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

Re: Public releases prefixes off-by-default (was RE: Unprefixing CSS properties)

From: Brad Kemper <brad.kemper@gmail.com>
Date: Thu, 17 Nov 2011 15:50:59 -0800
Message-Id: <DE6776F0-A87E-499A-852F-9A93E13AF9DD@gmail.com>
Cc: "Tab Atkins Jr." <jackalmage@gmail.com>, "robert@ocallahan.org" <robert@ocallahan.org>, www-style <www-style@w3.org>
To: Brian Manthos <brianman@microsoft.com>
On Nov 17, 2011, at 3:10 PM, Brian Manthos <brianman@microsoft.com> wrote:
>> -----Original Message-----
>> From: Tab Atkins Jr. [mailto:jackalmage@gmail.com] 
>> Sent: Thursday, November 17, 2011 3:09 PM
>> To: Brad Kemper
>> Cc: Brian Manthos; robert@ocallahan.org; www-style
>> Subject: Re: Public releases prefixes off-by-default (was RE: Unprefixing CSS properties)
>> On Thu, Nov 17, 2011 at 3:02 PM, Brad Kemper <brad.kemper@gmail.com> wrote:
>>> Maybe for things you haven't already provided as prefixed in general builds. Once you give something to authors and they find it very useful, I think you get anger in response to taking it away.
>> Oh gosh, yes.  I wouldn't suggest *taking* away stuff that's already
>> out there.  That's far too extreme.
> Um....
> So that proposal might fix new stuff but then is completely toothless for the problem at hand.
> Partial solutions are better than nothing I guess...

[moved Brian's quote down, to make it easier to respond to in a readable way]

I am also warming up to the idea of a '-draft-' prefix that could be overridden by UA-specific prefixes, such as '-webkit-'. Most of the time, a newer version of a draft would not change meaning without changing syntax too, but if it did (as degrees did in linear-gradient), then the WG could elect to reccomend a '-draft2-' prefix for the item in question. This wouldn't help that much with existing properties and values, since authors would still need to include prefixes for older versions of browsers, but I think it is worth reconsidering when moving forward. When the draft reached CR, authors could then have an unbuggy clean version to work with unprefixed. 

This idea suits me better than only having experimental version available in beta versions, where implementation experience and author experience becomes much more limited. 
Received on Thursday, 17 November 2011 23:51:41 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:52 UTC