Intent to Ship: Touch-action: pinch-zoom CSS property

74 views
Skip to first unread message

Dave Tapuska

unread,
Sep 27, 2016, 2:29:25 PM9/27/16
to blink-dev
dtap...@chromium.org https://github.com/w3c/pointerevents/pull/99

Yes the spec is a pull request. It is currently
But Microsoft's shipping implementation is a defacto standard here. See explanation here. Support the touch-action: pinch-zoom CSS property. The user agent MAY consider touches that begin on the element for the purposes of continuous zooming and immediately execute the default action instead of waiting for an event handler to not cancel it. Allow a page to consume single finger gestures; but permit the user-agent to handle multiple finger gestures. Firefox: Positive Edge: Shipped Safari: No public signals Web developers: Positive . Ongoing technical constraints
Chrome has a slight implementation difference between Edge and Chrome. In Chrome once a user begins an action is doesn't change until all the fingers are lifted from the display. Edge allows the touch-action (for pinch-zoom purposes) to be redetermined when the second finger goes down. Chrome chose this model as it simplifies the complexity in the code path. We are working with Microsoft to determine what the correct behavior should be. This is tracked by issue 649034 Yes http://crbug.com/626086 https://www.chromestatus.com/features/5670200775016448

Philip Jägenstedt

unread,
Sep 30, 2016, 7:24:14 PM9/30/16
to Dave Tapuska, blink-dev
Something missing after "Yes the spec is a pull request. It is currently"?

Rick Byers

unread,
Oct 1, 2016, 7:56:02 PM10/1/16
to Philip Jägenstedt, Dave Tapuska, blink-dev

FWIW, I think shipping this even if it never lands in the official PE spec is the right thing to do.  This is really unusual, so it's probably wort some elaboration.

1.  We have agreement from all PE implementors (Edge, Gecko, blink) that this API should be broadly supported, and that the semantics should match what IE/Edge has shipped for years.

2.  The delay in getting this into the PE spec is due to explicit limitations in the PEWG charter scope as requested by Microsoft.  We will get this into some spec eventually, it'll just take time to figure out how to do it properly because ... lawyers.

3.  It's somewhat urgent.  As we encourage use of passive touch listeners or pointer events without this, developers are disabling support for pinch-zoom when there's no good reason, which can be an accessibility and usability concern.  In some cases (eg. AMP documents other than those served by Google search) it is also blocking passive touch listener adoption, and could block our ability to ship our planned forced-passive intervention (details forthcoming, but it's a 50% improvement in our primary scroll metric).

I hope this helps.   Of course I'm recusing myself from voting on this intent, since I wrote the spec PR.

Rick

Dave Tapuska

unread,
Oct 3, 2016, 12:53:14 PM10/3/16
to Philip Jägenstedt, blink-dev
Err; yes it appears I sent it prior to my editing fully that portion of the text. Hopefully Rick's explanation clarifies it.

--
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscribe@chromium.org.

Philip Jägenstedt

unread,
Oct 4, 2016, 4:48:27 AM10/4/16
to Dave Tapuska, blink-dev
Given that Edge has already shipped this and Gecko is positive, LGTM1.

PRs on a spec can easily be forgotten over time. Maybe this is already a high-profile issue so that it's guaranteed to move forward, but if not can you add some kind of tracking, maybe a crbug, that the input team is responsible for poking at every once in a while?

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

Chris Harrelson

unread,
Oct 5, 2016, 7:41:09 PM10/5/16
to Philip Jägenstedt, Dave Tapuska, blink-dev
LGTM2

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscribe@chromium.org.

--
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscribe@chromium.org.

Jochen Eisinger

unread,
Oct 10, 2016, 7:10:43 AM10/10/16
to Chris Harrelson, Philip Jägenstedt, Dave Tapuska, blink-dev
lgtm3

LGTM2

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

Reply all
Reply to author
Forward
0 new messages