W3C home > Mailing lists > Public > public-fx@w3.org > January to March 2016

Re: [css-animations][web-animations] steps() timing function sometimes unintuitive

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Tue, 8 Mar 2016 15:59:01 -0800
Message-ID: <CAAWBYDASX7VXu5gES5Z+jVP7Bt10tCcPnB3ME26KZaPJCiVGAQ@mail.gmail.com>
To: "L. David Baron" <dbaron@dbaron.org>
Cc: Brian Birtles <bbirtles@mozilla.com>, www-style list <www-style@w3.org>, Rachel Nabors <rachelnabors@gmail.com>, "public-fx@w3.org" <public-fx@w3.org>
On Tue, Mar 8, 2016 at 3:29 PM, L. David Baron <dbaron@dbaron.org> wrote:
> Another possibility, working from the idea that steps(N, start/end)
> is pretty broken and this is generally the desired way to do
> multiple steps (i.e., not step-start and step-end which are a single
> step) is that this function with equal steps could simply be:
>   steps(N)
> with no second argument at all.

That has backwards-compatibility concerns - currently, if you omit the
keyword it defaults to "end".

Hmm. If we did do this, it'd mean we'd slightly reinterpret the N
value - rather than indicating the number of segments, it would be the
number of *discrete jumps*.  They'd default to being equally spaced
across the duration, and you can provide a keyword to optionally
indicate that the first/last jump should instead occur at the
start/end of the duration.  (The start/end values aren't *terrible*,
they're just not what people actually want in common cases.)

If we think the back-compat isn't bad, tho, I do like this the best.
We'd then get to add a "step" keyword, too, which is a shorthand for
"steps(1)", and gives the default "non-animatable value" behavior.

~TJ
Received on Tuesday, 8 March 2016 23:59:49 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 8 March 2016 23:59:50 UTC