Re: [CSS3] General question about CSS3 vendor prefixes

Out of curiosity, what was the particular feature(s) that caused the
problem?

On Thu, Apr 28, 2011 at 1:56 PM, Boris Zbarsky <bzbarsky@mit.edu> wrote:

> We've had that sort of thing before, in the last 90s, and no one (at least
> of the UA vendors I know) really wants to go back to that.
>

Thank you for explaining this.

On Thu, Apr 28, 2011 at 3:46 PM, Brian Manthos <brianman@microsoft.com>wrote:

>  Suppose you use a not-ready-for-CR feature without prefix and some
> “misbehaving” browser allows you to do so.  All “behaving” browsers will
> not.  Your site starts out as “seeming compliant” (“look ma, no prefixes in
> the markup”) even though it only works in that “seemingly cutting edge, but
> actually misbehaving” browser.  Now the spec changes and reaches CR.  This
> “now actually cutting edge” formerly misbehaving browser auto-updates to
> match the CR spec – unprefixed before and after the conversion.  Your site
> breaks for customers that auto-update that browser, and remains working for
> those that don’t.   The spec gets blamed for changing.  The browser gets
> blamed for “breaking” you.  Your site gets blamed for tormenting users.  The
> web as a whole gets a criticized for not having solved the core problem
> after over a decade.  Bad situation for all involved.
>