17:11:19 [szilles]
- szilles has joined #css
- 17:11:27 [nimbu]
- vhardy: if we have auto-sizing in 1 or more dimensions how does it work
- 17:11:53 [nimbu]
- vhardy: we had a discussion with hyatt on mailing list, alexmog brought up more questions. this email is a summary of discussions at tha time
- 17:12:07 [nimbu]
- vhardy: the proposal was do we actually need to say anything about auto-sizing
- 17:12:31 [nimbu]
- vhardy: the main model i have been using for regions, the results should be identical if you have broken down regions into bits and pieces and moved them into regions
- 17:12:46 [nimbu]
- vhardy: so if I follow that model we dont need smthing
- 17:12:57 [nimbu]
- dbaron: the problem with that model is that the slicing depends on the size.
- 17:13:27 [nimbu]
- alan: if you have not specified height there wouldnt be any.
- 17:13:39 [nimbu]
- alexmog: paginating content in terms of flexible size is really complicated.
- 17:14:11 [nimbu]
- szilles: the two dimensions inline and block the rules are slightly different. in inline you would want it to be like a block, in block progression dimension you are going to use breaks to control…
- 17:14:39 [nimbu]
- alexmog: even more complicated is when width is undefined we can come up with a width measurement, the final layout is difficult to do if you dont know what the width to be.
- 17:15:08 [nimbu]
- alexmog: e.g. for float: right you need to figure out what the size is if it should be cleared or not.
- 17:15:19 [nimbu]
- dbaron: whats the proposal?
- 17:16:09 [nimbu]
- vhardy: start with region 1 and then put all content in region 1. if anything that does not fit, you have a first segment and reminder will flow into next regions. if you have regions with undefined height, it just gets all the content.
- 17:16:44 [nimbu]
- vhardy: at any point it would be like rest of the flow will be in the content.
- 17:16:58 [nimbu]
- smfr: say u have a bunch of ps and a 1000 px wide div.
- 17:17:15 [nimbu]
- smfr: the first region is flexible width wise
- 17:17:25 [nimbu]
- smfr: sounds like you would make first region 1000px wide
- 17:17:40 [nimbu]
- dbaron: what is the proposal for what intrinsic size would be?
- 17:17:54 [nimbu]
- dbaron: intrinsic size is not a function of layout
- 17:18:05 [nimbu]
- alexmog: intrinsic is not the right term here.
- 17:18:13 [hober]
- If the intrinsic width of a region is 0, the designer wins because their layout "fails fast" & they immediately know they need to specify a width
- 17:18:15 [nimbu]
- alan: the proposal is intrinsic size does not apply
- 17:19:03 [nimbu]
- alexmog: we should not be using the word intrinsic here.
- 17:19:28 [nimbu]
- dbaron: i sounds like you are talking about what the sizes of the regions are going to be. that is still a separate notion from intrinstic size
- 17:19:40 [nimbu]
- vhardy: intrinsic is used for replaced content 2.1
- 17:19:49 [nimbu]
- dbaron: we used for table,
- 17:19:51 [arronei_]
- arronei_ has joined #css
- 17:19:57 [nimbu]
- dbaron: i am talking about preferred width and min width
- 17:20:10 [nimbu]
- fantasai: we call it two diff things based on if we are on tables chapter or not
- 17:20:33 [nimbu]
- szilles: there are two situations, 1. normal flow 2. out of normal flow. the rules for width and height calc are diff in those.
- 17:20:47 [nimbu]
- szilles: u use intrinsic when you are out of normal flow and simple calc in normal flow
- 17:20:59 [nimbu]
- szilles: regions behave in exact same way.
- 17:21:38 [nimbu]
- alexmog: 3 options: 1. if region size is undefined it is zero. that option works very well the standard box model rules apply regardless of where region is.
- 17:21:52 [nimbu]
- the size will be calc using formula for box size.
- 17:22:08 [nimbu]
- problem is when accidentally size is not set on a region it has no visibility and it is difficult to understand where it is.
- 17:22:13 [hober]
- that's a feature, not a bug
- 17:22:40 [nimbu]
- 2. if it has a specific intrinsic size, 3000 by 850 it would be seful as you might have forgotten to set a size and then you can see it and then make it right size.
- 17:22:49 [bradk]
- maybe we need 'position:flow' or 'position:flow(ident)', and then all flows are non-floating blocks.
- 17:23:06 [dbaron]
- s/3000 by 850/300 by 150/
- 17:23:16 [nimbu]
- normal box models wont apply.
- 17:23:20 [nimbu]
- and thats not helpful.
- 17:23:28 [szilles]
- szilles has joined #css
- 17:24:15 [nimbu]
- alexmog: 3. size to content. we have to figure out what is going to fit into this region without exceeding max width max height if they are available.
- 17:24:34 [nimbu]
- alexmog: this is very complicated to implement, it would require more than 1 layout pass
- 17:25:05 [nimbu]
- alexmog: option 4. use normal box model calculations and if that ends up as still undefined with height, then use 300x850 intrinsic size.
- 17:25:14 [nimbu]
- vhardy: it is like replaced content right.
- 17:25:22 [nimbu]
- alexmog: it is not like the second option replaced content.
- 17:26:07 [nimbu]
- alexmog: the 4th option would be an implementation of normal box sizing where the box size is not considered at the point where it is on or not but at the end when it is resolved or not.
- 17:26:10 [tantek_]
- tantek_ has joined #css
- 17:26:25 [hober]
- s/850/150/
- 17:26:33 [nimbu]
- thanks :)
- 17:27:41 [nimbu]
- i would expect devs to slice area into bunch of rectangles and use grid or flex box and then add content to the.
- 17:28:17 [nimbu]
- vhardy: the first proposal, the width and height would be set to 0 if you do not set height or width.
- 17:28:34 [nimbu]
- alexmog: this is a strong arg for default size to not be strict width/height
- 17:28:50 [nimbu]
- alexmog: if this region is going to have content in there, it wont have a size the flexbox and grid would give it.
- 17:29:04 [nimbu]
- alexmog: for the pruposes of flex or grid it has to have a default size of 0.
- 17:29:10 [nimbu]
- alexmog: then grid would give it size
- 17:31:09 [nimbu]
- ???: for e.g. auto width should be as wide as containing block so edges touch on normal flow. those rules would be ignored as per proposal 1, and width would be set to 0.
- 17:31:59 [nimbu]
- ???: the containers are typically well respected, we dont set to default size or width 0. so, which one is it that you are trying to achieve here.
- 17:32:10 [nimbu]
- vhardy: the general question is what happens with auto widths and heights on regions
- 17:32:46 [nimbu]
- vhardy: the point that smfr made is that if you layout everything then your widths maybe off.
- 17:33:10 [nimbu]
- vhardy: your usecase would naturally work, intrinsic height to default to 0 it is hard to make it work.
- 17:33:50 [nimbu]
- florian: in general resolving to 0 is smthing simple but not useful
- 17:34:11 [nimbu]
- mollydotcom: it is established and understood in devs, but now we are changing expectation.
- 17:34:16 [sylvaing]
- s/???/rossen
- 17:34:45 [nimbu]
- alexmog: 0 is logical, if a region that is part of a chain of regions if it does not have any natural content, it creates 0 new rules for sizing it.
- 17:35:00 [nimbu]
- alexmog: it is a good default.
- 17:35:16 [nimbu]
- mollydotcom: default to 0 sounds to be very different from what you are saying
- 17:35:49 [nimbu]
- vhardy: i propose we actually tab this discussion and go back. the default behaviour is not useful, we need to consider usecases.
- 17:36:18 [nimbu]
- phil: I thought it worked with any kind of layout, some proposals sacrifice the notion that how the child layout will make use of available space given by parent layout
- 17:36:30 [mmielke]
- mmielke has joined #css
- 17:37:04 [nimbu]
- dbaron: vhardy's statement intrinsic size is 0 then size is 0 is incorrect
- 17:37:12 [nimbu]
- plenty cases otherwise
- 17:37:45 [nimbu]
- dbaron: for intrinsic widths, the obv sol do the intrinsic width computation do over entire contents of the flow.
- 17:38:13 [nimbu]
- dbaron: what we have not specified we have not specified intrinsic heights in block progression dimension. flexbox and this (some degree) depends on that concept.
- 17:38:43 [nimbu]
- dbaron: problem: they are a function of width not entirely intrinsic. 1 concept that is entirely intrinsic and another that is a function of width
- 17:39:21 [nimbu]
- dbaron: i can imagine a situation where you want to build heights in region by starting either from 1st or last. splitting at breaks and using as many regions from beginning or end to build some sort of intrinsics where you need them. It might give you useful behaviour for a bunch of htem
- 17:39:31 [nimbu]
- dbaron: there is a bunch of issues being conflated.
- 17:39:44 [nimbu]
- dbaron: we need def for intrinsic sizes and we need to specify what actual resolved sizes for regions are.
- 17:40:18 [nimbu]
- szilles i heard you say the cal of size algorithm it does occasionally use intrinsic size but not necessarily use intrinsic size. That is critical to making it work, and we have a def of intrinsic size.
- 17:40:34 [cjon]
- cjon has joined #css
- 17:41:10 [nimbu]
- alexmog: if whereever in the cal we need the intrinsic size we consider all of the content or remainder of content from this to that. that makes it clear it might be expensive to calculate. it is really a fallback, so not going to be used all the time, so expensive should not be a problem there.
- 17:41:23 [nimbu]
- vhardy: that works for me
- 17:42:31 [nimbu]
- ACTION: vhardy Rework the issue of resolved sizes for regions and come up with an alt solution for the group along with use cases and examples
- 17:42:32 [trackbot]
- Created ACTION-351 - Rework the issue of resolved sizes for regions and come up with an alt solution for the group along with use cases and examples [on Vincent Hardy - due 2011-08-01].
- 17:43:23 [szilles]
- szilles has joined #css
- 17:44:07 [fantasai]
- <br type="coffee">
- 17:47:56 [tantek]
- tantek has joined #css
- 17:48:18 [tantek_]
- tantek_ has joined #css
- 17:50:05 [tantek_]
- tantek_ has joined #css
- 17:50:52 [nimbupani]
- nimbupani has joined #css
- 17:54:53 [miketaylr]
- miketaylr has joined #css
- 17:57:26 [JohnJansen1]
- JohnJansen1 has joined #css
- 17:57:29 [szilles]
- szilles has joined #css
- 17:58:45 [JohnJansen1]
- JohnJansen1 has left #css
- 17:59:10 [anne]
- anne has joined #css
- 17:59:13 [JohnJansen1]
- JohnJansen1 has joined #css
- 18:00:16 [JohnJansen1]
- JohnJansen1 is JohnJansen
- 18:00:57 [JohnJansen1]
- JohnJansen1 has left #css
- 18:01:48 [JohnJansen]
- JohnJansen has joined #css
- 18:04:30 [szilles]
- szilles has joined #css
- 18:04:38 [florian]
- florian has joined #css
- 18:05:12 [nimbupani]
- vhardy: region model, so the current spec has taken the approach that region is a general relation of an element relates to content
- 18:05:47 [plinss_]
- plinss_ has joined #css
- 18:06:33 [mmielke]
- mmielke has joined #css
- 18:06:38 [nimbupani]
- vhardy: instead of working with children as source of generated boxes you would layout, you can get set of elements from source which will be used with layout algorithm
- 18:07:27 [nimbupani]
- vhardy: looking at use cases, our thinking now is 1. they are like printed pages and lay it out from page box to page box. they are different viewport areas.
- 18:07:51 [nimbupani]
- vhardy: mixing regions that have different type, region thats a table, a flex box and put them in a chain does not seem to make sense.
- 18:08:05 [nimbupani]
- vhardy: as you would not know which kind of segment you get.
- 18:08:21 [nimbupani]
- vhardy: limit regions to apply to things that are display-inside: block
- 18:08:31 [nimbupani]
- it could be table-cell or anything that has display-inside: block.
- 18:08:57 [nimbupani]
- vhardy: proposal restriction on regions to contain display-inside: block
- 18:09:15 [nimbupani]
- fantasai: i can see reasons for display-inside of flexbox.
- 18:09:27 [nimbupani]
- dbaron: the current flexbox is different from our implementation.
- 18:09:41 [nimbupani]
- fantasai: i can see wanting to have two flexbox containers and have a flow on them.
- 18:10:01 [nimbupani]
- dbaron: the way flexbox spec works is the container is display: flexbox, the items in it is display: block
- 18:10:03 [mielke]
- mielke has joined #css
- 18:11:16 [nimbupani]
- vhardy: the idea is to simplify the implementation. not limit the future.
- 18:11:22 [nimbupani]
- vhardy: there are usecases for tables as well.
- 18:11:49 [nimbupani]
- vhardy: then we have to deal with mixing different float types which are much more complicated.
- 18:11:58 [nimbupani]
- fantasai: I am okay as long as we can expand it in that direction
- 18:12:24 [nimbupani]
- fantasai: I can see why you would want other display types
- 18:12:31 [nimbupani]
- alex: i want to keep the first version of spec simple.
- 18:12:51 [nimbupani]
- vhardy: what we are doing is not painting us in a corner. we should have a note saying in future we would extend it.
- 18:13:08 [nimbupani]
- RESOLVED: for this version we are limiting regions to be block containers
- 18:13:22 [nimbupani]
- ACTION: vhardy add a note to expand this in the future for different types of containers
- 18:13:22 [trackbot]
- Created ACTION-352 - Add a note to expand this in the future for different types of containers [on Vincent Hardy - due 2011-08-01].
- 18:14:03 [szilles]
- szilles has joined #css
- 18:14:11 [nimbupani]
- vhardy: region breaks. CSS 2.1 has page breaks that only work on paged media. css3 column breaks and page breaks.
- 18:14:32 [nimbupani]
- vhardy: page breaks slices a column into two, the next column box moves to next page.
- 18:14:46 [nimbupani]
- vhardy: how do you break content when you layout between regions
- 18:15:29 [nimbu]
- vhardy: those questions also arise for multicol
- 18:15:45 [nimbu]
- vhardy: if u have breaks in ur content and nesting what does it mean for heirarchy
- 18:16:18 [nimbu]
- vhardy: current spec: we will add a new type of break and thats what content will honor when it lays out content. alex objected to it so we came up with alt proposals
- 18:16:39 [nimbu]
- vhardy: instead of saying i need to have a new break, state which breaks you will honour
- 18:16:53 [nimbu]
- vhardy: as a region you state which breaks you honor
- 18:17:05 [nimbu]
- vhardy: what that means is content is break aware and container aware.
- 18:17:36 [nimbu]
- vhardy: the content can annotate myself, you can then have container types and associated break types.
- 18:18:10 [nimbu]
- alex: is this clear, or do we need a picture
- 18:18:20 [bradk]
- picture
- 18:18:43 [ChrisL]
- picture and photo
- 18:19:17 [nimbu]
- alex: using either column breaks or page breaks producing multicol layout with regions there is no way to define currently contents going from one column like region to another column like region if you have a columnbreak, but if you have a page break then you need to go to a new page with a new container
- 18:19:29 [nimbu]
- alex: we need to communicate this region is a column and another region is a page.
- 18:19:57 [bradk]
- so, column-break would break to next region if both regions are each one-column blocks?
- 18:20:14 [nimbu]
- alex: if we decide we dont need that kind of precise control in regions coz that content would be designed in smaller pieces, with one kind of break. we would gravitate to never need column/page break difference. but only one kind of break.
- 18:20:30 [bradk]
- unless a block is identified as a page somehow?
- 18:20:56 [nimbu]
- vhardy: one of the extensions is to generalise ideas and do named breaks
- 18:21:07 [nimbu]
- vhardy: my content should know what kind of container it is laid out on.
- 18:21:26 [nimbu]
- vhardy: another approach is if your content just wants to know where it should break.
- 18:21:43 [nimbu]
- vhardy: second approach content does not know about container types or what kind of breaks are being used.
- 18:22:03 [nimbu]
- vhardy: second one is a simplifying assumption and want feedback.
- 18:22:20 [nimbu]
- vhardy: 1. breaks are container agnostic 2. breaks are container aware.
- 18:22:33 [nimbu]
- smfr: there has to be a precedent here? what does page layout do.
- 18:22:54 [nimbu]
- alex: yes, they do different kinds of breaks. page break and section break.
- 18:23:01 [nimbu]
- vhardy: ur q is on existing tools?
- 18:23:05 [nimbu]
- smfr: existing tools
- 18:23:39 [nimbu]
- alex: columns are onlu possible as full-page columns or full section columns, so page break is always a column break. there is no such thing as breaking multi-col block …
- 18:23:52 [nimbu]
- alan: in indesign you have a column break, page break, frame-break
- 18:24:05 [nimbu]
- alex: whats the diff between frame/page break
- 18:24:12 [nimbu]
- alan: frame break is like a region
- 18:24:19 [nimbu]
- florian: can u have nested frames
- 18:24:25 [bradk]
- frame breaks can jump across several pages to get to next frame to, right?
- 18:24:32 [nimbu]
- alan: I think so, i do not know if it makes a difference in the break scenario.
- 18:24:40 [tantek]
- tantek has joined #css
- 18:24:51 [nimbu]
- vhardy: frames are more like regions
- 18:24:56 [nimbu]
- alan: yes
- 18:25:38 [nimbu]
- plinss: dont confuse column breaks with region breaks as column only breaks to another col not a region
- 18:25:53 [nimbu]
- plinss: i dont see why you should respect other kinds of break.
- 18:26:04 [nimbu]
- alex: so how do you know the region is in the next page?
- 18:26:10 [nimbu]
- plinss: from the pagination model.
- 18:26:38 [nimbu]
- alex: what is a page?
- 18:26:49 [nimbu]
- fantasai: a page in css is a page box
- 18:27:22 [nimbu]
- fantasai: in mozilla those are page boxes. particular kind of css box
- 18:27:36 [nimbu]
- alex: in IE they are also kind of a box.
- 18:28:18 [nimbu]
- fantasai: a page break requests a break between page boxes.
- 18:28:43 [nimbu]
- alex: i would like to be able to write page view in a standard way not just for print preview but also for page reading (like online magazine)
- 18:29:04 [nimbu]
- plinss: if you are doing with divs you are not doing anything with css box model.
- 18:29:29 [nimbu]
- alex: you can have multiple regions in every page. if you are going to do page break in content and requires it continue to next page, no way to determine what next page is.
- 18:30:00 [nimbu]
- florian: you need something that gets you to next page
- 18:30:28 [nimbu]
- florian: if we want to support that we need something like that.
- 18:31:01 [nimbu]
- alex: if some content that you place in to region has page-break: avoid, is that smthing regions will not be capable of satisfying, or would we have smthing different.
- 18:31:08 [fantasai]
- break-before: region(mycustompaginator)
- 18:31:10 [nimbupani]
- nimbupani has joined #css
- 18:31:28 [nimbupani]
- smfr: why dont u use normal paged media?
- 18:31:51 [nimbupani]
- alex: if nothing more adv is available we would consider every region is a page from point of view of breaking
- 18:32:02 [nimbu]
- szilles: all the things u have said requires a content change.
- 18:32:40 [nimbu]
- szilles: the author wants to present this content if it were a page in the display he is presenting to the user
- 18:32:57 [nimbu]
- vhardy: if you want something foo, then you say break for 'foo'
- 18:33:15 [karl]
- karl has joined #CSS
- 18:33:25 [hyatt]
- hyatt has joined #css
- 18:33:38 [nimbu]
- alex: i dont get why you are objecting. moz has internally a special element to represent …
- 18:33:48 [nimbu]
- fantasai: we have a page box we have smthing in the rendering tree that represents a page box
- 18:33:58 [nimbu]
- alex: as an author how do you make an element a page box
- 18:34:00 [nimbu]
- fantasai: we cant
- 18:34:30 [nimbu]
- alex: with regions we are creating an opportunity for author to use page navigating algorithm.
- 18:34:45 [nimbu]
- plinss: do u want headers and footers to appear on random elements?
- 18:35:19 [nimbu]
- dbaron: there is an assumption that you should be able to do what you described without changing the content. i dont think thats true
- 18:35:57 [nimbu]
- florian: u might want to break sometimes on certain page breaks and not on some others.
- 18:36:21 [nimbu]
- vhardy: thats the intended functionality is to have more than 1 kind of breaks.
- 18:36:48 [bradk]
- What about regions inside regions inside regions inside regions? You need to be able to indicate level
- 18:36:48 [nimbu]
- plinss: you really are talking about a region break to a higher level (in nested regions).
- 18:37:14 [nimbu]
- fantasai: e.g. break-before: region('named flow');
- 18:38:16 [bradk]
- You might not know the region you are in, but know that you want to break 2 levels up. break-before:region(-2)
- 18:38:18 [nimbu]
- vhardy: take an action to add the usecase alex is talking about and measure up the actions
- 18:38:36 [nimbu]
- alex: only region break is also generally okay.
- 18:40:10 [mollydotcom]
- mollydotcom has left #css
- 18:40:16 [nimbu]
- alan: before we got to pagination, it sounded like plinss was saying if we have col breaks and region breaks and i have flow going to regions that are not going to multicol elements you would prefer col break to not break to next region.
- 18:41:19 [nimbu]
- vhardy: if you got page break and not in paged media it does nothing
- 18:41:42 [nimbu]
- alan: if you have a region situation and u have a column break i prefer my text to go to next region from auth perspective each region is a single column
- 18:41:57 [nimbu]
- alan: we have usecases where regions to use multicol layout
- 18:42:19 [nimbu]
- alex: in multicol spec does it say in paged media each page should be considered a column?
- 18:42:38 [nimbu]
- ACTION: howcome to answer if paged media each page should be considered a column?
- 18:42:38 [trackbot]
- Created ACTION-353 - Answer if paged media each page should be considered a column? [on Håkon Wium Lie - due 2011-08-01].
- 18:43:03 [mollydotcom]
- mollydotcom has joined #css
- 18:43:12 [alexmog]
- alexmog has joined #css
- 18:43:30 [nimbu]
- vhardy: option1 seems to be off the table.
- 18:43:59 [nimbu]
- vhardy: we agree we need breaks that are typed.
- 18:44:14 [szilles]
- szilles has joined #css
- 18:44:24 [nimbu]
- plinss: if we have nested region flows we need ability to break a level.
- 18:46:12 [nimbu]
- plinss: the types of usecases that alex is taking about i think we should have to work on getting browsers to render a page in paginated mode.
- 18:46:48 [fantasai]
- There was a suggestion to add 'overflow-style: paged' a while ago.
- 18:46:51 [hober]
- Yes, some kind of explicit "I want paginated mode" switch is much cleaner
- 18:46:57 [bradk]
- or a section of a page in paginated mode
- 18:48:13 [nimbu]
- alex: we are making it possible to create a webpage that gives you the same experience as looking at a magazine. give it same kind of control including control of column breaks and page breaks. we would need to create new break types
- 18:48:25 [dsinger]
- dsinger has joined #css
- 18:49:00 [Martijnc]
- Martijnc has joined #css
- 18:49:07 [nimbu]
- alex: if we switch browser to switch to page mode, the ui browser provides to navigate pages is light years away from what designers would want in page navigation experience
- 18:49:51 [nimbu]
- RESOLVED: we need breaks that are specific to containers they are part of.
- 18:50:25 [nimbu]
- ACTION: vhardy come up with a proposal for breaks either a page-break or a general type of break or both and related use cases
- 18:50:25 [trackbot]
- Created ACTION-354 - Come up with a proposal for breaks either a page-break or a general type of break or both and related use cases [on Vincent Hardy - due 2011-08-01].
- 18:50:44 [nimbu]
- smfr: we are not happy with behavior for iframe objects in spec
- 18:50:54 [lhnz]
- lhnz has joined #css
- 18:51:06 [nimbu]
- smfr: issue 5
- 18:51:08 [hober]
- http://dev.w3.org/csswg/css3-regions/#the-flow-property
- 18:51:26 [nimbu]
- alex: it is not an issue any more. I added it in.
- 18:51:32 [hober]
- "For an <iframe>, an <object> or a <embed> element, the ‘flow’ property has a different behavior. The effect is similar to turning the ‘display’ property on the element to ‘none’ while moving the root element of the referenced document to the named flow."
- 18:52:34 [nimbu]
- alex: if we remove it, then it would be a non-standard thing like font-size-adjust. if it stays there, it is okay to have it optional. Another option there is to have an explicit way of saying for an element if the element that is added to flow or content added to flow.
- 18:52:45 [nimbu]
- alex: i like the second option as it is much more explicit.
- 18:53:05 [nimbu]
- smfr: you start introducing cross origin problems also q of what to do with script style elements and stuff
- 18:53:36 [nimbu]
- alex: the cross frame security does apply. We are not going to take content from iframe if its not in the same origin.
- 18:53:40 [nimbu]
- smfr: the spec should stay that.
- 18:53:52 [nimbu]
- dbaron: sounds a lot like seamless iframe stuff html5
- 18:54:04 [nimbu]
- dbaron: should we do that instead of redefining it
- 18:54:21 [nimbu]
- TabAtkins: seamless iframe becomes like an included
- 18:54:44 [nimbu]
- anne: the iframe height becomes height / width of the document. selectors bleed through.
- 18:54:50 [nimbu]
- alex: no default padding
- 18:55:18 [dbaron]
- http://dev.w3.org/html5/spec/the-iframe-element.html#attr-iframe-seamless
- 18:55:23 [hober]
- So in a region <iframe> would behave like replaced but <iframe seamless> would behave like alex has specced
- 18:55:24 [nimbu]
- ACTION: alexmog to look at seamless iframes and reconcile it with what we have in the spec. We should add the security consideration into iframe paragraph.
- 18:55:24 [trackbot]
- Created ACTION-355 - Look at seamless iframes and reconcile it with what we have in the spec. We should add the security consideration into iframe paragraph. [on Alex Mogilevsky - due 2011-08-01].
- 18:56:19 [bradk]
- link?
- 18:56:22 [nimbu]
- vhardy: we hve had diff proposals for floats and exclusions
- 18:56:28 [nimbu]
- vhardy: worked on a bunch of use cases.
- 18:56:43 [stearns]
- http://wiki.csswg.org/ideas/css3-floats-use-cases
- 18:56:50 [nimbu]
- vhardy: we have a model we would like to propose to the group
- 18:56:59 [bradk]
- thanks
- 18:57:27 [nimbu]
- vhardy: present this to be ready for editors draft that is a convergence of these two proposals.
- 18:58:17 [nimbu]
- ???: 2, 3,4 are essence of how we build the model and 1 is how we use it
- 18:58:28 [smfr]
- http://wiki.csswg.org/ideas/css3-floats
- 18:58:30 [arronei]
- s/???/rossen
- 18:58:54 [nimbu]
- rossen: how to create container model.
- 18:59:07 [nimbu]
- rossen: #3 how the shape gets affected by exclusion
- 18:59:28 [konaya]
- konaya has joined #css
- 18:59:50 [nimbu]
- rossen: in our case, we looked at a diff point of view, already have exclusions in css2 which are floats
- 19:00:39 [nimbu]
- rossen: overloading the float property was redundant. we still had to use wrap-mode in ms proposal to trigger an exclusion
- 19:01:04 [nimbu]
- rossen: in our case the wrap-type was defaulting to around. so it was enough to say float: position to create an exclusion but in reality it needs both
- 19:01:20 [karl]
- RRSAgent, minutes?
- 19:01:20 [RRSAgent]
- I'm logging. Sorry, nothing found for 'minutes'
- 19:01:34 [karl]
- RRSAgent, pointer?
- 19:01:34 [RRSAgent]
- See http://www.w3.org/2011/07/25-css-irc#T19-01-34
- 19:01:34 [nimbu]
- rossen: agreement is wrap-mode should be enough to create an exclusion as reg floats are already exclusions, we dont need another.
- 19:02:04 [nimbu]
- rossen: it would be interesting to combine what wrap mode really means outside or inside of shape
- 19:02:25 [nimbu]
- rossen: if you want table cell circular, you can apply shape on the inside of an element without affecting the outside default shape/box model
- 19:02:53 [nimbu]
- rossen: the last version of proposal we had, wrap-mode: default as if wrapping was not involved.
- 19:03:22 [nimbu]
- vhardy: this is a stub, i need more work on that.
- 19:03:29 [nimbu]
- alex: so initial is around?
- 19:03:32 [nimbu]
- rossen: initial is none
- 19:03:48 [nimbu]
- szilles: if its auto floats can behave the way they do today
- 19:04:04 [nimbu]
- vhardy: i propose we dont get into this. we need to work out the details. we have not discussed it.
- 19:04:08 [nimbu]
- rossen: thats fine by me
- 19:04:50 [nimbu]
- dbaron: i am uncomfortable about making resolutions about details when I do not understand the models yet
- 19:05:14 [nimbu]
- vhardy: wrap-mode: inside it will wrap inside (shows slides)
- 19:05:35 [nimbu]
- vhardy: wrap-mode: all-left would control the wrapping internally for the left side.
- 19:06:18 [nimbu]
- vhardy: we resolved having 1 property trigger an exclusion rather than 2 proposed by ms
- 19:06:29 [nimbu]
- alex: maybe we dont need resolutions on these fine grained issues.
- 19:06:41 [nimbu]
- vhardy: i think the resolution is to have a single trigger than 2.
- 19:06:49 [nimbu]
- vhardy: not on the specific properties
- 19:07:40 [nimbu]
- dbaron: it seems like you are proposing the model where anything wraps around anything. i have not seen a lot to explain how thats supposed to work. i saw the spec which says this syntax is supposed to do this thing but it is no where close to doing that thing.
- 19:07:43 [stearns]
- stearns has joined #css
- 19:07:49 [nimbu]
- vhardy: we will get to that.
- 19:09:05 [nimbu]
- ACTION: vhardy make the default value for wrap-mode be auto.
- 19:09:05 [trackbot]
- Created ACTION-356 - Make the default value for wrap-mode be auto. [on Vincent Hardy - due 2011-08-01].
- 19:09:23 [nimbu]
- rossen: next point is containing model for exclusions
- 19:09:52 [nimbu]
- rossen: we didnt make any change to css to containing model. based on position property containing block is computed just like css2 block.
- 19:10:04 [nimbu]
- rossen: there is one exception, that is the new position: page-value that we are adding
- 19:10:21 [nimbu]
- arronei: we can talk about that later when we get to css position
- 19:10:26 [arno1]
- arno1 has joined #css
- 19:10:51 [nimbu]
- alex: what it is saying is that the scope of what an exclusion affects its containing block.
- 19:11:28 [nimbu]
- alex: the important difference this makes compared to other containing model is that it could be defined an element affects anything that it visually overlaps.
- 19:11:49 [nimbu]
- alex: it could be defined an element with exclusion affects an element further down in the content flow and nothing before it.
- 19:12:07 [nimbu]
- alex: that would be very limiting in positioning exclusions backwards and forward which is a very common use case
- 19:12:30 [szilles]
- szilles has joined #css
- 19:12:45 [nimbu]
- dbaron: so you are proposing that things that create exclusions are still positioned according to normal css rules
- 19:13:14 [nimbu]
- dbaron: 2 divs both contain text, the second div has margin-top: −5em wrap shape smthing to make a circle that is 5em radius and they both have text.
- 19:13:23 [nimbu]
- dbaron: where do you position the second text?
- 19:13:46 [nimbu]
- alex: we are getting there to describe the processing model which is where we resolve contradictory layout situation.
- 19:14:03 [nimbu]
- alex: can we go over issue 5 then issue 1 as issue 1 is most difficult issue
- 19:14:26 [nimbu]
- rossen: so, issue 5 was ordering of exclusions. both the container models is same as css2.
- 19:14:41 [nimbu]
- rossen: everything in the scope of the container will be affected by exclusion
- 19:14:42 [nmccully]
- nmccully has joined #css
- 19:16:23 [nimbu]
- rossen: overlapping exclusions: 1. first draft mentioned taking doc order which was not intuitive 2. we are staying with z-index ordering. that appeared as double dependency from exclusions from within the content that would influence the outer exclusions or content outside. but we are scoping the effect of exclusions to containing block then ordering per z-index becomes much more easier.
- 19:16:46 [nimbu]
- szilles: show the example.
- 19:17:05 [nimbu]
- rossen: it is visual vs. content order by default.
- 19:17:46 [nimbu]
- alex: based on z-index, if its not set, elements that over lap prev elements then later elements create exclusions over prev elements. we should try to avoid relayout.
- 19:18:13 [nimbu]
- dbaron: so when you say exclusions affect only other things in containing block I presume you mean everything descendant from containing block. It includes siblings and so on
- 19:18:27 [nimbu]
- dbaron: this idea of using z-index you are potentially crossing stacking context
- 19:19:30 [nimbu]
- dbaron: with subtree model, you can be in situations where excl a should affect b but not vice versa as a is a descendant of b's containing block but b is not a descendant of a's containing block.
- 19:20:20 [nimbu]
- rossen: if u are on some level of the tree, and u have containing block which have exclusions on each side. and a set of eclusions coming from its container.
- 19:21:02 [nimbu]
- rossen: u can order the exclusions in the containing block. you can only take exclusions that are higher than your z-index. once you take those exclusions in your container then you need to resolve with exclusions inside.
- 19:21:29 [nimbu]
- rossen: even if you resolve the exclusions inside, it does not affect exclusions that are coming above the containing block.
- 19:23:02 [mollydotcom]
- mollydotcom has joined #css
- 19:23:08 [dbaron]
- just trying to figure out if this z-order model makes sense given that
- 19:23:39 [nimbu]
- phil: you have a positioned element, and it has some -ve z-index so its going to be behind rest of the content.
- 19:26:43 [smfr]
- ChrisL: mollydotcom is taking some
- 19:26:57 [ChrisL]
- ty
- 19:27:01 [nimbu]
- rossen: it may or may not wrap around D depending on z-index
- 19:27:07 [nimbu]
- rossen: we know D can never wrap around C
- 19:27:12 [nimbu]
- vhardy: is D a containing block?
- 19:27:16 [nimbu]
- alex: all are abs pos
- 19:27:39 [nimbu]
- rossen: the answer here is we apply priority based on z-order at container levels
- 19:27:53 [nimbu]
- rossen: if smthing affects B based on z then everything in B would be affected by that.
- 19:28:11 [nimbu]
- rossen: there should be no way D would wrap around C, so ther eis no way C would wrap around D.
- 19:28:25 [nimbu]
- phil: what if there is opacity on D?
- 19:28:50 [nimbu]
- alex: D does not have to use wrap-mode it does not have to create an exclusion.
- 19:29:12 [nimbu]
- rossen: exclsion is only when you want to "exclude"
- 19:29:41 [nimbu]
- alex: discussion on using -index or other kind of index is a separate discussion.
- 19:30:08 [nimbu]
- alex: issue is to determine order on exclusions and we have come up with z-index.
- 19:30:25 [nimbu]
- rossen: the prev containing model was different that was exposing a number of complexities
- 19:30:39 [nimbu]
- alex: we should consider all hte complicated z-index combination and see if it works
- 19:30:39 [Zakim]
- Zakim has left #css
- 19:30:52 [nimbu]
- szilles: what you saying is that the drawing order is the model of exclusions
- 19:31:01 [nimbu]
- szilles: z-index affects the drawing order.
- 19:31:55 [nimbu]
- glazou: we will continue after lunch
- 19:33:11 [mollydotcom]
- ChrisL: here's a drawing http://twitpic.com/5vs3ra
- 19:33:22 [ChrisL]
- molly, thanks!
- 20:12:10 [sylvaing]
- sylvaing has joined #css
- 20:28:53 [arronei]
- arronei has joined #css
- 20:30:36 [cjon]
- cjon has joined #css
- 20:32:21 [TabAtkins_]
- TabAtkins_ has joined #css
- 20:34:05 [sylvaing]
- yeah, sorry. seems you dropped
- 20:34:39 [arno]
- arno has joined #css
- 20:37:56 [smfr]
- smfr has joined #css
- 20:38:33 [szilles]
- szilles has joined #css
- 20:38:36 [mollydotcom]
- mollydotcom has joined #css
- 20:40:05 [dbaron]
- dbaron has joined #css
- 20:40:21 [glazou]
- glazou has joined #css
- 20:41:10 [kojiishi]
- kojiishi has joined #css
- 20:41:16 [TabAtkins_]
- ScribeNick: TabAtkins_
- 20:42:16 [alexmog]
- alexmog has joined #css
- 20:42:18 [sylvaing]
- rossen
- 20:42:47 [TabAtkins_]
- rossen: [something about z-order] Did that make sense to everybody?
- 20:42:47 [nimbupani]
- nimbupani has joined #css
- 20:42:52 [nmccully]
- nmccully has joined #css
- 20:43:05 [anne]
- anne has joined #css
- 20:43:19 [TabAtkins_]
- vhardy: We'll just take all the feedback and integrate it into the next draft.
- 20:43:41 [TabAtkins_]
- rossen: Back to issue 1, processing model.
- 20:43:42 [florian]
- florian has joined #css
- 20:43:47 [TabAtkins_]
- rossen: Currently we take the easy way out.
- 20:44:21 [TabAtkins_]
- rossen: We treat exclusions as out-of-flow, so you lay out once to pick up any "auto" positions that exlusions may anchor to, then you position exclusions, then you fill in the rest.
- 20:44:26 [tantek]
- tantek has joined #css
- 20:44:42 [TabAtkins_]
- rossen: The obvious problem with this is the accumulation of error; exclusions will push around their own or other's auto positions.
- 20:45:08 [TabAtkins_]
- rossen: There are discussions about a better processing model that would keep the auto positions closer to the right place in the flow, but so far there's been nothing solid enough for us to write down.
- 20:45:08 [florian]
- florian has left #css
- 20:45:11 [florian]
- florian has joined #css
- 20:45:30 [TabAtkins_]
- rossen: That is the processing model we have for oof with position:absolute|fixed|page.
- 20:46:14 [TabAtkins_]
- alexmog: The model rossen is describing, I'm proposing to have it as non-normative atm; it describes something specific enough that simple cases can be interop, but it leaves room for a more advanced processing model later.
- 20:46:17 [jdaggett]
- jdaggett has joined #css
- 20:46:23 [TabAtkins_]
- alexmog: We'll tighten it up as we get impl experience.
- 20:46:39 [TabAtkins_]
- arronei_: If you make it non-normative, you can't test it or depend on it.
- 20:47:04 [TabAtkins_]
- alexmog: We can make it normative to the extent of the simple bits, but I don't think we can fully fill it in yet.
- 20:48:32 [TabAtkins_]
- rossen: position:static has only one difference - you actually lay out with the flow.
- 20:49:27 [TabAtkins_]
- rossen: For position:static exclusions, they're laid out with the normal flow pass.
- 20:49:39 [TabAtkins_]
- rossen: If all you hae is static exclusions, you don't need a second pass, as they're laid out as part of the content.
- 20:49:49 [TabAtkins_]
- alexmog: What happens if they have a negative margin?
- 20:49:58 [TabAtkins_]
- rossen: Same as today - they'll overlap the previous content.
- 20:50:14 [TabAtkins_]
- alexmog: Doesn't that contradict what exclusions should be doing?
- 20:50:42 [bradk]
- It would be like a float, more or less, if static? Only affecting the leading edges?
- 20:50:45 [TabAtkins_]
- rossen: They only push around other stuff in the abspos case.
- 20:51:22 [TabAtkins_]
- rossen: [draws a diagram]
- 20:51:40 [bradk]
- Can someone post a photo once there is something to look at on the whiteboard?
- 20:52:37 [TabAtkins_]
- rossen: While laying out text, you lay out line by line. When you encounter a static exclusion, you're in the middle of layout out one line (previous lines are already done).
- 20:53:11 [TabAtkins_]
- rossen: If this was a regular div, the block would consume all the space to its sides, pushing following text below it. Exclusion just lets following text fill in space around it.
- 20:54:15 [arno1]
- arno1 has joined #css
- 20:54:16 [TabAtkins_]
- rossen: Moving on to further issues.
- 20:54:30 [TabAtkins_]
- rossen: issue 7, what does "shrink-to-fit" mean for floats with shapes?
- 20:54:33 [szilles]
- szilles has joined #css
- 20:55:33 [TabAtkins_]
- rossen: We defined an easy way out here - use the bounding box of the exclusion shape as the "shrink-to-fit" for the floater.
- 20:56:00 [TabAtkins_]
- alexmog: Nobody really liked the idea that they have to calculate the closest intersection of arbitrary shapes - too hard of a problem.
- 20:56:13 [TabAtkins_]
- alexmog: So it should be acceptable to treat exclusions as rectangles for the purpose of shrink-to-fit.
- 20:57:27 [TabAtkins_]
- alexmog: [something explaining the above point that I missed]
- 20:57:46 [TabAtkins_]
- rossen: I think the shape doesn't correspond to the border box - it could maybe define its own bounding rect.
- 20:58:24 [TabAtkins_]
- alexmog: It shouldn't get bigger than what's calculated with bounding boxes.
- 20:58:29 [TabAtkins_]
- alexmog: You can always make it smaller.
- 20:58:42 [TabAtkins_]
- szilles: I'm slightly confused. If I have a wrap-shape, it has a size. Why do I need shrinkwrap at all?
- 20:59:10 [vhardy]
- vhardy has joined #css
- 20:59:14 [TabAtkins_]
- rossen: We're talking about floats that are themselves shaped.
- 20:59:27 [TabAtkins_]
- dbaron: The shapes can have percenetages, so the shape is based on the size itself.
- 20:59:48 [TabAtkins_]
- dbaron: If the shape is a circle centered on the box with a radius of 50%, the shape is a function of the size.
- 21:00:06 [TabAtkins_]
- rossen: If that was a div with width:50% inside of the float, the same effect would occur.
- 21:01:16 [TabAtkins_]
- rossen: One more difference from the original exclusions spec was a property named "flow-wrap" that allowed elements to not pay attention to exclusions.
- 21:01:22 [TabAtkins_]
- rossen: We should maybe rename it.
- 21:01:46 [miketaylr]
- miketaylr has joined #css
- 21:01:54 [TabAtkins_]
- rossen: A lot of people thought it was related to a "wrap-mode" property.
- 21:03:00 [TabAtkins_]
- rossen: We also have Daniel's proposal that proposed a fairly different way to introduce exclusions.
- 21:03:10 [TabAtkins_]
- vhardy: I think your presentation is a convergence of the two proposals.
- 21:03:16 [TabAtkins_]
- rossen: Yeah, it got assimilated.
- 21:03:30 [TabAtkins_]
- vhardy: So I think the next thing is to work on a new ED.
- 21:03:36 [TabAtkins_]
- arno: And that's a wrap.
- 21:03:41 [stearns]
- perhaps the name for "ignore wrap" should follow whatever we end up with for "cancel-underline"
- 21:03:44 [TabAtkins_]
- /groan
- 21:04:18 [TabAtkins_]
- alexmog: I think we came to the conclusion that the name of the module should be "CSS3 Floats".
- 21:04:32 [TabAtkins_]
- alexmog: It eventually needs to be a bigger doc that describes much mor eof th eprocessing model,
- 21:04:47 [TabAtkins_]
- ... describing floats that don't overlap too.
- 21:07:17 [TabAtkins_]
- alexmog: [something about how extending 'float' might work]
- 21:07:26 [TabAtkins_]
- arno1: Would that be in CSS3 Floats or something else?
- 21:07:42 [TabAtkins_]
- vhardy: I think we should produce a first draft that covers what we know we want and have.
- 21:07:54 [TabAtkins_]
- vhardy: Then later figure out how we want to slice it with gcpm and other float-related things.
- 21:08:27 [TabAtkins_]
- arno1: My concern is that if this is float-related, we'd have to close on everything float-related to make this work.
- 21:08:38 [TabAtkins_]
- alexmog: I had the same concern. I think we can do Exclusions separately from Floats.
- 21:09:08 [TabAtkins_]
- alexmog: there is an issue with making Exclusions a Rec without having anything with floats.
- 21:09:28 [TabAtkins_]
- alexmog: But I think vincent has the right idea that we should write the draft now, and when we see it, it should be easy to tell what the right name is.
- 21:09:57 [TabAtkins_]
- dbaron: My general feeling about this draft is that it's very complicated to implement, and it's possible to write something much simpler to implement with most of the functionality.
- 21:10:08 [TabAtkins_]
- dbaron: I may want to look into going that way, but I haven't had a chance to do that yet.
- 21:10:29 [TabAtkins_]
- vhardy: If we produced a new draft that you could comment on, would that be ok?
- 21:10:42 [TabAtkins_]
- dbaron: I think at this point the best way forward for me is to produce a new proposal.
- 21:11:21 [TabAtkins_]
- florian: If we have proposals that look close enough, it's worth merging them. But we're early enough on right now that if there's something significantly different, it's worth exploring that separately.
- 21:12:08 [TabAtkins_]
- alexmog: I would be really glad if we could remove things from the draft and still meet the goals.
- 21:12:27 [TabAtkins_]
- florian: David, is what you have in mind the current draft with some things removed, or something totally different?
- 21:12:59 [TabAtkins_]
- dbaron: Similar things, but with a significantly different processing model.
- 21:13:51 [TabAtkins_]
- dbaron: I'm thinking that I want to separate the reordering aspects, and then not have the reordering implications within layout.
- 21:14:03 [TabAtkins_]
- dbaron: So if you need reordering, that's a separate process, and then layout is still one pass.
- 21:14:46 [TabAtkins_]
- dbaron: The part of this draft that scares me is the part that's not yet written.
- 21:15:12 [TabAtkins_]
- dbaron: It's hard to know I have comments on it before I read it. There may be a big gap somewhere that implies a big complicated chunk of processing model.
- 21:15:18 [TabAtkins_]
- dbaron: I've only thought about it for the last few days.
- 21:15:47 [TabAtkins_]
- vhardy: The processing model we've described is a bit different than what's currently in the ED.
- 21:15:54 [TabAtkins_]
- dbaron: I couldn't find much of a processing model.
- 21:16:12 [smfr]
- http://wiki.csswg.org/ideas/css3-floats-use-cases
- 21:16:18 [TabAtkins_]
- florian: Do you think that what you want fits within the shape of the existing proposal?
- 21:16:25 [arronei]
- http://wiki.csswg.org/ideas/css3-floats
- 21:16:34 [TabAtkins_]
- dbaron: There are some examples that I know don't fit, but they're incomplete examples right now.
- 21:16:43 [TabAtkins_]
- rossen: Have you checked out the wiki examples yet?
- 21:16:46 [TabAtkins_]
- dbaron: Not yet.
- 21:17:18 [TabAtkins_]
- rossen: If you could look, it would be great to have a review and see what's missing or whats overcomplicated.
- 21:17:32 [TabAtkins_]
- rossen: If you have an alternate proposal, I'd like to see if it solves these use-cases.
- 21:17:54 [TabAtkins_]
- vhardy: We should take an action item to review this page.
- 21:18:34 [glazou]
- jdaggett: can you call in?
- 21:18:50 [szilles]
- szilles has joined #css
- 21:19:13 [TabAtkins_]
- Topic: Writing Modes
- 21:19:17 [glazou]
- jdaggett: can you hear us?
- 21:19:55 [TabAtkins_]
- fantasai: I think there are several issues open right now.
- 21:19:59 [TabAtkins_]
- fantasai: One is the name of the property.
- 21:20:07 [TabAtkins_]
- fantasai: Second is the defualt orientation at a codepoint level.
- 21:20:17 [TabAtkins_]
- fantasai: The third is whether we're doing context-based resolution of punctuation, and if so, how?
- 21:20:34 [TabAtkins_]
- fantasai: The fourth is if we're giving additional controls at level 3 (beyond what you can currently do with text-orientation).
- 21:20:45 [TabAtkins_]
- jdaggett: Let's start iwth sylvain's issue.
- 21:21:00 [TabAtkins_]
- jdaggett: I think #3 and #4 are dependent on a concrete proposal for #2.
- 21:21:21 [TabAtkins_]
- szilles: Two more issues are Nat's comments on TCY
- 21:22:40 [TabAtkins_]
- szilles: I think you suggested separating the text-combine prop into one that did manual TCY and one that set up automatic TCY.
- 21:22:47 [TabAtkins_]
- szilles: They may want to be over different ranges.
- 21:22:57 [glazou]
- jdaggett: (we're in a big room, 30 people in the room, some of us are pretty far from the microphone… tell us if you don't hear us)
- 21:23:03 [TabAtkins_]
- jdaggett: Why do you need that distinction, Nat?
- 21:23:13 [dbaron]
- (TCY == Tate-chu-yoku)
- 21:23:22 [TabAtkins_]
- nat: Right now we have text-combine:horizontal, then a bunch of values for controlling the auto case, i guess.
- 21:23:41 [TabAtkins_]
- nat: Is that what we want? Just specify it on, and then specify the details (do letters, do digits, combine X many)
- 21:23:55 [TabAtkins_]
- nat: And then, if the lettes in the span meet that criteria, combine them?
- 21:24:18 [TabAtkins_]
- florian: One case was that, or a date, you don't wnat to have to add extra spans around the day and month and year, but rather just one aroudn th eyear.
- 21:24:31 [TabAtkins_]
- nat: The other issue I had was the defualt implied value.
- 21:24:59 [TabAtkins_]
- fantasai: The values are "none", "all" (tcy the whole span), and then various options that control what things are combined.
- 21:25:32 [jdaggett]
- nat's post: http://lists.w3.org/Archives/Public/www-style/2011Jul/0405.html
- 21:25:47 [dbaron]
- http://dev.w3.org/csswg/css3-writing-modes/#text-combine
- 21:26:00 [TabAtkins_]
- nat: I thought if I tried to implement this, it would be easier if I didin' thave to worry about having a state machine go through the text.
- 21:26:26 [TabAtkins_]
- fantasai: You can't combine, say "all" and "latin 4" - it's not syntactically valid.
- 21:26:30 [fantasai]
- ACTION fantasai: Add syntax examples
- 21:26:30 [trackbot]
- Created ACTION-357 - Add syntax examples [on Elika Etemad - due 2011-08-01].
- 21:26:37 [TabAtkins_]
- nat: i think some examples would help here.
- 21:26:58 [TabAtkins_]
- nat: So when it's not all, you can do auto TCY in any span.
- 21:27:25 [TabAtkins_]
- nat: A second point, why do you have, in the alphanumeric tolerance section, a "1.1em" figure. That seems arbitrary.
- 21:27:50 [TabAtkins_]
- fantasai: koji was talking with hyatt about that issue. hiragina apparently, if you ask for half-width glyphs, won't fit within 1em if you put them side-by-side.
- 21:28:04 [TabAtkins_]
- szilles: if that's true, why not just do it when they ask for it?
- 21:28:29 [TabAtkins_]
- fantasai: This is for whether you scale or not. This number is close enough that you usually won't have to scale.
- 21:28:34 [stearns]
- s/hiragina/hiragana
- 21:29:10 [TabAtkins_]
- kojiishi: The idea of scale is to compress glyphs so that they fit into 1em.
- 21:29:28 [TabAtkins_]
- kojiishi: But there are some fonts that get slightly gretaer than 1em when you set two half-width next to each other.
- 21:30:09 [TabAtkins_]
- kojiishi: So if two chars are much larger than 1em, authors would likely want to compress them. But if they're just a tiny bit over (maybe accidentally, like described here), you'd want to just leave them alone.
- 21:30:25 [TabAtkins_]
- nat: Okay. I think having the built-in tolerance is orthogonal to the scaling issue.
- 21:30:26 [dbaron]
- (not Hiragino? I think they were talking about the font)
- 21:31:14 [TabAtkins_]
- nat: If you're tolerant of the idea that you're using glyphs that odn't fit into 1em, you're probably okay with not scaling, so you should just say no scaling.
- 21:31:44 [TabAtkins_]
- kojiishi: Even if you do scale, though, you probably still don't want to scale if you're just very slightly over 1em.
- 21:31:52 [ChrisL]
- ChrisL has joined #css
- 21:31:56 [karl]
- karl has joined #CSS
- 21:32:32 [TabAtkins_]
- nat: But you're saying to scale it to 1.1em, not 1em.
- 21:32:43 [TabAtkins_]
- nat: You can have an "auto-scale" that has some tolerance in there.
- 21:32:57 [TabAtkins_]
- florian: Can we just have a scale with no tolerance, and let you add the tolerance in specifically?
- 21:33:33 [arronei_]
- arronei_ has joined #css
- 21:34:01 [ChrisL]
- rrsagent, here
- 21:34:01 [RRSAgent]
- See http://www.w3.org/2011/07/25-css-irc#T21-34-01
- 21:34:06 [TabAtkins_]
- szilles: If they're images, you really notice the artifacts. If they're outlines you won't notice as much.
- 21:34:28 [TabAtkins_]
- szilles: Can we say something other than "scale"?
- 21:34:38 [TabAtkins_]
- fantasai: squish/compress/scale-x?
- 21:34:40 [bradk]
- condense, horizontal scaling
- 21:34:43 [TabAtkins_]
- fantasai: compress
- 21:35:30 [TabAtkins_]
- jdaggett: If someone wants this tolerance, they should be involved in the discussion.
- 21:35:49 [TabAtkins_]
- florian: I don't think it's useless, but it seems to be going a bit too far right now. Push it to level 4.
- 21:35:53 [dbaron]
- (above) Nat: condensed is bad because it implies a different font selection
- 21:36:26 [TabAtkins_]
- fantasai: The tolerance is used in the scaling (removed now), and also in the "used glyphs" section.
- 21:36:42 [TabAtkins_]
- fantasai: Which says to use half/third-width glyphs if available; if not available, scale.
- 21:37:04 [bradk]
- Adobe calls it "horizontal scaling" in their programs, IIRC, FWIW
- 21:37:12 [TabAtkins_]
- fantasai: What might happen is that you say you want to use third-width glyphs, but it ends up being rendered with a font without third-width glyphs.
- 21:37:30 [TabAtkins_]
- fantasai: The default action, then, should be to go ahead and scale, as that's minimally invasive to the rest of the page design.
- 21:37:51 [TabAtkins_]
- nat: Is that how it's supposed to work? I didn't get the idea that things had fallback.
- 21:37:58 [TabAtkins_]
- fantasai: You can combine "scale" and "use-glyphs".
- 21:38:27 [TabAtkins_]
- szilles: Why wouldn't you scale to 1em?
- 21:38:42 [TabAtkins_]
- florian: You'd scale to 1em, but you wouldn't scal ei fyou're "close enough" to 1em.
- 21:39:47 [szilles]
- szilles has joined #css
- 21:39:49 [TabAtkins_]
- szilles: [an example with "1.3" and TCY]
- 21:39:50 [bradk]
- Is it tragic to allow the glyphs tooverflow if they are slightly too wide?
- 21:40:06 [TabAtkins_]
- jdaggett: In the original document this was based, there's an example for that.
- 21:40:35 [TabAtkins_]
- szilles: I know that stuff like TCY'd "1.357" is actually done.
- 21:40:49 [jdaggett]
- original document == jis 4051 spec
- 21:40:53 [TabAtkins_]
- nat: I think we should just have a "use glyphs" option, and if glyphs don't exist, just do fallback.
- 21:41:40 [TabAtkins_]
- fantasai: We can say that if you find the right glyphs, just don't worry about compressing. Assume that the glyphs are the right width.
- 21:42:49 [TabAtkins_]
- kojiishi: If the first font has the glyphs (slightly off the right width) but the second font doesn't, the first font would grab the glyphs and be slightly off-width, while the second would synthesize and be exactly right.
- 21:42:58 [TabAtkins_]
- nat: I think you're far off in edge-cases here.
- 21:43:24 [TabAtkins_]
- szilles: I think it's pretty simple. If it has half-width glyphs, it has half-width glyphs. Just use them.
- 21:43:58 [TabAtkins_]
- nat: Why are we doing anything if the half-width glyphs dont' exist?
- 21:44:19 [TabAtkins_]
- szilles: CSS generally tries to best capture the authors' intent.
- 21:45:20 [TabAtkins_]
- kojiishi: If the author says "digits 3", and the half-widths are slightly more than half an em, you'll get slightly more than 1em-width TCY.
- 21:45:43 [TabAtkins_]
- fantasai: Is it possible to know if the font has third-width glyphs?
- 21:45:50 [TabAtkins_]
- nat: There's a feature in OT that the UA can query.
- 21:46:21 [TabAtkins_]
- jdaggett: The problem with these features is that for different features, different sets of glyphs are available.
- 21:46:32 [TabAtkins_]
- jdaggett: For third/fourth-width, often only digits are available.
- 21:46:46 [TabAtkins_]
- szilles: What about numeric punctuation (periods and commas)?
- 21:46:59 [TabAtkins_]
- nat: Generally not. Most only have half-width punctuation.
- 21:47:19 [TabAtkins_]
- szilles: So you could do years and integers, but not decimal numbers.
- 21:47:21 [TabAtkins_]
- nat: Right.
- 21:47:38 [TabAtkins_]
- szilles: I think we just need a clear processing model of whe nthings are decided.
- 21:48:05 [TabAtkins_]
- szilles: I don't think we want to say "let's make as tring, and then try to compress it". I think you want to first check for half-width glyphs or whatever, and then build out of what exists.
- 21:48:58 [TabAtkins_]
- szilles: So I want a clear statement of what you're looking for from the font, and underwhat conditions.
- 21:49:13 [TabAtkins_]
- fantasai: So what I'm hearing is that we should use the correct glyphs if the font has them, otherwise synthesize.
- 21:49:47 [TabAtkins_]
- fantasai: So if I'm doing "122", if the font has third-width glyphs, I use them. If it doesn't, I scale the "1" and "2" glyphs and then combine them, rather than making "122" and then scaling.
- 21:50:25 [TabAtkins_]
- florian: Also, the current spec specifies just that they should fit into 1em, whihc allows cherry-picking half and quarter-width glyphs and combining them, which is not what you want.
- 21:50:29 [bradk]
- Doesn't kerning and tracking affect if two 1/2-width glyphs fit into one em?
- 21:52:42 [TabAtkins_]
- fantasai: We can tell if the font has a "third-width" feature, but we don't know if individual characters have third-width glyphs. So if I wanted "IBM" in third-width glyphs, I'd request those glyphs from the font as third-width glyphs, but I may or may not actually get third-width glyphs back.
- 21:53:12 [TabAtkins_]
- fantasai: So once I get them back, if they're wider than 1/3em, I need to scaled them to 1/3 em.
- 21:53:59 [TabAtkins_]
- fantasai: I think we can make an exception that, for half-width glyphs, don't measure and just use what you get. You may get proportional, which may be close enough.
- 21:54:05 [TabAtkins_]
- bradk: Can kerning affect this?
- 21:54:14 [TabAtkins_]
- nat: You don't kern monospace CJK fonts.
- 21:54:31 [TabAtkins_]
- jdaggett: In theory, it's an issue; in practice, you don't ever kern these types of fonts. I don't think we need to worry about it.
- 21:54:58 [jdaggett]
- ... you don't generally kern these *glyphs*
- 21:55:16 [jdaggett]
- i.e. third-width glyphs
- 21:55:35 [TabAtkins_]
- kojiishi: I think it may still be useful to have an option to not scale glyphs, as it may be ugly sometimes.
- 21:56:19 [TabAtkins_]
- florian: Perhaps if you say "use-glyph no-scale", you just use exactly what's given back by the font, no measurement. It may be too big, but shrug. Then "use-glyph" will scale if they're too big, and "compress" will just always scale the full-size.
- 21:56:24 [TabAtkins_]
- fantasai: Makes sense.
- 21:57:19 [fantasai]
- ACTION fantasai: edit as above
- 21:57:19 [trackbot]
- Created ACTION-358 - Edit as above [on Elika Etemad - due 2011-08-01].
- 21:57:52 [TabAtkins_]
- RESOLVED: Remove the concept of glyph-size tolerance from this level of the spec.
- 21:58:11 [jdaggett]
- whatever you spec out, you need to try it on some simple examples and include them in the description
- 21:58:19 [TabAtkins_]
- fantasai: So for fonts with propotional, but not half-width glyphs, if you say "use-glyphs", should we just use the glyphs directly?
- 21:58:28 [jdaggett]
- the markup should be easy and the implementation should be fairly easy
- 21:58:47 [jdaggett]
- nat: yeah, you don't want multiple passes to support TCY
- 21:59:16 [TabAtkins_]
- szilles: I think, since the user can say "no-scale" if they really don't want it, you should have a single default that works decently enough for everyone.
- 21:59:52 [szilles]
- szilles has joined #css
- 22:00:46 [TabAtkins_]
- kojiishi: The tolerance came in to address the common case where the TCY want to use two digits without scaling.
- 22:03:45 [TabAtkins_]
- nat: If we have different scaling values (depending on number of digits), it's too complicated for the value you get.
- 22:03:59 [TabAtkins_]
- nat: I agree that it may be useful in some cases, as Koji says, but I don't think its' worthwhile to address.
- 22:04:24 [TabAtkins_]
- nat: Koji brings up the possibility that some people may want to scale only when it looks terrible, and let it not scale if it's "almost there".
- 22:04:39 [TabAtkins_]
- nat: But I don't think it's worth dealing with that explicitly. That should be something a UA could decide.
- 22:05:00 [TabAtkins_]
- florian: Or we could go with steve's idea of adding a tolerance to scale.
- 22:05:11 [TabAtkins_]
- szilles: I mainly just think we're spending too much time on this.
- 22:05:33 [TabAtkins_]
- fantasai: And the default is the UA tries to do whatever it can to make things fit. No limitations.
- 22:06:56 [TabAtkins_]
- nat: Moving on, it looks like you can omit the integer in "digits" etc. and it defaults to 2. I think it would be better to be explicit.
- 22:06:59 [TabAtkins_]
- fantasai: Okay.
- 22:08:17 [szilles]
- szilles has joined #css
- 22:08:19 [TabAtkins_]
- kojiishi: You're not requiring an integer for "all", right?
- 22:08:30 [bradk]
- cute
- 22:08:48 [hober]
- bradk: i think you mean 'kawaii'
- 22:08:49 [TabAtkins_]
- fantasai: Right.
- 22:09:21 [TabAtkins_]
- nat: We should just make "all" be very simple. It does *all*.
- 22:09:40 [szilles]
- Add Usecases for 2 or 3 digits in auto t-c-y and have an example with mix of digits and non-digits; e.g., "1.2"
- 22:09:46 [florian]
- s/steve's idea of adding a tolerance to scale/steve's idea of adding a parameter to no-compress, rather than using tolerances/
- 22:11:39 [TabAtkins_]
- fantasai: If you say "use-glyphs" and there's only a single char, the current spec says to use a full-width glyph if one is availab.e
- 22:11:45 [TabAtkins_]
- nat: And I think you shouldn't do that.
- 22:12:12 [TabAtkins_]
- nat: glyphs substituation may cause problems (undefined?)
- 22:12:40 [TabAtkins_]
- nat: Are you supposed to scale up the glyphs to 1em, or what?
- 22:13:06 [TabAtkins_]
- florian: i think we've said that TCY and text-transform interacts.
- 22:13:48 [TabAtkins_]
- fantasai: Spec says that any text-transform features are turned off for combined text of more than 1 char.
- 22:14:20 [TabAtkins_]
- florian: Then that's fine. You can use text-transform to get single large chars, but have smaller chars for more TCY.
- 22:14:38 [TabAtkins_]
- fantasai: jdaggett you were saying that for text-transform we should write up proposals first and then discuss?
- 22:15:04 [TabAtkins_]
- jdaggett: Yeah, I did some experiments. Spec says currently that you have to use the VR2 feature of the font, but in practice that doesn't really work.
- 22:15:05 [jdaggett]
- http://lists.w3.org/Archives/Public/www-style/2011Jul/0401.html
- 22:15:07 [nmccully]
- nmccully has joined #css
- 22:15:09 [TabAtkins_]
- s/text-transform/text-orientation/
- 22:15:45 [TabAtkins_]
- jdaggett: In the post I summed up the key points:
- 22:16:03 [TabAtkins_]
- jdaggett: Latin will rotate, greek and cyrillic won't, and there are many other cases where it's not clear that what exists is what you want.
- 22:16:08 [jdaggett]
- http://lists.w3.org/Archives/Public/www-style/2011Jul/0402.html
- 22:16:35 [TabAtkins_]
- jdaggett: In discussion with Erik Mueller, we thought it mad emore sense to specify a property (possibly going into unicode) that says whether a character is upright or rotated.
- 22:16:46 [TabAtkins_]
- jdaggett: Then we can use that property to define the gray areas here.
- 22:16:53 [stearns_]
- stearns_ has joined #css
- 22:16:54 [TabAtkins_]
- jdaggett: symbols, punctuation, currency, etc.
- 22:16:58 [TabAtkins_]
- fantasai: That sounds reasonable.
- 22:17:36 [TabAtkins_]
- jdaggett: I just don't think this can be a derived property from the existing values.
- 22:17:53 [TabAtkins_]
- fantasai: Agreed - until it's a unicode property, we'll have to have it be derived with a large list of exceptions.
- 22:18:33 [TabAtkins_]
- jdaggett: And then the logic of text-orientation will be to use that property, then apply the "vert" opentype feature to the upright runs.
- 22:18:59 [TabAtkins_]
- nat: And you do that so fonts which disagree with the feature can achieve that?
- 22:19:01 [TabAtkins_]
- jdaggett: Yes.
- 22:19:21 [TabAtkins_]
- nat: I'm comfortable with that.
- 22:19:35 [TabAtkins_]
- jdaggett: This property effectively already exists, but it's just custom right now in different tables.
- 22:19:59 [TabAtkins_]
- jdaggett: IE already has it (though clearly out of date; it doesn't do non-BMP). Webkit has a similar table, similarly wrong.
- 22:20:07 [ChrisL]
- has Unicode consortium been asked to add this property?
- 22:20:21 [TabAtkins_]
- jdaggett: It would be great to pull together our individual dbs and help make a proposal to unicode.
- 22:20:45 [TabAtkins_]
- nat: InDesign has something similar too.
- 22:21:22 [ChrisL]
- (I take what jdagett said to mean "not yet but we will")
- 22:22:00 [TabAtkins_]
- jdaggett: I don't think we need to get into the individual heuristics yet.
- 22:22:21 [TabAtkins_]
- nat: When unicode didn't provide two codepoints for something that was two codepoints in shift_jis, we have to be able to differentiate.
- 22:22:35 [TabAtkins_]
- nat: Some fonts have different designs for the two. In some fonts you want to rotate, and in others you don't.
- 22:23:12 [TabAtkins_]
- fantasai: We can get info for Japanese fairly easily, but I'm concerned about other languages like Chinese where we don't have much if any information.
- 22:23:26 [TabAtkins_]
- nat: It's different, and unfortuantely the standards are even more loose.
- 22:24:15 [TabAtkins_]
- nat: We don't use font metrics, but it's very complicated to deal with this.
- 22:24:44 [TabAtkins_]
- jdaggett: to handle that sort of thing, we could call out those specific codepoints and deal with those in a separate category.
- 22:24:59 [TabAtkins_]
- jdaggett: So CSS3 can deal with japanese well, but the separation lets us make better rules for Chinese and such later.
- 22:25:25 [TabAtkins_]
- szilles: You and I, jdaggett, have proposed being able to provide a delta on the unicode table.
- 22:25:49 [TabAtkins_]
- jdaggett: I think it would be better to wait and propose the rpoeprty first, then went through and discussed how you might want to modify it.
- 22:26:19 [TabAtkins_]
- szilles: Right; I brought it up as a reminder of the concept.
- 22:26:54 [TabAtkins_]
- jdaggett: Right. it would be nice to eventually let authors say exactly what defaults they want ("never rotate english text"), so they didn't need to put a bunch of markup around specific areas.
- 22:27:07 [TabAtkins_]
- szilles: It would also allow all the default tables out there to be incorporated.
- 22:28:17 [TabAtkins_]
- fantasai: I have a concern that the rest of the draft (the layout parts) really ought to be stabilized somehow.
- 22:28:42 [TabAtkins_]
- fantasai: We have a lot of work left to do on text-orientation, but I think the layout bits should be locked down and not held back by this work.
- 22:29:34 [TabAtkins_]
- fantasai: Another month or two i can handle, but if this'll take us until next year, I'm concerned.
- 22:29:38 [TabAtkins_]
- jdaggett: I don't think it'll take that long.
- 22:29:49 [TabAtkins_]
- fantasai: One more item now.
- 22:29:59 [TabAtkins_]
- fantasai: Sylvain was talking about the name of 'writing-mode' property.
- 22:30:08 [szilles]
- szilles has joined #css
- 22:30:20 [TabAtkins_]
- sylvaing: The writing-mode property says it defines block-progression (then why isn't in named "block-progression").
- 22:30:38 [TabAtkins_]
- sylvaing: then it says that the writing mode is deifned by three properties, one of which is called "writing-mode", which is confusing.
- 22:30:56 [TabAtkins_]
- fantasai: I think we shouldn't change the name, but if you have a better term...
- 22:31:06 [TabAtkins_]
- sylvaing: There used to be a block-progression property.
- 22:31:27 [TabAtkins_]
- fantasai: Yes, because at the time writing-mode was a shorthand that overrides "direction", which was a very bad thing. We've had this discussion a few times.
- 22:31:34 [TabAtkins_]
- fantasai: And it's incompatible with SVG and IE6.
- 22:31:37 [TabAtkins_]
- sylvaing: What is?
- 22:32:06 [TabAtkins_]
- fantasai: We could call it block-progression, but then writing-mode would be an alias.
- 22:32:59 [TabAtkins_]
- dbaron: Is this a naming discussion, or a substantive one?
- 22:33:31 [TabAtkins_]
- fantasai: Kinda both. Given legacy and other reasons, I think the name should stay, but I'm willing to change the terminology.
- 22:34:08 [TabAtkins_]
- szilles: I agree that this would be better called "block-progression". But history/SVG means that it should stay consistent and be "writing-mode".
- 22:34:35 [ChrisL]
- svg also uses the term 'block progression direction' (from xsl) btw
- 22:34:44 [TabAtkins_]
- fantasai: I'd just like to avoid property aliases.
- 22:34:47 [ChrisL]
- but it does use the writing-mode property, yes
- 22:34:54 [TabAtkins_]
- dbaron: value aliases are much eaiser than property aliases.
- 22:39:04 [bradk]
- meeting is over for today?
- 22:39:15 [stearns_]
- no - just on break
- 22:39:21 [ChrisL]
- i believe another 2.5 hours
- 22:39:44 [bradk]
- OK, thanks
- 22:40:15 [bradk]
- It sounded like everyone was heading out to go boating, but I guess that is later
- 22:46:57 [JohnJansen]
- correct, Brad. That's at 7 tonight.
- 22:56:23 [fantasai]
- ScribeNick: fantasai
- 22:56:31 [fantasai]
- Topic: Gradients
- 22:56:38 [fantasai]
- Tab: 3 issues to resolve
- 22:57:04 [fantasai]
- Tab: First one is repeating gradients when the distance between the first and last stop approaches zero
- 22:57:12 [vhardy]
- vhardy has joined #css
- 22:57:29 [fantasai]
- TabAtkins: zw gradients are not a problem without repeating, but is a problem for repeating because you can't repeat them on the same point infinitely
- 22:57:45 [fantasai]
- TabAtkins: Right now I punt on the issue by making repeating gradients requreid to have minimum 1px width
- 22:57:53 [fantasai]
- TabAtkins: That guarantees we can tile the space like we should
- 22:58:02 [fantasai]
- TabAtkins: Don't know if it's ideal, but it stops the problem of infinity
- 22:58:27 [fantasai]
- TabAtkins: Other possibility that preserves continuity is that when we hit zero width, we pick the average color you'd get
- 22:58:43 [fantasai]
- TabAtkins: Seems kinda complicated for an edge case, but I don't realy care, just want a decision
- 22:58:52 [fantasai]
- TabAtkins: And to be consistent with svg
- 22:59:00 [fantasai]
- Florian: Pixels aren't px
- 22:59:12 [fantasai]
- TabAtkins: I'm fine with saying it's a hairline wide
- 22:59:21 [fantasai]
- Florian: So you do fallback at device pixel
- 22:59:30 [fantasai]
- Brian: WD says to use first oclor stop
- 22:59:39 [fantasai]
- TabAtkins: Yeah, but that's not continuous behavior
- 22:59:53 [ChrisL]
- device pixels is better. because svg scaling can mean that a 1px is really big
- 23:00:06 [fantasai]
- [fast talking]
- 23:00:22 [fantasai]
- Brian: Non-repeating radial gradients folow last color stop rule
- 23:00:30 [fantasai]
- dbaron: Butwith those you always fill the area outside with that color
- 23:00:40 [fantasai]
- TabAtkins: As you approach the limit, you get that result
- 23:00:44 [fantasai]
- Brian: So it's continuity issue.
- 23:00:57 [fantasai]
- TabAtkins: I'm ok with either way, whatever way implementers would like to do, let's do
- 23:01:05 [fantasai]
- shepazu: SVG, when it hits zero, says that nothing is rendered
- 23:01:12 [fantasai]
- shepazu: That kinda punts on that
- 23:01:24 [fantasai]
- shepazu: I don't know that SVG says anything about approaching zero
- 23:01:43 [ChrisL]
- it does not, no. its either a zero bounding box or it isn't
- 23:01:47 [fantasai]
- shepazu: The averaging thing doesn't seem like what ppl want to do
- 23:02:09 [fantasai]
- fantasai: I think that makes the most sense; asyou zoom out that's what you'd get
- 23:02:24 [fantasai]
- shepazu: It'd be mixed with other color.. start color end color harmonious with page
- 23:03:03 [fantasai]
- TabAtkins: If you do 1px red blue and repeat it, it looks purple. If you make it smaller, can assume it'll look purple
- 23:03:31 [fantasai]
- Brian: 1px seems too large to me. But making the spec normative at 1px, but allow better resolution
- 23:03:33 [smfr]
- smfr has joined #css
- 23:03:44 [fantasai]
- TabAtkins: So the clamp is at 1px or below.
- 23:03:55 [nmccully]
- nmccully has joined #css
- 23:03:55 [anne]
- anne has joined #css
- 23:03:58 [ChrisL]
- px or device pixel?
- 23:04:05 [fantasai]
- Florian: So this would allow them to go to device pixel, but not force it
- 23:04:06 [fantasai]
- CSSpx
- 23:04:10 [ChrisL]
- eww
- 23:04:19 [ChrisL]
- device pixels is better. because svg scaling can mean that a 1px is really big
- 23:04:41 [fantasai]
- Brian: Do you want inconsistent rendering across zoom
- 23:05:02 [fantasai]
- Brian: If you'll get purple pixels sooner at one zoom level than another zoom level
- 23:05:23 [fantasai]
- ...
- 23:05:35 [fantasai]
- TabAtkins: Gradients are a vector format in any case
- 23:05:49 [fantasai]
- shepazu: If you zoom in until 1px is the whole screen, then what?
- 23:06:21 [fantasai]
- TabAtkins: You're allowed to clamp at a smaller size
- 23:06:40 [fantasai]
- Brad: Makes sense. Like with text you get better resolution as you zoom in.
- 23:06:48 [ChrisL]
- <g transform="scale(1000,1000)"><rect width="1px" height="1px"fill="url(#gradient"/></g>
- 23:06:52 [fantasai]
- TabAtkins: It ends up being a quality-of-implementation issue.
- 23:07:16 [fantasai]
- dbaron: As the length of the repeating gradient approaches zero, you should average the color.
- 23:07:24 [fantasai]
- Florian: Need to make sure it's not going to do that at 20px
- 23:07:42 [dbaron]
- s/As the/You could just specify that as the/
- 23:07:43 [ChrisL]
- I don't want the 1000 by 1000 device pixels rect above to be a solid colour
- 23:08:09 [fantasai]
- plinss talks about billionths of a CSSpx
- 23:08:56 [ChrisL]
- @shepazu could you read out my example please and make sure people understand it
- 23:08:58 [fantasai]
- dbaron: I think it should be device pixels, because if you zoom out ...
- 23:09:24 [fantasai]
- plinss: Say that the UA can substitute an average color if the gradient length is small [...]
- 23:09:49 [florian]
- florian has joined #css
- 23:09:56 [mmielke]
- mmielke has joined #css
- 23:11:01 [fantasai]
- fantasai: Just define what happens at zero. Everything above zero is handled by pixel-rounding, which we don't speicfy
- 23:11:10 [fantasai]
- TabAtkins: So should I specify averaging color?
- 23:11:29 [fantasai]
- dbaron: Need to specify color space to average in
- 23:11:51 [fantasai]
- shepazu: So what happens when you zoom in?
- 23:11:57 [fantasai]
- TabAtkins: That depends on implementation
- 23:12:02 [ChrisL]
- @dbaron yes, you do
- 23:12:04 [fantasai]
- TabAtkins: If you zoom enough you hit rounding issues
- 23:12:41 [ChrisL]
- @tab that is always the case. welcome to the world of non-financial computing which uses fixed precision arithmetic
- 23:12:43 [fantasai]
- TabAtkins: I want to make sure ChrisL's case is handled
- 23:13:00 [fantasai]
- TabAtkins: is allowed to handle, don't know if I can require it
- 23:13:13 [fantasai]
- ..
- 23:13:19 [fantasai]
- TabAtkins: I can't must without being precise
- 23:14:21 [fantasai]
- shepazu: It's not as important to specify the behavior at 1px when ti's been defined as 1px, it's what the rendering is.
- 23:14:22 [dbaron]
- I think Peter's suggestion was good.
- 23:14:38 [fantasai]
- shepazu: Important part was ... as you zoom in that 1px width is now 50 pixels it should have the whole gradient
- 23:14:48 [fantasai]
- shepazu: You can test that -- chris just wrote a test for that.
- 23:15:03 [fantasai]
- shepazu: You haven't been dealing much with things like scale, but you're going to be, so you're going to run into this problem
- 23:15:21 [ChrisL]
- css transforms
- 23:15:33 [fantasai]
- Dean: We all know the implementers are going to do the best they can.
- 23:16:52 [fantasai]
- plinss: This is very simply specified.
- 23:16:58 [fantasai]
- plinss: You try to not overspecify it.
- 23:17:02 [ChrisL]
- s/gradient"/gradient)"
- 23:17:38 [fantasai]
- plinss: When the UA has knowledge of the output resolution, it's allowed to substitute an average color for the repeating color when the device does not have the resolution to capture the resolution correctly.
- 23:17:50 [fantasai]
- plinss: It lets everybody do the right thing to the best of their ability.
- 23:18:22 [plinss_]
- s/the resolution correctly/the gradient correctly/
- 23:18:34 [fantasai]
- RESOLVED: Accept plinss's proposal.
- 23:19:06 [fantasai]
- TabAtkins: Image values is becoming divergent in implementation stability. Some features like gradients widely implemented, others have no impls or are just beginning implementation.
- 23:19:35 [fantasai]
- TabAtkins: In the interest of getting gradients unprefixed as soon as they're sufficiently stable, I'd like to either pull gradients out into a Gradients spec, or go through and kick a bunch of stuff into css4-images
- 23:19:52 [fantasai]
- TabAtkins: pecifically, the image() function, the cross-fade() function, and image-* properties
- 23:20:29 [ChrisL]
- which is larger, the bits you cut out or the bits you leave in?
- 23:20:32 [fantasai]
- JohnJansen: I think pulling out Gradients would make it easier.
- 23:20:42 [fantasai]
- sylvaing: If you move the others, you're still splitting the document.
- 23:20:52 [fantasai]
- ChrisL, about equal
- 23:21:45 [fantasai]
- fantasai: There's some things that should move just as fast as gradients. Also you need to define <image> type.
- 23:22:04 [fantasai]
- TabAtkins: CR can reference WD, so it wouldn't delay CR, just REC
- 23:22:18 [fantasai]
- TabAtkins: I just want to get Gradients to a point where we can kill prefixes
- 23:22:32 [fantasai]
- TabAtkins: I don't care either way; WG decide for me.
- 23:23:04 [sylvaing]
- http://dev.w3.org/csswg/css3-images/
- 23:23:40 [fantasai]
- TabAtkins: So, we would keep 4.1. 4.3 is implemented in Mozilla, so keep that there. We can test it for now.
- 23:23:45 [fantasai]
- TabAtkins: 4.2 and 4.4 would go
- 23:23:49 [fantasai]
- TabAtkins: Gradients stay
- 23:24:02 [fantasai]
- TabAtkins: 6 stays; object-* implemented already, rest is image sizing algorithm
- 23:24:14 [fantasai]
- TabAtkins: Rest can be punted to level 4
- 23:24:28 [ChrisL]
- I suggest splitting into css3 graidients and css3 images. because css4 images sounds like its years away, just from the name
- 23:24:55 [fantasai]
- dbaron: Things without 2 impls should be at-risk
- 23:25:09 [fantasai]
- ChrisL, we'll publishing Selectors 4 next month ;)
- 23:25:31 [fantasai]
- TabAtkins: 9 is a separate issue we need to discuss
- 23:26:01 [fantasai]
- smfr: Seems we need cross-fade() to interpolate gradients
- 23:26:28 [fantasai]
- Florian: We could pull back into 3 if it stabilizes before CR
- 23:26:37 [ChrisL]
- @smfr you mean to interpolate as in a transition, not to draw the gradient?
- 23:27:04 [fantasai]
- plinss: Any objections to splitting along the lines described?
- 23:27:18 [smfr]
- ChrisL: my point is that interpolation of gradients and images should be in the same spec
- 23:27:22 [fantasai]
- RESOLVED: Split css3-images as described
- 23:27:58 [fantasai]
- TabAtkins: 3rd issue is about gradients directly
- 23:28:14 [bradk]
- I am
- 23:28:18 [fantasai]
- TabAtkins: I've been messing with keyword definitions for gradients
- 23:28:30 [fantasai]
- TabAtkins: Right now you can specify linear gradients by keyword or angle
- 23:28:50 [fantasai]
- TabAtkins: Some very informal polls on twitter, if you ask someone what a gradient with top should do, they agree with the spec
- 23:28:59 [fantasai]
- TabAtkins: If you ask them about angles, they agree with the spec
- 23:29:17 [fantasai]
- TabAtkins: If you ask them both at the same time, they say they should be the same, despite that being inconsistent.
- 23:29:25 [fantasai]
- TabAtkins gives numbers
- 23:29:51 [fantasai]
- TabAtkins: Given this, I think the angles are very easy to see, but the way keywords work right now is confusing
- 23:30:02 [fantasai]
- TabAtkins: So I'm thinking we should drop keywords and work them out better for level 4
- 23:30:12 [TabAtkins_]
- data:text/html,<div id=one></div><div id=two></div><style>div {margin: 50px; border: thick solid black;width: 520px;height: 300px;}#one {background: -webkit-linear-gradient(top left, red, white, blue);}#two {background: -webkit-linear-gradient(-60deg, red, white, blue);}</style>
- 23:30:17 [fantasai]
- TabAtkins: We have an issue that corner-to-corner gradients, for example, are not sufficiently pretty.
- 23:30:25 [bradk]
- he're's mine: http://www.bradclicks.com/cssplay/linear-gradient/corner-to-corner-gradient.png
- 23:30:55 [fantasai]
- TabAtkins: The top one is what corner-to-corner gradients currently do
- 23:31:08 [fantasai]
- TabAtkins: The gradient bands are perpendicular to the diagonal gradient line
- 23:31:23 [fantasai]
- TabAtkins: But the lower picture is what people actually expect/want
- 23:31:57 [fantasai]
- TabAtkins: The two are different by the angle being reflected over the line y=x
- 23:32:07 [dbaron]
- data:text/html,<div class=one></div><div class=two></div><style>div {margin: 50px; border: thick solid black;width: 520px;height: 300px;}.one {background: -moz-linear-gradient(top left, red, white, blue);}.two {background: -moz-linear-gradient(-60deg, red, white, blue);}</style>
- 23:32:10 [fantasai]
- TabAtkins: one is 30deg other is 60deg
- 23:32:13 [ChrisL2]
- ChrisL2 has joined #css
- 23:32:16 [dbaron]
- (for those who don't want to use WebKit to look at it :-)
- 23:32:26 [fantasai]
- Brad: Another way to say is that for corner to corner, the hypothetical midpoint connects the other two corners
- 23:32:58 [fantasai]
- TabAtkins: With the current spec, the further the rectangle is from a square, the more it looks like a sideways gradient rather than a corner-to-corner one
- 23:33:19 [fantasai]
- TabAtkins: Given this issue, I want to drop keywords for now and address them in level 4
- 23:33:42 [fantasai]
- bradk: Keywords are the most popular way of doing this right now, mostly vertical or horizontal gradients.
- 23:34:15 [fantasai]
- Brian: Instead of up top and bottom, use upwards and downwards and instead of left and right use leftward rightward
- 23:34:24 [fantasai]
- Brian: And remove the paired keyword corner options
- 23:35:23 [fantasai]
- plinss: If we do partial keyword option, then we're locking in our syntax and it'll be incompatible with our future syntax.
- 23:35:54 [fantasai]
- plinss: We'll get into a situation where we dislike our legacy keywords and can't change them.
- 23:36:03 [fantasai]
- TabAtkins: Could use a different function
- 23:36:16 [fantasai]
- dbaron: Authors really want gradients. At some point we need to stop fiddling with it and declare it ready.
- 23:36:18 [arno1]
- We could just have "horizontal" and "vertical" as keywords
- 23:36:29 [fantasai]
- Florian: We're not throwing out all of it, just part of it
- 23:36:35 [fantasai]
- Florian: just the part we don't agree on it.
- 23:37:13 [dbaron]
- anne: We did that with border-radius, and by the time we got it sorted out they weren't popular anymore. :-)
- 23:37:14 [fantasai]
- TabAtkins: We solve the cases we know we need now, get them unprefixed, then see if the remaining cases are useful enough to care
- 23:38:06 [fantasai]
- Brad: the only gradient generators online use the keyword types right now, and because we changed the angle definition, you can't get something that works
- 23:38:15 [fantasai]
- Brad: People use prefixed versions for years.
- 23:38:22 [fantasai]
- TabAtkins: We're not going to use prefixed versions
- 23:38:37 [fantasai]
- Brad: Because we changed meaning of degress, you can't make a backwards compat version
- 23:38:54 [fantasai]
- TabAtkins: You write the prefixed versions with the old syntax, do unprefixed version with spec version
- 23:39:45 [fantasai]
- various try to explain that prefixed versions aren't going to change; they'll change when they drop the prefix
- 23:40:26 [smfr]
- Zakim: gavel
- 23:40:37 [fantasai]
- smfr, you forgot to use a comma
- 23:40:44 [fantasai]
- :P
- 23:41:16 [fantasai]
- plinss: How vendors deal with them and their customers is up to them.
- 23:41:44 [fantasai]
- Brad: I don't think we can pretend what we decide here is not going to have ramifications. If we don't have a committment that they won't do that... otherwise we wind up with something unusable
- 23:41:48 [fantasai]
- TabAtkins: That's why nobody will do that.
- 23:42:12 [fantasai]
- dbaron: We've done it plenty of times, but we probably wouldn't do it with this one
- 23:42:22 [fantasai]
- Anne: It worked fine for border-radius, -moz-opacity, etc.
- 23:42:42 [fantasai]
- Brad: I'd like to keep the keyword going and figure it out before we publish it.
- 23:43:03 [fantasai]
- Arron: Do you have a problem with upwards/downwards/etc
- 23:43:39 [fantasai]
- Brad: As plinss pointed out, it'll make it harder for us to get a consistent syntax later when we add corner-to-corner gradients
- 23:44:00 [fantasai]
- TabAtkins: The future has potential. I'm confident I'm not blocking my ability to expand in the future.
- 23:44:11 [fantasai]
- Brad: I don't see the syntax as being problematic.
- 23:45:11 [fantasai]
- Brad: I don't think it needs to change.
- 23:46:15 [fantasai]
- Brad: It's intuitive for a lot of people, and they've learned it, and do we really want to hold up the spec for this.
- 23:46:53 [fantasai]
- ...
- 23:47:08 [fantasai]
- smfr: I'm concerned about having multiple gradient functions. Do I need linear-gradient or corner-gradient?
- 23:47:12 [fantasai]
- smfr: confusing
- 23:48:16 [fantasai]
- TabAtkins: I think we can have a syntax that's different enough there won't be a parsing ambiguity
- 23:48:42 [fantasai]
- Brad: "from top" would be close to current syntax and resolve ambiguity
- 23:49:14 [fantasai]
- Brad: "left" -> "rightwards" is more different
- 23:49:35 [fantasai]
- plinss: The question was about whether to drop the keywords, not about what they should be
- 23:49:55 [fantasai]
- plinss: So do we drop the keywords, or go offline and figure it out
- 23:50:10 [fantasai]
- TabAtkins: I want to drop the corner keywords
- 23:50:24 [fantasai]
- plinss: Sounds like we're don't have consensus
- 23:50:47 [fantasai]
- plinss: Take it to email/telecon
- 23:50:54 [fantasai]
- plinss: Any other gradient issues?
- 23:51:02 [fantasai]
- Brian: Premultiplied? Keep it don't keep it have options?
- 23:51:26 [fantasai]
- TabAtkins: I think premultiplied produces most attractive gradients in common cases blending with transparent.
- 23:52:02 [fantasai]
- TabAtkins: I can go either way, depending on implementers
- 23:52:21 [fantasai]
- fantasai: If you go other way, then make 'transparent' magic.
- 23:52:40 [fantasai]
- TabAtkins: I don't want to make transparent magic.
- 23:54:00 [fantasai]
- ?: Drop 'transparent' keyword, make everybody use rgba()
- 23:54:14 [dbaron]
- Tab: If we're going to go non-premultiplied, then disallow use of 'transparent' and make everybody use rgba()
- 23:54:17 [fantasai]
- fantasai: That's horrible. You're making the author do the work.
- 23:54:30 [fantasai]
- ??: You have the same problem with transitions.
- 23:54:47 [fantasai]
- fantasai: gradient from color to tranparent is very common use case
- 23:55:12 [fantasai]
- smfr: Reality is that if we go with premultiplied we won't have it working correctly on Mac for another year or so
- 23:55:15 [ChrisL]
- ChrisL has joined #css
- 23:55:29 [fantasai]
- anne: If it's just a year, I say head for the future.
- 23:55:57 [fantasai]
- Vincent: Example of where premultiplied looks better?
- 23:56:16 [fantasai]
- TabAtkins: yellow to transparent, in premultiplied goes from yellow, fading to transparent
- 23:56:31 [nimbupani]
- http://jsfiddle.net/rK9Pd/11/show/
- 23:56:34 [fantasai]
- TabAtkins: in non-premultilied, goes through an ugly grayish greeny color
- 23:56:36 [nimbupani]
- (check in webkit vs opera)
- 23:57:17 [fantasai]
- TabAtkins: If you go red - transparent - blue, if you want to make it work correct in non-premultiplied, you have to write red, transparent-red, transparent-blue, blue, placing transparent-blue and transparent-red at the same spot in the gradient
- 23:57:51 [fantasai]
- plinss: Why don't we want to do premultiplied?
- 23:58:03 [fantasai]
- TabAtkins: It's not natively available on some platforms
- 23:58:10 [fantasai]
- smfr: We'd have to get libraries to add it
- 23:58:29 [glazou]
- glazou has joined #css
- 23:59:28 [fantasai]
- Brian: Alan Gresley was asking about non-premultiplied gradients because there are some cases that you'd want that result
- 23:59:53 [fantasai]
- TabAtkins: You could simulate it by manually arc it through the color space, it's tricky, but doable, and a very uncommon case in comparison
- 00:00:13 [fantasai]
- Brian: I had good results with Opera
- 00:00:35 [fantasai]
- TabAtkins: With Opera and IE we'll have 2 impls, so let's stick with that.
- 00:00:45 [fantasai]
- RESOLVED: Use premultiplied colors for gradients and transitions
- 00:01:20 [fantasai]
- shepazu: At the outset of this effort, there was discussion about remaining compatible with SVG gradients.. was that abandoned?
- 00:01:26 [fantasai]
- TabAtkins: No. SVG doesn't have alpha colors.
- 00:01:40 [fantasai]
- shepazu: Talking about geometry
- 00:02:03 [fantasai]
- TabAtkins: Yeah, I don't think that's useful. But I was going to talk tomorrow about using SVG paint servers in CSS or CSS gradients in SVG.
- 00:02:18 [fantasai]
- shepazu: So an engine supporting both will have to support two different things.
- 00:02:20 [fantasai]
- TabAtkins: Yes.
- 00:02:33 [fantasai]
- plinss: Ok, let's discuss that tomorrow. Grids.
- 00:02:36 [fantasai]
- Topic: Grids
- 00:04:04 [fantasai]
- Phil: My name is Phil and I'm from MS.
- 00:04:59 [fantasai]
- Phil: We recently published new editor's draft. Hoping none of it's controversial so we can go through it quickly.
- 00:05:11 [fantasai]
- Phil: With one possible reduction in functionality in grid template property
- 00:05:38 [fantasai]
- Phil: specifically, assinging display types to grid pseudos
- 00:06:30 [fantasai]
- ACTION Phil: Post notes to www-style or www-archive so they can be put in the minutes
- 00:06:30 [trackbot]
- Sorry, couldn't find user - Phil
- 00:06:39 [smfr]
- http://dev.w3.org/csswg/css3-grid-align/
- 00:06:48 [fantasai]
- Phil: 7.2 covered explicitly defined grid cells, creating named grid cells
- 00:06:59 [fantasai]
- Phil: You could place children of the grid into them. We're removing that
- 00:07:15 [fantasai]
- Phil: Removing grid-stacking property, which said which layout this explicitly-defined grid cell would be using
- 00:07:33 [fantasai]
- Phil: The other mode it had was layer, which was the default layout type for a grid cell so items would layer on top of others
- 00:07:52 [fantasai]
- Phil: When we presented that at MV ...
- 00:08:11 [fantasai]
- Phil: Talked about creating flows inside a grid cell
- 00:08:30 [fantasai]
- Phil: And assigning display types to the grid cell
- 00:08:30 [glazou]
- glazou has joined #css
- 00:08:51 [fantasai]
- Phil: This was about creating presentation-specific structure through declaration of these grid cells, trying to remove concept form the grid layout spec
- 00:09:25 [fantasai]
- Phil: We also added new paragraph about grid cell concept. We still have logicla notion of a grid cell, but it's just an alias syntax for referring to a region of the grid.
- 00:09:32 [fantasai]
- Phil: But we're saying it's not stylable
- 00:09:39 [fantasai]
- Phil: It's just a way to name a spot on the grid
- 00:10:01 [fantasai]
- Phil: In section 6.4 , repeating columns and rows. e got some ffeedback that the name dlines syntax we had didn't make sense inside repeating syntax
- 00:10:21 [fantasai]
- Phil: Since it defined that only the firs toccurance of the name would be honore, and purpose of repeat syntax is to replay the grid lines over
- 00:10:34 [fantasai]
- Phil: Added issue as te whether to remove that ability from there
- 00:10:44 [fantasai]
- Phil: Any feedback, send it, otherwise we'll remove
- 00:10:54 [fantasai]
- Phil: We also had another isuse on the grammar, for grid columns and grid rows
- 00:11:02 [fantasai]
- Phil: So we changed from ... to ...
- 00:11:15 [fantasai]
- Phil: I'll let you figure it out; trust me the new one is better.
- 00:11:29 [fantasai]
- Phil: Section 7.1 anonymous grid cells, we just added some language about relation between grid cells
- 00:11:38 [fantasai]
- Phil: Now it's just logical container, what does it do to grid items inside.
- 00:11:44 [fantasai]
- Phil: Define them as containing block
- 00:11:57 [fantasai]
- Phil: And we said how they came ito being, added language defining dminesions of the grid cell etc.
- 00:12:35 [fantasai]
- Phil: Next is explicitly defined grid cells, 7.2 is gone
- 00:12:43 [fantasai]
- Phil: Defining grid cells with a template is still there
- 00:12:48 [fantasai]
- Phil: Still use the ascii-art syntax
- 00:12:59 [fantasai]
- Phil: Keeping around grid-cell proeprty, just don't have pseudo-element selector
- 00:13:05 [fantasai]
- Phil: Can still define grid, and put things in it
- 00:13:08 [fantasai]
- Phil: witht hat
- 00:13:18 [fantasai]
- Phil: Section 7.5 automatic placement of grid items.
- 00:13:27 [fantasai]
- Phil: It's no longer what we're thinking, so removed note
- 00:13:34 [fantasai]
- Phil: ...
- 00:13:51 [fantasai]
- Phil: Importance of automatic placement, some language about fixup of grid; it matches language in flexbox
- 00:14:07 [fantasai]
- Phil: And we noted that if we don't have this fature, the fixup isn't useful
- 00:14:21 [fantasai]
- Phil: Everything just gets dumped into first grid row. We're planning to leave in, so this is just an observation
- 00:14:31 [fantasai]
- Phil: I just renamed section 8.1 .. size of grid items.
- 00:14:39 [fantasai]
- Phil: Needs more work; need to specify box model calculations
- 00:14:53 [fantasai]
- Phil: So if going to be stretched, questions like whta if you have a replaced element with intrinsic ratio ...........
- 00:15:25 [fantasai]
- Phil: 10 calculating isze of grid tracks. Don't know if anyone is implementing, or thinking of implementing, but if so, we've published osme pseudo-code about sizing these bits of the grid.
- 00:15:31 [fantasai]
- Phil: Some bugs in it still
- 00:15:42 [fantasai]
- Phil: We'll update a few more times before pushing to WD
- 00:15:48 [fantasai]
- Phil: Lastly all these changes captured in Appendix A
- 00:16:21 [fantasai]
- Phil: That is everything we have changed. The biggest piece is obviously the inability now to create these grid cells that you can give a display inside to control how they layout their contents, does anybody have objections to removing concept from the spec?
- 00:16:31 [fantasai]
- TabAtkins: Ok with it, majority of my use cases don't need it.
- 00:16:48 [fantasai]
- TabAtkins: But if I ant to use this with regions, I need to nwo insert dummy elements to position them with grid ...
- 00:16:58 [fantasai]
- TabAtkins: I greatly disagree that I should put dummy divs in my doc
- 00:17:15 [fantasai]
- Markus: We think that should be defined somewhere other than grid
- 00:17:42 [fantasai]
- TabAtkins: I hate junk put into your page for the sole purpose of styling.
- 00:18:29 [fantasai]
- Tab tries to explain the difference between semantic markup and stylistic presentation to the MS folks
- 00:18:54 [fantasai]
- Steve: Aren't templates a little bit of both?
- 00:19:55 [fantasai]
- TabAtkins: As long as we keep in mind that we might want to do this more generalized in the future, then I'm cool.
- 00:20:00 [bradk]
- If the CSS is creating the pseudo-elements, then conceivably more regions can be created to accomodate more content.
- 00:20:03 [fantasai]
- Vincent: We have a resolution on this from the morning.
- 00:20:14 [fantasai]
- ...
- 00:21:02 [fantasai]
- Alex: One issue that we discovered ... which was that alignment in grid is currently different from alignment in flexbox
- 00:21:25 [fantasai]
- Alex: What we discussed yesterday flexbox alignment, we kinda liked the idea of what grid is doing now. We're going to come up with a proposal that will make something consistent between the two.
- 00:21:37 [fantasai]
- TabAtkins: Don't think they need to be different when both flexing and aligning stuff
- 00:22:00 [fantasai]
- Markus: Q for Peter, we experiment with named lines, and ended up with very long strings.
- 00:22:31 [fantasai]
- Markus: Is ther ea way to shortcut this somehow? Authoring becomes awkward.
- 00:22:45 [fantasai]
- Phil: This example uses template
- 00:23:08 [fantasai]
- Phil: It only takes one letter to name a position
- 00:23:22 [fantasai]
- Phil: with named grid lines
- 00:23:35 [fantasai]
- Phil: You end up putting 4 strings for each item that you had
- 00:23:55 [fantasai]
- Phil: It gets a little verbose
- 00:24:04 [fantasai]
- Phil: I think in practice if you have a grid, and have a large number of grid items
- 00:24:24 [fantasai]
- Phil: And don't want to renumber them, you probably won't use named lines anyway
- 00:24:33 [fantasai]
- TabAtkins: What wins if you use grid-rows, grid-columns
- 00:24:41 [fantasai]
- Phil: If the grid cell exists, then it wins
- 00:24:53 [fantasai]
- TabAtkins: What properties apply to grid templae ??
- 00:25:41 [fantasai]
- ...
- 00:25:50 [fantasai]
- Phil: You can only use ascii letters in grid template
- 00:26:04 [fantasai]
- Florian: Disallowing punctuation doesn't disallow Chinese characters or accented characters...
- 00:26:22 [fantasai]
- TabAtkins: Most reasonable layouts won't use 26, but last year I was looking at some very complicated layouts that were pushing it
- 00:26:29 [fantasai]
- Arron: I had 180 in a layout I did
- 00:27:12 [fantasai]
- TabAtkins: I think you should say any single character
- 00:27:21 [fantasai]
- fantasai: grapheme cluster
- 00:27:36 [fantasai]
- ...
- 00:27:37 [dbaron]
- I wouldn't allow more than characters that are allowed to start identifiers.
- 00:28:02 [fantasai]
- Phil: Would you use numbers ... ?
- 00:29:52 [fantasai]
- dbaron: That includes most of Unicode except some ascii punctuation and stuff
- 00:30:30 [fantasai]
- plinss: Where it gets unweildy with grid names, it's not so much where they're defined as where they're used
- 00:31:18 [dbaron]
- template: "北北北北" "西中中东" "西中中东" "西中中东" "西南南东";
- 00:31:19 [fantasai]
- ...
- 00:31:28 [fantasai]
- plinss: I'm not concerned about the verbosity of the grid columns declaration
- 00:31:35 [fantasai]
- plinss: I fyou line it up like that, I think it's readable
- 00:31:41 [fantasai]
- plinss: Later when you do the grid-column ...
- 00:31:59 [fantasai]
- Phil: You have up to for lines for positioning, so that's not too bad.
- 00:32:10 [fantasai]
- Phil: But if you have a lot of grid items, that gets quite large
- 00:32:17 [fantasai]
- plinss: Depends on how large your grid is
- 00:32:48 [fantasai]
- Phil: No issues open on it, these are just some comments we got back
- 00:32:58 [fantasai]
- plinss: Maybe you can have something that maps grid cells back into named grid lines
- 00:33:09 [fantasai]
- plinss: Right now you can only do that with template.
- 00:33:21 [fantasai]
- plinss: What if you could define a grid-cell "foo" defiend by these named lines
- 00:33:36 [fantasai]
- plinss: You're not defining the grid template, so not single-letter names; use multi-character names
- 00:33:47 [fantasai]
- Phil: We had ::grid-cell() pseudo that could do that
- 00:33:55 [fantasai]
- Phil: It was tyleable enough that you coudl specify the ...
- 00:34:07 [fantasai]
- Phil: Ultimately we pulled that out
- 00:34:24 [fantasai]
- Phil: Anything else?
- 00:34:37 [fantasai]
- TabAtkins: Alex and I were talking about dropping the 'fr' unit and just using the flex() function
- 00:35:07 [fantasai]
- for flexbox
- 00:35:27 [fantasai]
- Alex: Flex function can do everything fr unit can do in flex, but fr unit is just a shortcut for one of the use of flex function
- 00:35:40 [dbaron]
- Isn't 'fr' a shortcut for one of the uses of the flex() function that requires all three arguments?
- 00:35:47 [fantasai]
- Alex: In a grid fr is a basic of how it is calculated
- 00:36:10 [fantasai]
- Alex: it would be .. also allow flex function in grid, in which case you could replace fr unit with it or have both
- 00:36:27 [fantasai]
- Phil: right now we have minmax function, and cna mix that with min-content, max-content keywords
- 00:36:39 [fantasai]
- Phil: Would we introduce flex() function there?
- 00:36:51 [fantasai]
- Alex: flex function has flexibility and preferred size
- 00:36:54 [fantasai]
- Alex: ...
- 00:37:21 [fantasai]
- Alex: Auto and flexibility at once.. produces the same calculation .. maximum, however preferred size of column based on content
- 00:37:29 [fantasai]
- Phil: There's just an issue we haven't resolved in the spec now
- 00:37:59 [fantasai]
- Alex: For flexbox you don't specify 50 values for an items, so flex() being longer than a unit is not a big deal, but on a grid probably woudln't want to replace fr unitl with flex
- 00:38:03 [fantasai]
- Alex: Make it even longer
- 00:38:19 [fantasai]
- Alex: So if in grid definition fr is shortcut for flex(), we could kepe it for flex
- 00:38:37 [fantasai]
- TabAtkins: I'm uncomfortable with two of us using different notions of flex
- 00:39:02 [fantasai]
- dbaron: fr right now is equivalent to flex() on a basis of 0, and 0 is not the defautl for that third argument
- 00:39:12 [fantasai]
- TabAtkins: I changed flex() ...
- 00:39:24 [fantasai]
- TabAtkins: If you say flex(1), you start from zero
- 00:39:35 [fantasai]
- TabAtkins: That's not what my spec says right now
- 00:39:44 [fantasai]
- Alex: ...
- 00:39:50 [fantasai]
- TabAtkins: Ok, we need to discuss this
- 00:40:00 [fantasai]
- dbaron: I agree with Alex, but I think that's another reason to keep fr.
- 00:40:21 [dbaron]
- s/.../the omitted preferred width should default to initial value/
- 00:40:38 [fantasai]
- Arron: We submitted our proposal for floats and positioning, and we decided to split those. Alex volunteered me to do the editing
- 00:40:41 [arronei]
- http://www.interoperabilitybridges.com/css3-positioning/
- 00:40:57 [glazou]
- glazou has joined #css
- 00:41:05 [fantasai]
- Arron: So you can take a look at it.Not too many differences from 2.1
- 00:41:24 [fantasai]
- Arron: Most of it's what 2.1 says, with some little clarifications of handling our new value. So you won't see a lot of differences.
- 00:41:30 [fantasai]
- Arron: Will jump straight to some changes.
- 00:41:38 [fantasai]
- Arron: So our first change that we have here is page positioning.
- 00:41:52 [fantasai]
- Arron: What this defines is a way to go straight for the initial containing block, much like fixed positioning does but without fixing on the page
- 00:42:00 [fantasai]
- Arron: So size and positioning based on initial containing block
- 00:42:15 [fantasai]
- Arron: It'll go all the way to the initial containing block, and positioning off of that box.
- 00:42:27 [fantasai]
- Arron: That's the first part.
- 00:42:39 [fantasai]
- smfr: How is it different from position: fixed
- 00:42:54 [glazou]
- http://www.interoperabilitybridges.com/css3-positioning/ (trailing slash DOES matter)
- 00:43:06 [fantasai]
- Rossen: If we create a region each the size of containing block, you want to have one positioned to the current region
- 00:43:19 [fantasai]
- Rossen: Don't want to rely on how many relative and abspos elements in your ancestor chain
- 00:43:32 [tantek]
- lol: http://twitter.com/getify/status/95648509319069696 <center> vs CSS
- 00:43:48 [fantasai]
- Rossen: Different from position: fixed because in scrolling media, it's fixed to initial contianing block. So element scrolls along with the rest of the content
- 00:44:03 [fantasai]
- Rossen: Similar to abspos content that has no positioned ancestor
- 00:44:13 [fantasai]
- Rossen: Whereas fixed replicates
- 00:44:52 [fantasai]
- Rosse: Position page was specific to design so you can target only the current page, and unlike position: fixed; position: paged; elements ... you can position negatively, and you can overflow, and that's just fine, whereas fixed position that just clips
- 00:45:03 [fantasai]
- dbaron: So it's positioned on the current page relative to its placeholder.
- 00:47:05 [fantasai]
- fantasai: My concern is that you'll get a layout that makes sense on paged media, but breaks in scrolled. Opposite problem of fixed positioning.
- 00:47:32 [jdaggett]
- jdaggett has joined #css
- 00:47:46 [fantasai]
- fantasai: e.g. if you use this on a 15-page document, positioning 15-20 things throughout the document across the 15 pages, looks fine
- 00:48:11 [fantasai]
- fantasai: Load that into a scrolled view, and everything piles on top of each other in the first screenful, and then scrolls away with the rest of the document empty
- 00:48:39 [fantasai]
- ...
- 00:48:51 [fantasai]
- Alex: We should have discusison of what are different aspects of paged media
- 00:49:02 [fantasai]
- Alex: Some means it's non-interactive, you're on paper
- 00:49:07 [fantasai]
- Alex: Some of it means you're paginated
- 00:49:21 [fantasai]
- Alex: From pov of layout, it's tempting to apply paged media to regions. But it's not paged media
- 00:49:58 [fantasai]
- Alex: Should we have a new media type?
- 00:50:03 [fantasai]
- fantasai: I think so.
- 00:50:20 [fantasai]
- Arron: What I"m trying to ge there is, we have this defintion right now, it may need more work as fantasai points out.
- 00:50:47 [fantasai]
- Arron: What I'd like to do is put it up on W3C as an actual editor's draft, so we can start to shape something that works
- 00:51:02 [fantasai]
- plinss: Anybody objecting?
- 00:51:12 [fantasai]
- RESOLVED: Put CSS3 Positioning draft on dev.w3.org
- 00:51:21 [fantasai]
- Arron: That's pretty much it. Please review once it's up
- 00:51:33 [fantasai]
- plinss: Some flexbox issues to get back to?
- 00:51:37 [fantasai]
- TabAtkins: Haven't had time.
- 00:51:44 [fantasai]
- Alex: Directions and alignment; need to ...
- 00:52:01 [fantasai]
- plinss: Meeting closed.
- 00:52:05 [smfr]
- smfr has left #css
- 01:07:31 [karl]
- karl has joined #CSS
- 01:19:36 [dino]
- dino has joined #css
- 02:21:49 [smfr]
- smfr has joined #css
- 02:23:23 [smfr]
- smfr has left #css
- 03:11:30 [miketaylr]
- miketaylr has joined #css
- 03:53:38 [jdaggett_]
- jdaggett_ has joined #css
- 05:27:51 [arronei]
- arronei has joined #css
- 05:44:56 [plinss_]
- plinss_ has joined #css
- 05:52:10 [glazou]
- glazou has joined #css
- 05:54:27 [tantek]
- tantek has joined #css
- 06:19:22 [dino_]
- dino_ has joined #css
- 06:24:44 [dino_]
- dino_ has joined #css
- 06:38:43 [konaya]
- konaya has joined #css
- 06:44:34 [kojiishi]
- kojiishi has joined #css
- 07:33:22 [kojiishi]
- kojiishi has joined #css
- 07:37:54 [Jorrit]
- Jorrit has joined #css
- 07:38:09 [Jorrit]
- Jorrit has left #css
- 10:03:22 [Ms2ger]
- Ms2ger has joined #css
- 11:29:01 [karl]
- karl has joined #CSS
- 11:29:25 [karl]
- karl has joined #CSS
- 11:47:49 [Martijnc]
- Martijnc has joined #css
- 13:09:22 [miketaylr]
- miketaylr has joined #css
- 13:37:53 [anne]
- anne has joined #css
- 13:52:19 [nimbupani]
- nimbupani has joined #css
- 15:05:59 [oyvind]
- oyvind has joined #css
- 15:06:04 [oyvind]
- oyvind has left #css
- 15:09:04 [dino]
- dino has joined #css
- 15:39:02 [cjon]
- cjon has joined #css
- 15:40:11 [JohnJansen]
- JohnJansen has joined #css
- 15:44:52 [Martijnc]
- Martijnc has joined #css
- 15:46:28 [nimbupani]
- nimbupani has joined #css
- 15:49:11 [anne]
- anne has joined #css
- 15:49:21 [plinss_]
- plinss_ has joined #css
- 15:56:59 [JohnJansen]
- JohnJansen has left #css
- 15:58:20 [dino]
- dino has joined #css
- 15:59:38 [tantek]
- tantek has joined #css
- 15:59:55 [tantek]
- heading over - ETA 10min
- 16:01:11 [tpod]
- tpod has joined #css
- 16:03:57 [arronei_]
- arronei_ has joined #css
- 16:04:20 [arno]
- arno has joined #css
- 16:04:59 [bradk]
- bradk has joined #css
- 16:06:18 [bradk]
- hello?
- 16:06:53 [Ms2ger]
- Good evening
- 16:07:24 [bradk]
- Is the meeting starting?
- 16:08:03 [ed]
- in a moment
- 16:08:06 [Ms2ger]
- tantek said he'd be there in a couple of minutes
- 16:08:08 [sylvaing]
- sylvaing has joined #css
- 16:08:24 [plinss_]
- we'll be using the #fx channel for today's meeting
- 16:09:12 [bradk]
- Please let me know when there is a skype call ready to connect.
- 16:09:52 [tpod]
- tpod has joined #css
- 16:10:36 [hober]
- me too :0
- 16:10:40 [hober]
- err, :)
- 16:11:53 [dbaron]
- dbaron has joined #css
- 16:11:57 [tpod]
- How was the boat?
- 16:11:57 [smfr]
- smfr has joined #css
- 16:12:05 [anne]
- meeting is in #fx
- 16:12:20 [sylvaing]
- Anne sank one Microsoft employee in the lake: success
- 16:12:40 [sylvaing]
- Brad, i'm on Skype
- 16:13:00 [TabAtkins_]
- TabAtkins_ has joined #css
- 16:13:08 [tpod]
- anne: Pls set topic to say #fx
- 16:16:59 [szilles]
- szilles has joined #css
- 16:17:17 [florian]
- florian has joined #css
- 16:17:54 [shans]
- shans has joined #css
- 16:20:49 [tantek]
- tantek has joined #css
- 16:24:08 [alexmog]
- alexmog has joined #css
- 16:27:02 [dsinger]
- dsinger has joined #css
- 16:38:17 [szilles]
- szilles has joined #css
- 16:38:41 [ChrisL]
- ChrisL has joined #css
- 16:56:26 [szilles]
- szilles has joined #css
- 17:01:13 [shepazu]
- shepazu has joined #css
- 17:02:54 [stearns__]
- stearns__ has joined #css
- 17:11:39 [szilles]
- szilles has joined #css
- 17:49:21 [szilles]
- szilles has joined #css
- 17:57:55 [alexmog]
- alexmog has joined #css
- 18:02:02 [szilles]
- szilles has joined #css
- 18:02:59 [fantasai]
- dbaron: Is there a way to make it obvious when font-fallback happens? I considered adding Ahem to the fallback list, but it doesn't cover all of Unicode.
- 18:03:21 [dbaron]
- fantasai, we have a new api for determining the fonts that actually got used for a range of text
- 18:03:39 [dbaron]
- fantasai, beyond that, not sure
- 18:15:37 [hyatt]
- when you don't even know the height of your block yet
- 18:22:59 [konaya]
- konaya has joined #css
- 18:26:46 [szilles]
- szilles has joined #css
- 18:43:25 [szilles]
- szilles has joined #css
- 18:58:58 [arno]
- arno has joined #css
- 19:06:28 [dbaron]
- hyatt, well, yesterday I was the only one against it...
- 19:06:54 [hyatt]
- seems like a different layout model to me
- 19:07:08 [hyatt]
- like using left/top seems fine etc.
- 19:07:20 [hyatt]
- but it's not really laying out at the same time a normal positioned object would
- 19:07:43 [szilles]
- szilles has joined #css
- 19:07:45 [hyatt]
- just trying to understand
- 19:07:47 [hyatt]
- if it's 2-pass
- 19:07:48 [hyatt]
- or
- 19:07:51 [hyatt]
- i dunno
- 19:08:11 [dbaron]
- it's either 2-pass and suboptimal or abitrary-N-pass and correct
- 19:08:56 [dbaron]
- the draft hasn't specified which yet
- 19:13:00 [fantasai]
- :)
- 19:14:32 [szilles]
- szilles has joined #css
- 19:20:31 [hyatt]
- dbaron: is the expectation that positioned floats affect positioned descendants as well or only the normal flow
- 19:20:52 [hyatt]
- there's some confusing sentence in there about block formatting contexts
- 19:21:13 [hyatt]
- "A positioned-float box intersects with other elements in the same or other block formatting contexts. "
- 19:21:14 [dbaron]
- positioned descendants of what?
- 19:21:19 [hyatt]
- i have no idea what that sentence means
- 19:21:25 [dbaron]
- (That said, my response should probably be: you're asking *me*?)
- 19:21:29 [hyatt]
- "or other block formatting contexts"
- 19:21:44 [hyatt]
- sounds like they expect it to actually affect things other than just your normal flow descendants
- 19:22:02 [arron]
- arron has joined #css
- 19:22:12 [hyatt]
- but maybe they just mean normal flow objects that establish bfcs
- 19:22:13 [hyatt]
- like overflow
- 19:22:15 [hyatt]
- or columns
- 19:34:55 [tpod]
- tpod has joined #css
- 19:41:38 [tpod_]
- tpod_ has joined #css
- 20:03:44 [nimbupani]
- nimbupani has joined #css
- 20:10:40 [stearns]
- stearns has joined #css
- 20:35:03 [anne]
- anne has joined #css
- 20:35:04 [szilles]
- szilles has joined #css
- 20:35:20 [tantek]
- tantek has joined #css
- 20:37:30 [smfr]
- smfr has joined #css
- 20:38:12 [TabAtkins_]
- TabAtkins_ has joined #css
- 20:41:19 [ChrisL]
- ChrisL has joined #css
- 20:43:17 [alexmog]
- alexmog has joined #css
- 20:58:06 [szilles]
- szilles has joined #css
- 21:33:43 [tantek_]
- tantek_ has joined #css
- 21:45:11 [szilles]
- szilles has joined #css
- 22:25:46 [szilles]
- szilles has joined #css
- 22:30:07 [kimberlyblessing]
- kimberlyblessing has joined #css
- 22:43:29 [szilles]
- szilles has joined #css
- 22:54:34 [szilles]
- szilles has joined #css
- 23:07:15 [tpod]
- tpod has joined #css
- 23:09:38 [tpod_]
- tpod_ has joined #css
- 23:34:59 [szilles]
- szilles has joined #css
- 23:41:24 [tpod]
- tpod has joined #css
- 00:06:16 [tpod_]
- tpod_ has joined #css
- 00:25:46 [alexmog]
- alexmog has joined #css
- 00:25:53 [tantek]
- tantek has joined #css
- 00:43:20 [miketaylr]
- miketaylr has joined #css
- 00:55:03 [stearns]
- stearns has joined #css
- 00:55:10 [stearns]
- stearns has left #css
- 01:11:13 [dino]
- dino has joined #css
- 03:21:59 [arron]
- arron has joined #css
- 03:31:22 [dbaron]
- dbaron has joined #css
- 03:35:21 [nimbupani]
- nimbupani has joined #css
- 04:20:45 [anne]
- anne has joined #css
- 04:51:03 [tantek]
- tantek has joined #css
- 05:11:20 [alexmog]
- alexmog has joined #css
- 05:13:43 [TabAtkins_]
- TabAtkins_ has joined #css
- 05:14:21 [florian]
- florian has joined #css
- 05:14:50 [florian]
- florian has left #css
- 05:15:19 [plinss_]
- plinss_ has joined #css
- 05:21:31 [anne]
- anne has joined #css
- 05:30:58 [arronei_]
- arronei_ has joined #css
- 05:38:56 [arron]
- arron has joined #css
- 06:24:46 [homata]
- homata has joined #CSS