[css3-writing-modes] leaving Tr fallback undefined

>From the minutes of the CSSWG Telcon on 16 Oct 2013 [1]:

> fantasai: On the topic of writing modes, glazou you asked what was
>           holding it up.
> fantasai: It's a minor detail so we can make it undefined.
> fantasai: The spec has it undefined and the argument is for making it
>           not undefined.
> glazou:   There's a disagreement on what to do and if everyone agrees on
>           leaving it undefined for now we can move to LC.
> glazou:   If they don't and it takes another month for a solution, we
>           can't for to LC.
> glazou:   So my question was can we leave it undefined.

I don't think there's really "undefined" behavior here, we're arguing
whether to allow optional fallback.  Optional behavior is kind of sucky
for authors but I think I'm more concerned that Koji feels that while
the Writing Modes spec makes it optional, in his view UTR50 *requires*
it [2]:

> 1. SHOULD|MUST render the fallback rotated sideways.
> 2. MUST render the fallback upright.
> 3. MAY render the fallback upright, or MAY render rotated sideways.

> UTR50 says #1:
> # Tr: Same as Tu except that, as a fallback, the
> # character can be displayed with the code chart
> # glyph rotated 90 degrees clockwise.

Nothing in the wording of UTR50 implies to me option (1), namely that
this fallback is *required*, and discussions with other Unicode UTC members
suggests that it's not a commonly held opinion.  But I think we need a
clarification from the UTC before moving forward.  The UTC is meeting
the week before TPAC which is one reason I think it makes sense to 
wait until then to try and resolve this.  In the meantime, it would be
useful to hear from other implementors whether they've implemented this
fallback and, if so, their reasons for implementing it.  It would be 
extremely helpful to hear from Microsoft, Apple and Amazon about this.

I'm not comfortable with leaving this optional if the combination of
Writing Modes plus UTR50 effectively makes this fallback required for
all user agents.  I think we need to resolve this before publishing LC.

> SteveZ: There's still information to be gathered, does the working
>    group have consensus that we need to get LC out since we need
>    comments, recognizing no decision on this issue?
> <sgalineau> also, LC will encourage i18n and others to review this
>     issue soon anyway
> fantasai: If we don't publish tomorrow, we can't have a period that
>   ends before or during TPAC.
> SteveZ: They don't wait to end to comment.
> ChrisL: Yes they do.
> SteveZ: So we're already too late :)
> 
> fantasai: I think we did this with an issue with fonts where we left
>   it open for LC.
> fantasai: This is a similar level of detail.
> <fantasai> vertical synthesized italics

No, we resolved all issues before publishing the Fonts LC.  Daniel
felt we should resolve contentious italics issue before LC.  In the
end we resolved to leave it undefined.

I should note here that the Internationalization group (the group most
likely to have comments on this spec) just completed a review of the
Writing Modes spec in August.  We can ask them to review it again
before TPAC with the explanation that the Tr fallback issue is still
unresolved.

Regards,

John Daggett

[1] http://lists.w3.org/Archives/Public/www-style/2013Oct/0460.html
[2] http://lists.w3.org/Archives/Public/www-style/2013Oct/0366.html

Received on Thursday, 17 October 2013 03:02:20 UTC