- From: Alan Stearns <stearns@adobe.com>
- Date: Thu, 2 Oct 2014 20:59:20 +0000
- To: Håkon Wium Lie <howcome@opera.com>
- CC: Christoph Päper <christoph.paeper@crissov.de>, www-style list <www-style@w3.org>
On 10/2/14, 1:38 PM, "Håkon Wium Lie" <howcome@opera.com> wrote: >Also sprach Alan Stearns: > > > > > Yes, that’s correct. Perhaps you’d add a class to the elements >that should > > > > snap their lines (this would work for opt-out as well). I’m still >in favor > > > > of opt-in, given Dave’s evaluation that there would be fewer >elements > > > > using the grid than not. Do you want to argue for the opt-out >scheme with > > > > a single property? > > > > > >The one-property solution described in CSS Books gives you both opt-in > > >and opt-out so there's not much reason to argue. > > > > I’m perfectly willing to continue arguing for two properties instead of > > one :) > > > > I’m interested to know whether Mr. Päper has a preference. > >My point is that the solution with the single property supports both >opt-in and opt-out. Noone has proposed a single-proporty solution with >only opt-out. As I point out below, your single-property solution is only opt-in for a named (or keyworded) grid. That’s not opt-in enough. > > > >Opt-in: > > > > > > p, dt, dd, li, blockquote, pre { baseline-snap: root } > > > > This only allows for opt-in for the root grid (or some other named >grid). > > If (as I’d like) we aren’t going to do named grids for the first level, > > then you need two properties for opt-in. > >'root' is just a keyword, it's not part of a generic grid-naming >system with namespace complexities and such. It still introduces all the complexity of specifying and implementing how box and line snapping occur when more than one grid can be referenced. > >One extra keyword is a small price to pay to avoid an extra property. The one keyword only works if what you want to use is a root grid. I expect that most grids used will be established on an element below the root or body. > > >Also, by having that extra keyword we solve Liam's use case. I don’t think it does, for the reasons I pointed out yesterday [1]. > >And, it's implemented. The implementation in WebKit uses two properties, but I’d rather discuss what we should specify. Thanks, Alan [1] http://lists.w3.org/Archives/Public/www-style/2014Oct/0026.html
Received on Thursday, 2 October 2014 20:59:50 UTC