- From: Charles McCathieNevile <chaals@opera.com>
- Date: Wed, 20 Jun 2012 13:48:56 +0200
- To: "public-coremob@w3.org" <public-coremob@w3.org>
(Thus I can close ACTION-1 in trackbot - ahead of the due date :) ). Right now, there are various CSS properties people want to use but which are only available as prefixed implementations. Most of the stuff that today's cool kids want is -webkit-something (which says a lot about where the cool kids are looking), but the current methodology of the CSS working group means that having something prefixed is likely to be around for a while. If coremob is unsuccessful, it doesn't matter what we do. But is it has a real impact on the implementation priorities of the top handful of mobile browser manufacturers, and identifies things currently available only as prefixed properties (or prefixed APIs), there are a few things we could do (not necessarily mutually exclusively): 1. We recommend prefixed versions as applicable based on the functionality we want. This is a nasty solution, because it motivates vendors to support other people's prefixes. This has already happened in a few cases. It upsets developers, and it reduces the potential for the prefix system to do its job of allowing experimentation followed by interoperable implementation of the standards. But it is the rational pragmatic approach for vendors to take, so it is likely to recur if we follow this path. 2. Only recommend things when they are available unprefixed. We could say what we *want* to recommend, as a way to clarify people's minds, but wait until the CSS group has changed from prefixed to unprefixed versions. 3. We convince the CSS group to change their approach to prefixing. This is possible and has been discussed - we can certainly add weight to any proposal we think is a real improvement. 4. We convince developers and vendors to change their approaches to work strictly in line with the existing policy on prefixes. This is unlikely, as it requires significant changes to the way the market functions. It seems that we should avoid doing 1 because it has unpleasant results, and 4 because it is like asking people to stop wanting to be paid. IMHO we should work on a combination of doing 2, and 3 wherever there is a solution that seems likely to improve things. cheers Chaals -- Charles 'chaals' McCathieNevile Opera Software, Standards Group je parle français -- hablo español -- jeg kan noen norsk http://my.opera.com/chaals Try Opera: http://www.opera.com
Received on Wednesday, 20 June 2012 11:49:38 UTC