Re: wading into the Prefix morass...

David Singer:
> On Feb 20, 2012, at 11:53 , Daniel Glazman wrote:
> 
>> if we do that, will Apple remove such early prefixes from WebKit even if they are shipped and used in the wild? Without a VERY firm "yes", I will increase the problem only, not decrease it…
> 
> I think if a vendor introduces an idea, that then transitions into a CSS WG item, they ought to recognizer their vendor prefix while it's theirs, and the CSS prefix once it goes to the W3C.  I don't think they should be required to then drop *support* for the vendor prefix, but of course, I agree, it would be most helpful if they then *evangelized* the CSS prefix and 'ceased to mention' their old vendor prefixed version.  Is that what you are asking?

It probably answers the question he was asking – in the way he presumed.

Vendor (prefix) lock-in comes from 

  idea → implementation → release → proposal → standard

which is a (simplified) scheme some players seems to favor, especially when they see early implementation (however imperfect) as an business advantage. The ideal open standard, W3C process, however, would be

  idea → proposal → implementation → standard → release
                      ↖ draft ↙

The resulting actual process, currently, is more like

     a)↗  implementation  ↘
  idea               ↑   release
     b)↘ proposal → draft ↙

Once you hit the “release” stage you* hardly can remove prefixes that exist at this point, whatever they may be called.

*read: anyone who wants to keep or increase their market share.