Re: Constraints 2014 new slides

See my main comment to Dominique,  but I wanted to answer your remaining 
questions.

On 3/27/14 12:09 PM, Eric Rescorla wrote:
> I'm assuming that this rather than your mailing list post is now the
> most up to date version of your proposal?

Yes, though the slides/new slides difference is cosmetic. My 
presentation will be the same, except the examples have "video: { }" 
around them.

> These syntaxes appear to be incompatible in the sense that if you were 
> to supply a new-style syntax to an old implementation, it would likely 
> choke and vice versa.

Unknown WebIDL dictionary members MUST be ignored by old browsers, so 
this should be 100% compatible, in theory.

Unfortunately, Chrome has a bug because it throws a " Malformed 
constraints object.", which is in clear violation of WebIDL and should 
be fixed ASAP!

This may delay a transition, but doesn't make it impossible. Bugs happen.

> The actual algorithm for processing optional constraints seems
> pretty complicated. At the point where this is an unordered wish
> list, why not just let the browser evaluate the preferences any
> way it pleases?

I agree! That's why I softened the language on default behavior in the 
slides. In other words, when "prefer" is not specified, I see no reason 
to restrain how the UA implements, other than perhaps to say that it 
SHOULD weigh multiple matches more than fewer matches.

> In summary, I oppose adopting this proposal at this time. I'm not a big
> fan of the current syntax, but if we're going to change a bunch of stuff
> we should take a serious look at what we are optimizing for and
> then build something that matches that.

If you think you can get bigger changes through, I welcome you to try. I 
feel I tried that already, and it didn't work for me, so I'm aiming for 
less this time.

.: Jan-Ivar :.