[whatwg] Support for page transitions

On Tue, Jun 7, 2011 at 7:45 AM, Mikko Rantalainen
<mikko.rantalainen at peda.net> wrote:
> The things I don't want to have in this specification (page author control):
>
> - actual transition animation ("slide the next page from the left")
> - transition duration
> - ability to specify easing for transition movement
>
> Instead there should be a method for defining that submitting a form
> with a given button, the UA should use transition to "next page".
> Hitting another button on the same form should use transition to
> "previous page" and hitting some link should use "closing" transition.

It would make sense for the author to be able to control this too.
You can already do in-page transitions using CSS, and the same syntax
could be reused for page transitions:

http://www.w3.org/TR/css3-transitions/

> Note that the "next page" button may or may not match with rel="next"
> and as such, I think that there should be additional method for
> specifying this kind of relation.

What are cases where it wouldn't match?

> I think that it would make sense to use "next page" transition for
> rel="next" by default, but there's a need to attach "next page"
> transition to interactive elements other than rel="next".

What need?

> I think that this could be sensible to have in HTML instead of just in
> the CSS (or some other method) because it's possible that other software
> but just the styling system could use the information about target type
> for links and buttons.

Offhand, it seems sensible to reuse rel; let each platform work out
the default transition animation for each link type (perhaps none in
most cases); and allow authors to override the transition animation on
a per-link basis.  Selectors like a[rel~=next] would be useful here
for authors.  On the desktop, you don't usually have this sort of
next-page animation, so it would be weird if pages exhibited that
behavior unless the author specifically requested it.  On Android or
other particular platforms, it might make sense as the default.  But
it definitely makes sense to me to put this in CSS.

Received on Tuesday, 7 June 2011 15:37:29 UTC