Re: vendor keyword history

On Mon, Dec 15, 2008 at 11:20 AM, Steven Simpson <ss@comp.lancs.ac.uk> wrote:
> A case for URI-based namespaces?:
>
> @namespace somevendor "http://example.com/ns/";
>
> body {
>  -somevendor-myproperty: 23%;
> }

In practice, vendors seem to use descriptive enough prefixes that no
one is likely to collide anytime soon, at least with any vendor with
reasonably large market share.  Even if there is a collision, it will
only have an effect if the vendors happen to introduce exactly the
same vendor-prefixed property.  Even then it will only have a
*negative* effect if the properties have significantly different
semantics -- in many cases they don't, when the prefix only indicates
a property that's not yet standardized rather than one the vendor has
made up entirely.

It seems like no change is needed here unless there's some
demonstrated problem.  Other standards use even laxer restrictions on
vendor-specific identifiers, like saying that all vendors use a single
shared prefix such as X-, and I haven't heard of many problems arising
in practice.

Received on Monday, 15 December 2008 17:15:22 UTC