Re: Add 'manipulation' touch-action property?

Actually, sorry - I think the grammar isn't quite right.  You have: "auto |
none | [pan-x || pan-y || manipulation]" but I think it should be "auto |
none | [pan-x || pan-y ] | manipulation".  I.e. manipulation cannot be
combined with the pan-x/pan-y values.  At least that's my reading of the
(more complex) IE grammar:
http://msdn.microsoft.com/en-us/library/ie/hh772044(v=vs.85).aspx

Thanks,
   Rick


On Tue, Mar 4, 2014 at 9:31 PM, Rick Byers <rbyers@google.com> wrote:

> Looks great, thank you very much for the super-fast turn-around!  We're
> already beginning to implement this in chromium (http://crbug.com/349016).
>
> Rick
>
>
> On Tue, Mar 4, 2014 at 8:53 PM, Jacob Rossi <Jacob.Rossi@microsoft.com>wrote:
>
>> Hi folks,
>>
>> Per our discussion in today's meeting, I've added "manipulation" to the
>> latest editor's draft using the language proposed by Rick below. Let me
>> know if there are issues with the change.
>>
>> https://dvcs.w3.org/hg/pointerevents/rev/018f1b69c985
>>
>> -Jacob
>>
>> -----Original Message-----
>> From: Jacob Rossi [mailto:Jacob.Rossi@microsoft.com]
>> Sent: Wednesday, February 26, 2014 4:12 PM
>> To: Patrick H. Lauke; Rick Byers
>> Cc: public-pointer-events@w3.org
>> Subject: RE: Add 'manipulation' touch-action property?
>>
>> I have no problems with this addition iff we have strong consensus and
>> support by implementers. We have an opportunity for this type of change
>> given we intend to publish another LC draft soon and it would be helpful
>> for interop.
>>
>> But I wouldn’t block on the spec’s progress for this. So if we don’t
>> reach consensus very soon or if a 2nd interoperable implementation doesn’t
>> soon arise, then I’d rather move this to V2.
>>
>> I think Rick’s text is fine as-is for me.
>>
>> -Jacob
>>
>>
>> On 26 Feb 2014, at 01:33, Rick Byers <rbyers@google.com> wrote:
>> Hi,
>> I'd like to propose we add IE's 'touch-action: manipulation' [1] property
>> to the pointer events specification.  We've avoided the Microsoft-specific
>> properties because they're explicitly out of scope for the working group,
>> but I think we can word the definition of 'manipulation' such that it's
>> within the scope of the group and consistent with the existing language.
>>  In particular, how about something like:
>>
>> manipulation: The user agent MAY consider touches that begin on the
>> element only for the purposes of panning and continuous zooming.  Any
>> additional behaviors supported by 'auto' are out of scope for this
>> specification.
>>
>> I'm happy to go into why I think this is important, but my biggest
>> reasons out out of scope for this group.  Instead we can focus on:
>>  - improves compatibility with existing sites (manipulation seems to be
>> the most commonly used of the non-standard properties)
>>  - acts as a nice shorthand replacement for the more awkward
>> 'touch-action: pan-x pan-y'
>>
>> Alternately, I'd even be content to spec 'manipulation' as a synonym for
>> 'pan-x pan-y' but say that user agents may choose to associate with it
>> additional semantics that are out of scope with this specification.
>>
>> Thanks,
>>    Rick
>>
>> [1] http://msdn.microsoft.com/en-us/library/windows/apps/hh767313.aspx
>>
>
>