Re: SMIL is dead,.. long live the SMIL

On Tue, Mar 17, 2015 at 9:16 PM, Tab Atkins Jr. <jackalmage@gmail.com>
wrote:

> On Tue, Mar 17, 2015 at 5:35 PM, Charles Lamont
> <charles@gateho.gotadsl.co.uk> wrote:
> >>> The result is that I am totally confused. Is SMIL dead or not? Will the
> >>> SMIL animated SVG that I wrote, and which has featured in Wikipedia for
> >>> the last 6 years continue to work (where it does now) or not?
> >>> Is it intended that some time soon SMIL declarative animation (or
> >>> something 'backward compatible') will work in Internet Explorer or not?
> >>
> >> SMIL won't work in IE for the foreseeable future; I don't particularly
> >> expect them to change their position on this (but I could be
> >> surprised).  It will continue to work, as much as it does, in the
> >> other browsers; in particular, Chrome is dropping its native support
> >> and switching to browser-JS to run it instead.  This shouldn't come
> >> with much, if any, of a behavior change.
> >
> > Presumably, then, it would stop working in Chrome for someone who has
> > disabled Javascript?
>
> I have no idea.  You can test this by turning off JS and trying to use
> the <marquee> element; it's already implemented via browser JS in
> Chrome.


If this is implemented using blink-in-js, it will work even if you disable
JavaScript for the page.
The emulation code lives in a different sandbox from the main document.


> >> We (Chrome) aren't interested in any further changes or improvements
> >> to SMIL, though.  Further improvements to animations should be done by
> >> additions to Web Animations, and if we need declarative ways to access
> >> such improvements, we'd prefer it be done via additions to CSS
> >> Animations.
> >
> > I still don't understand what this actually means:
> >
> >
> http://www.w3.org/TR/web-animations/#relationship-to-other-specifications
> >
> > I don't know what you mean by "... if we need declarative ways ...".
> >
> > This page:
> >
> > https://en.wikipedia.org/wiki/SVG_animation
> >
> > makes the problem with the WG's approach very clear. Looking at the
> > examples, the SMIL example is clear and readable; the CSS example is
> > less so, for reasons vehemently expressed by DD & JM earlier in this
> > discussion; and the scripted example is even more verbose and cannot
> > even be demonstrated: "No example as uploads with ECMAScript are
> > barred". Scripting is not always possible. Declarative animation, where
> > it can be used, is usually simpler to write, read and maintain.
>
> Those examples are extremely terrible.  The CSS one repeats itself
> twice, due to prefixes; the -moz one is definitely unnecessary, and
> the -webkit one may or may not be (it works unprefixed on my machine,
> but that might be due to a flag I have turned on).  Once you remove
> those duplications, the example becomes similar in size to the SMIL
> one (less lines, but longer; total text length is roughly comparable).
> The scripting example is similarly terrible; it's being done with
> plain JS, not the Web Animations API.  Written properly, the JS
> example will be roughly the same size and complexity of the other two
> examples.
>
> ~TJ
>
>