Re: Alex Russell on Polyfills and some other thoughts...

On Saturday, 17 November 2012 at 16:50, Brian Kardell wrote:

>  
> On Nov 17, 2012 11:45 AM, "Marcos Caceres" <marcosscaceres@gmail.com (mailto:marcosscaceres@gmail.com)> wrote:
> >  
> >  
> >  
> >  
> > On Saturday, 17 November 2012 at 15:34, François REMY wrote:
> >  
> > > | I prefer the proposed api go into a prefixed or data- or x-
> > > | form so that it is parallel with the proposal, so if it evolves
> > > | to the point where it becomes native there is no risk of the
> > > | unexpected.
> > >  
> > > That's what I was afraid when I spoke about implementing faithfully an
> > > ever-changing spec. Prefixing is probably a key solution to avoid problems,
> > > but we can also burry prefixing inside the code generator.
> >  
> >  
> > Having said that, we've seen that prefixing sometimes makes no difference with regards to defacto standards (e.g., CSS prefixes being adopted across browsers, and browsers having to support those prefixes indefinitely to remain compatible with content that is no longer being maintained on the Web).  
> Right, those are problems with browser vendor prefixing. This suffers from none of those problems, because it is decoupled from browser implementations.
> Also, I think the IETF has also stopped recommending people use x-* in HTTP headers for the same reason. I guess we need to consider the options very carefully here.  
> Those arguments have been had and won, x- is the extensibility mechanism for html/css (easy enough to do the same in js prollyfills too) - its been had on the lists. IETF HTTP headers are entirely unrelated.

I'm ok with "x-", just wanted to make sure we didn't miss anything. I'm not too experienced with that particular form of extensibility, so if everyone is confident that is the way to go, then I'm happy to go along with it.

Received on Saturday, 17 November 2012 16:55:54 UTC