W3C home > Mailing lists > Public > www-style@w3.org > December 2016

Re: [css-logical-props] Is this spec ready to ship?

From: Peter Moulder <pjrm@mail.internode.on.net>
Date: Mon, 19 Dec 2016 22:49:39 +1100
To: Christian Biesinger <cbiesinger@google.com>, www-style list <www-style@w3.org>
Message-ID: <20161219114939.GA12984@mail.internode.on.net>
Apparently, it was already decided in Sydney f2f 2016-05-11 that existing
impls had already committed us to the existing behaviour of margin-*, and
that new properties could later be introduced for directions in terms of
parent or containing block.

The only remaining issues for the size,padding,border,margin properties
that I'm currently aware of are:

  - #691: whether it's computed or used 'direction', and hence whether
    the computed value of 'text-orientation' is relevant.

    (Marked as closed, but see recent comments on the bug.)

  - #800: meaning of percentages on logical props.

  - Relatedly, meaning of 'inherit' on logical props.

    (Raised in http://www.w3.org/mid/20141230144943.GA26766@crum.dbaron.org
    with supplementary comment in
    http://www.w3.org/mid/20141231055217.GA6969@crum.dbaron.org
    but I didn't find a github issue for it, and current css-logical-props
    doesn't mention inheritance.)

    For both percentages and 'inherit', the easiest thing to implement is
    that it's as if the resolved physical property had that cascaded value,
    and it seems likely that all existing implementations would have done
    it this way.  In the case of 'inherit', this answer also simplifies
    the question of what "all:inherit" means when the parent element has
    a different writing mode.

    (Though I'd guess that it's not all that costly for an implementation
    to keep track of whether the winning declaration came from a physical
    or logical property, if there were an authoring advantage to doing so.)

    Also somewhat relevant to the decision is the issue of whether
    inspection of a declaration for the 'all' property gives a result that
    mentions logical properties
    (http://www.w3.org/mid/54A0E862.6080903@mcc.id.au).

I'd suggest avoiding shipping offset-* unprefixed, as this still differs
between the spec and at least Firefox.  (I haven't tested other impls.)

pjrm.
Received on Monday, 19 December 2016 11:50:32 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:09:05 UTC