Re: [css3-flexbox] One final round of bikeshedding on property/value names?

On Wed, Apr 25, 2012 at 10:02 AM, Alex Mogilevsky <alexmog@microsoft.com> wrote:
> ± From: Tab Atkins Jr. [mailto:jackalmage@gmail.com]
> ± Sent: Thursday, April 19, 2012 11:27 AM
> ±
> ± On Tue, Apr 17, 2012 at 7:44 PM, Alex Mogilevsky <alexmog@microsoft.com>
> ± wrote:
> ± > ± From: Tab Atkins Jr. [mailto:jackalmage@gmail.com] ± Sent: Monday,
> ± > April 16, 2012 4:34 PM ± ± 'flex-align' becomes 'content-align'
> ± > ± 'flex-item-align' becomes 'box-align'
> ± > ± 'flex-line-pack' becomes 'content-pack'
> ± > ± 'flex-pack' becomes 'content-justify'
> ± >
> ± > I am not a fan of moving to generic properties, I can't say I am perfectly
> ± happy with current naming. 'flex-item-align' and 'flex-line-pack' aren't the
> ± best names I've seen.
> ± >
> ± > Do we have better ideas, or can we apply Fanatai's thinking within the
> ± "flex-" set?
> ±
> ± Let's see...
> ±
> ± flex-align => flex-group-align
> ± flex-item-align => flex-align
> ± flex-pack => flex-group-justify
> ± flex-line-pack => flex-group-pack
> ±
> ± Or, try it the other way around:
> ±
> ± flex-align => flex-align
> ± flex-item-align => flex-box-align
> ± flex-pack => flex-justify
> ± flex-line-pack => flex-pack
> ±
> ± I'm not sure if either of these are actually an improvement over what we
> ± currently have.  :/  I'm inclined to just keep the current names until we get
> ± the proper generic names.
>
> I think I'll be with majority if I say I am not perfectly happy with the names but I would be just fine living with current set.
>
> Just in case any tweaking happens to naming, my preferences are...
>
> 0) The most used properties should be shorter. If all four could be one word each (with "flex-" prefix of course) it would be kewl.
>
> 1) *flex-align* -- perfect as is.
>
> It can be questioned which axis of flexbox "align" should describe.
> If there is any analogy with 'text-align' or HTML 'align', these are applied on axis that is orthogonal to block flow direction, exactly as it is used on flexbox.
>
> 2) *flex-item-align* -- it makes the most sense to be "flex-???-align" (as in most proposed options). "item" looks unusual, but there is 'list-item' for a precedent. "child" would be confusing (who's child?). "box" doesn't say what it applies to either.
>
> Making 'flex-align' apply to flex item and having a longer name for flexbox-level align goes against my preference (0)
>
> 3) *flex-pack* -- can have a number of alternatives, about equally attractive:
>
>        -compact
>        -fill
>        -justify
>        -arrange
>        -adjust
>        -shift
>        -cluster
>
> None sound much better to me, but I am already used to "-pack"...
>
> 4) *flex-line-pack* could actually become one word (maybe one of the above synonyms). I don't mind it being long though (it should be rare), and I think it would sound better if it was 'flex-wrap-???", showing clearly that it only affects content that wraps.
>
> 'flex-wrap-pack' would sound reasonable and intuitive.
> 'flex-wrap-align' would actually sound reasonable too.
>
> "-pack" works because same values apply, with the same effect.
> "-align" could work because it is in the same direction as 'flex-align'.
>
> I could live with either.
>
> That's about as much as I would be interested in changing. Not much, but if an awesome name comes up, I want to hear))

I'm digging the names that fantasai used in her generic alignment
draft <http://fantasai.inkedblade.net/style/specs/css3-align/#overview>.

content-justify/align => align the contents of a box en masse (maps to
flex-pack and flex-line-pack respectively)
box-justify/align => align a box within its parent, in the primary or
secondary axis. (box-align maps to flex-item-align)
child-justify/align => set the defaults for box-align/justify
(child-align maps to flex-align)

All the names are sensical, and the rules for remembering which one
does what are simple.  They're all two-word as well.

I'm pretty convinced now that these generic properties address our
needs in Flexbox, and will also work well in Grid.

The only criticism I might offer is that, while the child/content
distinction makes sense, it's subtle.  I might prefer
default-justify/align instead, as that's *really* clear what it's
doing.

~TJ

Received on Thursday, 26 April 2012 00:32:33 UTC