- From: Anton Prowse <prowse@moonhenge.net>
- Date: Fri, 06 Jun 2008 00:21:48 +0200
- To: www-style@w3.org
Anton Prowse wrote: > There are (at least) eight ambiguities or errors in description in 9.9.1 of stacking contexts, stacking levels and z-index > http://dev.moonhenge.net/css21/spec/z-index/ I have added to this collection the following issue, which has urgency in that the behaviour of the first beta release of IE8 differs from all other major browsers and yet cannot be said to be violating the current specification. Within a given stacking context, 9.9.1 mandates separate "stacking levels" (painting layers) for positioned elements (three layers, corresponding to negative/positive/auto-0 z-index) and floats (one layer). However, it fails to specify whether relatively positioned floats fall into the painting layer for floats or into the appropriate painting layer for relatively positioned elements. IE8b1 decides that floats belong in the float layer, irrespective of position or z-index. Everybody else (including IE<=7) allows position to win out over float. The latter approach is of course preferable because IE8b1's approach renders the z-index property impotent for relatively positioned floats. It would be desirable to resolve this issue or discuss it with Microsoft before IE8 is finalized. The corresponding bug report for IE8 is https://connect.microsoft.com/IE/feedback/ViewFeedback.aspx?FeedbackID=348959 (registration required). Thanks, Anton Prowse http://dev.moonhenge.net
Received on Thursday, 5 June 2008 22:22:01 UTC