- From: Alan Stearns <stearns@adobe.com>
- Date: Mon, 30 Jul 2012 11:35:45 -0700
- To: "L. David Baron" <dbaron@dbaron.org>, "www-style@w3.org" <www-style@w3.org>
On 7/30/12 11:17 AM, "L. David Baron" <dbaron@dbaron.org> wrote: >On Monday 2012-07-30 16:53 +0000, Rossen Atanassov wrote: >> This is an update to Issue 15183 >>(https://www.w3.org/Bugs/Public/show_bug.cgi?id=15183). >> >> When reviewing the processing model with the co-editors of the spec we >>couldn't find any technical reason to keep this issue active. The issue >>statement is not correct since we do not require nor suggest that >>exclusions follow the CSS 2.1 absolute positioning. We simply don't >>forbid it and when used in that combination authors can achieve >>compelling typographic designs. >> >> We propose resolving the issue as 'invalid'. > >The underlying issue is that having an exclusion model without a >connected collision-handling model is broken, because it leads >authors to build designs that are extremely inflexible, and only >work at the specific page size for which they designed it. Exclusions can be used with a flexibly-sized grid layout, allowing one grid item to affect the inline content of other grid items. We have examples of this in the spec. A collision-handling model is not required to make this useful, and the layout can respond to page size changes. Exclusions can be used with a flexibly-sized flex layout, allowing one flex item to affect the inline content of other flex items. A collision-handling model is not required to make this useful, and the layout can respond to page size changes. Exclusions can be used with absolute positioning that is percentage-based (and thus not tied to a particular page size) where a collision is *intended* and made workable using the exclusions. Take a look at the ordering example in the spec. If the technical point to this issue is that exclusions only work with a specific page size, I believe the issue is invalid. If the technical point to this issue is that exclusions require a collision-handling model, I believe the issue is invalid. Thanks, Alan
Received on Monday, 30 July 2012 18:36:16 UTC