W3C home > Mailing lists > Public > www-style@w3.org > April 2010

RE: Flexbox - box orientation

From: Alex Mogilevsky <alexmog@microsoft.com>
Date: Thu, 29 Apr 2010 12:23:29 +0000
To: Tab Atkins Jr. <jackalmage@gmail.com>, www-style list <www-style@w3.org>
Message-ID: <B738CAD75DE3C64FAF5F477519FEACA005F007@TK5EX14MBXC141.redmond.corp.microsoft.com>
I agree with both points here
1) the spec should clearly say how 'direction' and 'block-progression' affect flexbox direction, orientation and alignment. It may be intuitive but deserves to be defined clearly.
2) it really doesn't make sense for "reverse" to affect both dimensions. If it is desired to have multi-line box to stack lines bottom to top, it probably needs a separate property ('box-line-progression:normal|reverse' ?)

-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf Of Tab Atkins Jr.
Sent: Friday, April 16, 2010 5:49 PM
To: www-style list
Subject: Flexbox - box orientation

I'm reading over the current flexbox spec very carefully while writing my revision, and I notice some oddities in how the before/after/start/end directions are determined.

First, for a standard horizontal box, start is left and before is top.
 For a standard vertical, start is top and before is left.  This is all reasonable and fine.  Now, for reverse boxes, *both* directions reverse - for a horizontal box start is right and before is bottom.
Is this useful?  Naively, I'd assume that you'd only want to reverse the left/right on a horizontal box, and the top/bottom on a vertical box.  It seems to me that the flow direction of the axis perpendicular to the orientation is a completely separate question to the flow direction of the axis parallel to the orientation.  (I'd like for there to be a natural mapping between the two, as it would allow a somewhat simpler model simmering in my head, but I'm not sure if such a mapping would be natural/useful.)

As well, in the very last paragraph of the Alignment chapter, the topic of ltr versus rtl comes out of nowhere as affecting the direction of start/end/before/after edges.  This is not mentioned anywhere else in the draft, so you could be forgiven for assuming that text direction plays no part in the determination of the various directions.

What, precisely, are the most useful direction associations here?
What does XUL use?  Do people like what XUL uses?  Is it useful to pay attention to text direction?


Received on Thursday, 29 April 2010 12:24:07 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:07:45 UTC