W3C home > Mailing lists > Public > www-style@w3.org > May 2011

[css3-exclusions] Feedback on exclusions document

From: David Hyatt <hyatt@apple.com>
Date: Tue, 10 May 2011 17:34:18 -0500
Message-id: <4186C29E-4F1B-4083-88D2-F9CD42106F7B@apple.com>
To: "www-style@w3.org CSS" <www-style@w3.org>
I am assuming wrap-shape-mode applies to block-level elements.  This would also allow it to be applicable to replaced elements like images.

Keywords like left/right/top-bottom are bothersome from a writing-modes perspective, since there's no way they would really be physical (you're really referring to line edges with left/right).

For all the subsequent wrap-* properties, I would say that they apply to "exclusions" rather than continuing to say "any block element," since I am assuming wrap-shape-mode is what makes you an exclusion.

For float exclusions, it feels weird to me that you have to give a direction when the float is already hugging one particular side.  I don't have a particularly great suggestion here... it just seems silly to me to have to say <img style="float:left; wrap-shape-mode: right"> when that's pretty much always what I'll want if I make an exclusion.  Maybe a keyword that just indicates to base it off the float placement would be useful.

The wrap-shape-image properties bug me in the sense that I would expect a simple floating <img> tag to be able to act as an exclusion and have content wrap around its particular image.  I think having to make a div instead is overly complicated, and it would defeat the browser's built-in user interface for working with images (e.g., Save Image As... / Open Image in New Window / Copy Image Address, etc.).  One possible idea for solving this issue would be to leverage wrap-shape:auto to not just use the content box in the case of images.

<img style="float:left; wrap-shape:auto"> could mean to use the image.

The Combining Exclusions section is really unclear.  I think you might need to get more rigorous regarding the rules, especially given that the border boxes of the blocks and the wrap shapes of the blocks can differ.

wrap-shape-order concerns me in that I still wish we could latch on to existing z-order concepts instead of having to invent yet another stack.

Received on Tuesday, 10 May 2011 22:35:03 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:45 UTC