W3C home > Mailing lists > Public > www-style@w3.org > May 2014

Re: [cssom-view] scope of the 'scroll-behavior' property

From: Simon Pieters <simonp@opera.com>
Date: Mon, 26 May 2014 12:51:35 +0200
To: "Robert O'Callahan" <robert@ocallahan.org>
Cc: "L. David Baron" <dbaron@dbaron.org>, www-style <www-style@w3.org>
Message-ID: <op.xggq39iwidj3kv@simons-mbp>
On Mon, 26 May 2014 11:50:50 +0200, Robert O'Callahan  
<robert@ocallahan.org> wrote:

>> It was the API's "auto" value that was under discussion. Currently the  
>> CSS
>> property does not have an "auto" value.
> Yes, I was referring to the CSS property. It seems to me having an 'auto'
> value for that property that falls back to UA/platform defaults would be
> better than making 'smooth' do that. In particular it seems that  
> currently
> "smooth" in the JS API overrides UA defaults but CSS  
> scroll-behavior:smooth
> does not override UA defaults. It would be better if they were  
> consistent.

No, the API "smooth" honors user preferences just like the CSS property.

In particular, the spec allows the UA to ignore the 'scroll-behavior'  
property, which makes both navigation scroll and API scrolls be instant.  
See http://dev.w3.org/csswg/cssom-view/#perform-a-scroll step 2.

Technically, the spec also allows the UA to follow the "smooth scroll"  
steps and let that be a zero-time no-animation scroll, since it's  

>>  But then I'm not sure what the use-cases are for the "initial" and
>>> "smooth"
>>> values of scroll-behavior.
>> "instant" is the initial value, and gives the same behavior as we have
>> today.
>> "smooth" allows the UA to do the scroll over a period of time in a
>> UA-defined manner for scrolls triggered from navigation and from the
>> scrolling APIs (unless the script specified "instant" or "smooth"
>> explicitly).
>> For use cases, please see http://www.w3.org/mid/op.
>> ww4mdlp2idj3kv@simons-macbook-pro.local
> Thanks. I guess that makes sense though it means scroll-behavior is a  
> very
> narrowly-useful property. I'm not sure it's worth having to be honest,
> assuming we add the JS API.

OK. I thought the CSS property was worth having, but if implementors  
disagree then there's not much point having it in the spec.

Simon Pieters
Opera Software
Received on Monday, 26 May 2014 10:52:08 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:51:27 UTC