[css3-flexbox] comments on "Resolve flexible lengths" from Tab Atkins Jr. on 2012-09-07 (www-style@w3.org from September 2012) (original) (raw)

On Sat, Aug 11, 2012 at 3:52 PM, L. David Baron <dbaron@dbaron.org> wrote:

Some comments on http://dev.w3.org/csswg/css3-flexbox/#resolve-flexible-lengths :

In step 5, "the sign choosing the flex ratio" is very confusing. I think it should explicitly point to step 1. (It took me quite a while to understand that that's what it meant; I initially assumed it was extraneous wording that duplicated the distinction in the bullet points following.)

We didn't add an explicit link, but instead revised the wording so that it's more explicit anyway.

I don't like the terminology "flex grow ratio" and "flex shrink ratio", because they're not ratios -- they're weights or portions, or in the case of flex-shrink, a number to be multiplied by a width in order to get a weight or a portion. I think it makes sense to call the number computed during step 5 a "ratio", but I don't think the numbers it's computed from should be called ratios.

We did s/ratio/factor/.

I also think it would be better if what the flex-shrink and flex-grow properties were described a little bit under the definitions of those properties -- at least to the point of explaining that flex-shrink is multiplied by width and flex-grow is not, rather than only in the algorithm where authors reading the spec are going to have more difficulty finding it. (It also might make more sense to have that definition along with the individual properties rather than under the shorthand -- although in this case, given that the shorthand is the preferred form, maybe it makes sense as it is with the prose under the shorthand.)

Yeah, we've just collected the explanations under the 'flex' shorthand, but they link up.

We've put in a line about how the flex shrink factor is multiplied by the flex basis.

~TJ

Received on Friday, 7 September 2012 01:19:09 UTC