Re: [css-ruby] Default font-size of ruby annotation with ruby-position: inter-character

On Mon, Jan 12, 2015 at 9:56 PM, Koji Ishii <kojiishi@gmail.com> wrote:

> On Mon, Jan 12, 2015 at 6:56 AM, Xidorn Quan <quanxunzhen@gmail.com>
> wrote:
>
>> On Sun, Jan 11, 2015 at 6:10 PM, Koji Ishii <kojiishi@gmail.com> wrote:
>>
>>> Clarified that it's UA dependent in Level 1[1]. How UAs render
>>> Bopomofo and how much UAs do so interoperably is very unclear to me,
>>> so I think this is the best for me as of this point. Please let me
>>> know if any.
>>>
>>> [1] http://dev.w3.org/csswg/css-ruby/#ruby-size
>>
>>
>> I guess you probably should use "font size" instead of simply "size"
>> here, because size may refer to the box size.
>>
>
> Done, can you review again?
>

It seems that the document on w3c's site hasn't been updated, but I see the
update in the github repo. You probably forgot to sync the two documents in
this commit:
https://github.com/w3c/csswg-drafts/commit/64bb2ea3c48a9455e2d21a38e197cfb02ddf4c11


>
> In addition, the spec says:
>>
>> > This level of the specification does not provide a mechanism to control
>> the size of ruby annotations.
>>
>> I think we do have mechanism for authors to control the font size of ruby
>> annotations, which is the font-size property.
>>
>
> Right, added "relative to ruby base." Does this resolve the concern?
>

Sorry, but I don't think so. Maybe something like, this level of the spec
dose not define any additional mechanism to automatically compute the font
size according to ruby properties or content of ruby annotations.


>
> It might probably be better to mention that the default style sheet is not
>> normative, and say that UAs are free to behave a different way for some of
>> the styles.
>>
>
>  It's already there; Appendix A says it's informative.
>

I meant that it might be better to mention in the Ruby Font Size section in
addition that font-size: 50% in the default style sheet is not normative,
and the font size of annotation is actually impl-dependent. But I feel that
impl-dependent behavior is evil for web standard. I'm not quite sure what
should happen here.

- Xidorn