- From: fantasai <fantasai.lists@inkedblade.net>
- Date: Tue, 22 May 2012 12:00:56 -0700
- To: www-style@w3.org
On 05/21/2012 06:29 PM, Tab Atkins Jr. wrote: > (I'm hoping that this doesn't start a new thread. For some reason > Ojan's message hasn't my inbox yet, an hour after it showed up in the > list archives.) > > On May 21 2012, Ojan said: >> I think this was probably an oversight in the recent changes to the spec, >> but at some point, the default value of flex-basis changed from 0px to auto. >> >> "‘flex:<positive-number>’ >> Equivalent to ‘flex:<positive-number> 1 0px’. This value makes the flex >> item flexible, and sets the flex basisto zero, resulting in an item that >> receives the specified proportion of the free space in the flex container. >> If all items in the flex container use this pattern, their sizes will be >> proportional to the specified flex ratio." >> >> That seems like the correct default behavior. Having auto as the preferred >> size is considerably slower and often not what the developer wants. It >> should not be the default value. >> >> "flex:auto == flex:1 1 auto" and "flex:none == flex: 0 0 auto" both seem >> fine to me as is, but the default value for flex should be "1 1 0px". > > Before we added flex-grow/shrink/basis, the initial value for 'flex' > was "none", or "0 0 auto". I suspect that's why gave flex-basis > 'auto' as its initial value. > > Now, though, everything flexes by default. I have no real opinion on > whether "absolute" or "relative" flex is the better default behavior. > > Anyone else have strong opinions either way? Yes, I think the default behavior should be 'auto'. If you're not specifying proportions explicitly and just want things to auto-fit, this is the right behavior. Also ignoring width/height by default seems like a really bad call. In other words, 'flex: auto' is a better default than 'flex: 1'. If you're individually tweaking 'flex-grow' and not getting expected results, it's your fault for not using the shorthand like you're supposed to! ~fantasai
Received on Tuesday, 22 May 2012 19:01:51 UTC