W3C home > Mailing lists > Public > public-coremob@w3.org > October 2012

Re: Inclusion of overflowScrolling in coremob specification

From: Tobie Langel <tobie@fb.com>
Date: Wed, 10 Oct 2012 09:29:21 +0000
To: Robert Shilston <robert.shilston@ft.com>, "SULLIVAN, BRYAN L" <bs3131@att.com>
CC: Paul Bakaus <pbakaus@zynga.com>, "public-coremob@w3.org" <public-coremob@w3.org>
Message-ID: <F9981AFB970564408FEB7DFCF62D440843677906@SC-MBX01-4.TheFacebook.com>
On 10/10/12 10:39 AM, "Robert Shilston" <robert.shilston@ft.com> wrote:

>Bryan,
>
>I think that polyfills show need, and that spec writers should review
>them and see where the spec needs to evolve.  I'm not sure if that's the
>CoreMob spec, or other specs that CoreMob has referenced.  I was trying
>to argue that "I just don't want to see useless cruft added to CSS" needs
>to be balanced with ensuring that browsers offer a great experience for
>users, and that if developers need features, then they ought to be
>considered for addition and whilst the spec gets longer it shouldn't be
>considered cruft.
>
>Rob

To second Rob's point, the best way to go about building specs is to offer
low level capabilities (e.g. IndexedDB), see what libs get built on top
(e.g. an async version of localStorage) and standardize high level specs
around these OSS market leaders. Like that, you get the low-level plumbing
where needed, and the high level feature that fits the 80% use cases and
benefits from being implemented by the browser directly (perf, one common
API, etc.).

With regards to momentum scrolling, there are clear need expressed by the
developer community. There's a plethora of JavaScript libraries enabling
momentum scrolling and related functionalities (pull to refresh, etc.).
Yet, because the performance requirements aren't met, it's driving
developers to native solutions. As mentioned before[1], it is one of the
key issues that brought us (Facebook) to move to a more native solution
for our mobile apps.

How exactly we address the issue of scrolling capabilities and performance
is open to debate. Is it strictly a quality of implementation issue? Are
there pieces missing from the specs? If so should these be addresses
though CSS? HTML? Etc.

But what's certain is that if we do not address it shortly in a performant
and interoperable way, we're going to drive developers away from the Web
platform an to native technologies.

Remember, momentum scrolling is to scrolling what touch is to click.


--tobie

---
[1]: http://lists.w3.org/Archives/Public/public-coremob/2012Sep/0021.html
Received on Wednesday, 10 October 2012 09:29:53 UTC

This archive was generated by hypermail 2.3.1 : Friday, 19 April 2013 17:36:47 UTC