RE: [css3-flexbox] flex-shrink defines a negative flex ratio; negative numbers are invalid

± From: Sylvain Galineau [mailto:sylvaing@microsoft.com] 
± Sent: Thursday, May 31, 2012 11:28 AM
± 
± [Tab Atkins Jr.:]
± > 
± > On Wed, May 30, 2012 at 11:12 PM, Sylvain Galineau 
± > <sylvaing@microsoft.com>
± > wrote:
± > > Usability question: is it potentially odd or confusing for the 
± > > negative flex ratio to be a positive integer and to make negative 
± > > values invalid? [1]
± > >
± > > [1] http://dev.w3.org/csswg/css3-flexbox/#flex-shrink-property

± > 
± > No clue.  I haven't been confused by it so far, but that's not a 
± > strong indicator.
± > 
± 
± It doesn't bother me unduly either but on reading the prose I just 
± found it odd to be told a negative ratio must be a positive integer. 
± And given the encouragement to use the shorthand I wondered if making 
± flex-shrink a negative could help make the value more readable by 
± clarifying which flex value does what.

It would make "flex:1 -1 100px" look cool and we certainly can do that. It would however create the first ever precedent of a property taking negative values only. And its meaning is not really "negative", it is ability to shrink... 

Maybe we should just use a different word in the spec?

Received on Thursday, 31 May 2012 21:19:38 UTC