W3C home > Mailing lists > Public > www-style@w3.org > February 2016

Re: [css-round][css-position] positioning offset property naming

From: Brad Kemper <brad.kemper@gmail.com>
Date: Sun, 28 Feb 2016 11:15:09 -0800
Cc: www-style list <www-style@w3.org>
Message-Id: <DABAF32B-28AE-4CA0-B230-93C36910A7E8@gmail.com>
To: fantasai <fantasai.lists@inkedblade.net>

> On Feb 12, 2016, at 7:02 PM, fantasai <fantasai.lists@inkedblade.net> wrote:
> At the F2F someone (Brad?) mentioned that we should change the prefix
> for polar-center / polar-distance to something else. I'm not convinced
> this is necessary, though I'm open to suggestions.

Yes, it was me. I think asking an author to mentally switch gears into thinking of polar coordinates instead of cartesian coordinates is an unnecessary mental hoop to have to authors to jump through, with no benefit to them. 

My view is that most authors are going to look at this as for what it can do. And all it really does (other than abspos centering an element on a [cartesian] point of the parent, which should be a separate operation, since it has independent utility on its own), all it REALLY does is apply a translation along an angled path. Given that with relative positioning there would be no centering step anyway, authors will see the central effect and purpose of the *-distance and *-angle properties as that they can move an element a certain distance along a certain angle. That does not require rethinking the cartesian coordinates used throughout CSS. 

Cartesian coordinates are even used for polar-origin, and remain as cartesian coordinates for other properties, such as translate, even when polar-* properties are in use. They were the primary reason given for making 0deg point up for linear gradient, instead of to the right as they would with a common protractor used in elementary math classes. I don’t see any strong argument for asking authors to think in different terms whenever they want to position something using an angle instead of an x y transation (as ‘left’ and ‘top’ would have given them for the same ending position).

So, some ideas for the property name:

anglular-distance, anglular-angle (or shorthand ‘angluar: <distance> | <angle>’)

anglular-move-distance, anglular-move-angle (or shorthand ‘angluar: <distance> | <angle>’)

move-distance, move-angle (or shorthand ‘move: <distance> | <angle>’)

radial-distance, radial-angle (or shorthand ‘ radial: <distance> | <angle>’)

angle-position: <distance> | <angle>

angle-offset: <distance> | <angle>

center: <distance> | <angle> (since distance percentages are based on the distance from the center of the element to the nearest edge of the containing block).

> However, I did want
> to point out that I'm planning to use 'offset-' as a prefix for the
> logical equivalents of top/bottom/left/right.

So maybe ‘offset-center’? ‘offset-radial’? ‘offset-angle’ and 'offset-distance’?

> This will also give us an 'offset' shorthand that resets them all,
> which is one of the annoying things about top/bottom/left/right atm--
> it's super annoying to set them all to e.g. 0 atm.

Sounds good.

> ~fantasai
> who really needs to finish the logical properties draft...
> among other things
Received on Sunday, 28 February 2016 19:15:45 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:09:00 UTC