Re: Simplified <picture> element draft

My overall impression of this is very good, and I am also a strong
supporter of srcset. As to testing, the obvious answer is "a lot". In my
opinion, some kind of fallback is essential, particularly in transitional
stages. It seems to me that, whatever that specification, there will need
to be 2 stages in its inception - namely a transitional requiring fallback,
and a strict no longer requiring it once accepted as at least somewhat of a

On Monday, November 25, 2013, Kornel LesiƄski wrote:

> On 25 November 2013 15:16:03 Aaron Gustafson <>
> wrote:
>  Thanks for your hard work on this, I am really liking the direction. What
>> do you think browser vendor reception will be?
> I haven't got any official responses, but I haven't heard any strong
> objections yet either.
> Developers I've asked like it better than src-N and the old <picture>
> (although not necessarily srcset), with disclaimer that their opinion isn't
> official opinion of their employer.
> The issues discussed currently are details whether <img> in fallback
> content should be always required (as explicit "controlling image"), what
> to do with scripts creating the fallback img dynamically, and how much
> testing the new element will require.
> We still need to work out how to get viewport switching syntax of src-N in
> <source>, and if and how <img src> in fallback content should be used for
> <picture>.
> --
> regards, Kornel

*Daniel J Chamberlin*  _  831·854·7813 _


Received on Tuesday, 3 December 2013 09:17:41 UTC