W3C home > Mailing lists > Public > www-style@w3.org > June 2011

Re: [css3-images] Summary of recent gradient issues

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Thu, 16 Jun 2011 16:12:56 -0700
Message-ID: <BANLkTikEwTR_sTOLF6u5pBGALahvtj9vSO+PDjSe1aEfYK_0Yg@mail.gmail.com>
To: Brad Kemper <brad.kemper@gmail.com>
Cc: fantasai <fantasai.lists@inkedblade.net>, "www-style@w3.org" <www-style@w3.org>
On Thu, Jun 16, 2011 at 3:54 PM, Brad Kemper <brad.kemper@gmail.com> wrote:
> On Jun 16, 2011, at 3:30 PM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote:
>> On Thu, Jun 16, 2011 at 3:17 PM, fantasai <fantasai.lists@inkedblade.net> wrote:
>>> Or rather,
>>>  Given linear-gradient(left, black, white), which side would you expect to
>>> be black?
>> Given my conclusion #1 (by themselves, keywords are intuitively
>> understood to be starting-points), I doubt this would give us a useful
>> answer.  I don't think the keywords are confusing by themselves, I
>> think they're confusing because they're inconsistent with the other
>> way to specify gradients.
> But the way you asked it implies a relationship between the keyword and angle that some of us do not assume. By framing the question that way, you akkready set up that relationship in the minds of the respondents. It can bias the answers.
> I would ask it the way fantasai asked, and then also ask, "Given linear-gradient(0deg, black, white), which side would you expect to be black?" if the same person answers both, separately then you can see if they Consider the second question linked to the first as strongly as you and Brian would.

Again, my objection is *not* to the value definitions on their own.  I
was obviously fine with each of them on their own, since I wrote them
that way.  It is precisely the two types of values in combination that
concern me.  Two locally-good decisions can be bad when combined due
to them being inconsistent with each other.

That's why I very specifically brought up the two in concert, to
explore how people intuitively relate the two.

> Also, I don't think twitter is going to give you a suitable sample size of responses. I'd like To see the question on survey monkey where each question is on it's own page, in that order, and then advertise the poll in more places (this list, CSS.info, several twitter accounts, ALA, blogs, etc).

I got 16 responses, which is a good return on the minimal investment I
was willing to put in.  If you'd like to post the question more
widely, feel free.  ^_^  (Though, if you post the question the way you
proposed, I don't think you'd get very useful information - we already
know the answer to what people consider "0deg" to be, and we know what
people consider "left" to be if they only think about that on its

Received on Thursday, 16 June 2011 23:13:51 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:46 UTC