Re: Pull Down to Refresh

I'm fairly close to having an implementation working in tip of tree Chrome.

I'm still working on exposing the exact overscroll amount
<https://docs.google.com/document/d/1-ZH0Jfqdj_q4vM67587xP39pVwelPL4DPmdROZ8ASAw/edit?usp=sharing>
to
Chrome, and tweaking the way we handle the disposition of the first touch
move
<http://lists.w3.org/Archives/Public/public-touchevents/2014Apr/0005.html>.

The demo is available here <http://tdresser.github.io/polymer-ui-p2r/>, but
it requires recent Chrome and does have some problems which will be
addressed by the fixes mentioned above.


On Wed, Jun 25, 2014 at 4:01 PM, Timothy Dresser <tdresser@google.com>
wrote:

> I'm fairly close to having an implementation working in tip of tree Chrome.
>
> I'm still working on exposing the exact overscroll amount
> <https://docs.google.com/document/d/1-ZH0Jfqdj_q4vM67587xP39pVwelPL4DPmdROZ8ASAw/edit?usp=sharing> to
> Chrome, and tweaking the way we handle the disposition of the first touch
> move
> <http://lists.w3.org/Archives/Public/public-touchevents/2014Apr/0005.html>.
>
>
> The demo is available here <http://tdresser.github.io/polymer-ui-p2r/>,
> but it requires recent Chrome and does have some problems which will be
> addressed by the fixes mentioned above.
>
>
> On Wed, Jun 25, 2014 at 11:51 AM, Rick Byers <rbyers@google.com> wrote:
>
>> We agree getting pull-to-refresh working well on the web is absolutely
>> critical, it's one of the top priorities for the chrome input team.
>>
>> I did some original brainstorming here
>> <https://docs.google.com/a/chromium.org/document/d/1cBYKZMCeeaT6pN3VYu9uox8AXbJKmOaEmpL4BTg5lT4/edit#heading=h.n1849rs1xni3>,
>> and we've started to execute on that by shipping our 'async touchmove on
>> scroll' model of touch events in Chrome 36 (http://crbug.com/346693).
>>
>> tdresser@ has been working on this recently and has some demos.  They're
>> not perfect yet, but we're getting closer.  Tim, can you share your current
>> status and links?
>>
>> Pull to refresh was indeed a topic at the input face-to-face (notes
>> coming in the next day or so), but we didn't dive into specific solutions
>> in much detail there.  Instead we agreed it's a problem that we should be
>> working to solve on multiple angles.  On Chromium we're trying to avoid the
>> temptation of providing a specific feature and instead focusing on
>> improving the platform primitives to enable a wide class of features like
>> this to be built, but this stance is controversial and could ultimately
>> fail.  The IE team (Jacob cc'd) has been working on their own approach that
>> better leverages threaded scrolling.  We're continuing to discuss it with
>> all the browser vendors and will hopefully converge on a set of solutions
>> over the coming months.  But in the meantime we're letting the scenario
>> drive concrete changes to Chrome.
>>
>> Rick
>>
>>
>> On Wed, Jun 25, 2014 at 4:10 AM, Christophe Jolif <
>> christophe.jolif@fr.ibm.com> wrote:
>>
>>> Hi all,
>>>
>>> Out of curiosity, is that something that might have been discussed at
>>> the Web input brainstorming face to face? This is really a challenge for
>>> both touch and pointer and solving this would really help out application
>>> developpers.
>>>
>>> Christophe
>>>
>>>
>>>
>>> From:        Sebastien Pereira2/France/IBM@IBMFR
>>> To:        "public-touchevents@w3..org" <public-touchevents@w3.org>,
>>> Date:        04/14/2014 03:01 PM
>>> Subject:        Pull Down to Refresh
>>> ------------------------------
>>>
>>>
>>>
>>> So far it is a challenge to implement a cross browser PD2R feature. With
>>> touch events it is still possible by re-implementing the scrolling in
>>> JavaScript, which leads to heavy/complicated code and does not perform well
>>> in particular when scrollable data is large and complex. With pointer
>>> events and IE there is a solution based on snap points and chained
>>> scrolling [1] to provide PD2R using native scroll.
>>>
>>> The problem with pointer events on other browsers will be that when
>>> touch-action gives control to the user agent, there will be no way for the
>>> application developer to implement PD2R other than re-implementing the
>>> scrolling in JavaScriptt, using a touch-action value that prevents native
>>> scrolling.
>>>
>>> >From a user experience standpoint the pull down to refresh (or pull up
>>> to whatever) is becoming a de facto standard. It would be great if we could
>>> come up with a common solution to allow applications to use pointer events
>>> to implement PD2R with the benefit of the native scrolling.
>>>
>>> For example, we could imagine that when a user reaches the top (or the
>>> bottom in case of a Pull Up 2 something) :
>>> - a "pointerpull" event would be sent,
>>> - the user agent gives control back to the application and resume the
>>> sequence pointermove* > pointerup > pointerout
>>>
>>> [1] *http://dwcares.com/pull-to-refresh-2*
>>> <http://dwcares..com/pull-to-refresh-2>
>>>
>>> Sebastien Pereira
>>> IBM JS Team
>>> *https://github.com/ibm-js* <https://github.com/ibm-js>
>>> ---
>>> 33-1-5875-3479  (803479)
>>> Sauf indication contraire ci-dessus:/ Unless stated otherwise above:
>>> Compagnie IBM France
>>> Siège Social : 17 avenue de l'Europe, 92275 Bois-Colombes Cedex
>>> RCS Nanterre 552 118 465
>>> Forme Sociale : S.A.S.
>>> Capital Social : 655.732.332,20 €
>>> SIREN/SIRET : 552 118 465 03644 - Code NAF 6202A
>>>
>>> Sauf indication contraire ci-dessus:/ Unless stated otherwise above:
>>> Compagnie IBM France
>>> Siège Social : 17 avenue de l'Europe, 92275 Bois-Colombes Cedex
>>> RCS Nanterre 552 118 465
>>> Forme Sociale : S.A.S.
>>> Capital Social : 655.732.332,20 €
>>> SIREN/SIRET : 552 118 465 03644 - Code NAF 6202A
>>>
>>
>>
>