W3C home > Mailing lists > Public > www-style@w3.org > January 2015

Re: [css-inline] Request for advice: Initial letter behaviour

From: Florian Rivoal <florian@rivoal.net>
Date: Thu, 29 Jan 2015 14:50:00 +0100
Cc: "CJK discussion (public-i18n-cjk@w3.org)" <public-i18n-cjk@w3.org>, W3C Style <www-style@w3.org>, Richard Ishida <ishida@w3.org>
Message-Id: <C6C7E268-67F6-4189-B71F-1657CA2477D0@rivoal.net>
To: Koji Ishii <kojiishi@gmail.com>

> On 19 Jan 2015, at 12:45, Koji Ishii <kojiishi@gmail.com> wrote:
> 
> On Sun, Jan 18, 2015 at 10:45 PM, Florian Rivoal <florian@rivoal.net> wrote:
>> 
>> Mainly 3 things:
>> 
>> 1) It sounds reasonable to me that we should be able to put ruby on drop/raised caps. However, it if never happens, there is no need to specify how it should work and to implement it. So I am looking for evidence that we should support it at all.
> 
> Isn't it needed, if never happens, to specify how drop/raised caps
> disables ruby automatically?
> 
> It's easy to say rare, but hard to prove none. Since it's rare, if
> existing impl behave the same way, we could spec that. Happen to know
> how existing impl behave? Gecko might need to use Nightly and turn
> flags on.

initial-letter is new, and I don't think any browser implements it yet (at least in public builds)

>> 2) What is supposed to happen on jukugo with group ruby, where only the first kanji of the jukugo has initial-letter applied to it?
>> 
>> p::first-letter {initial-letter: 2;}
>> <p><ruby><rb>今日</rb><rt>きょう</rt></ruby>は天気がいいから、散歩しましょう。</p>
>> 
>> I can think of two alternatives: either initial letter doesn't apply at all, or it extends to the whole ruby base. Probably the way we should spec it is to make ::first-letter not select the first character of a multi-character ruby base, and require an explicit selector like "p rb {initial-letter: 2;}". But before pushing for that, I wanted to check examples of actual usage.
> 
> That's a good point I didn't notice. As said above, I'm happy to take
> whatever impls do if they're interoperable.

Since browsers don't support initial letter, there is no interop to go by.


> BTW, I noticed that initial letter needs to take TCY as an atomic
> character. Is this covered?

I agree it is the same problem. initial-letter should not apply to a single letter that is part of a TCY. Either all or none of it, and as above, I'd prefer if applying initial-letter via first-letter did nothing in this case, and we required applying it explicitly to the whole element.

As far as I can tell, no spec handles this currently.

>> 3) If we have ruby on a drop/raised cap, what font size should it use? Same as ruby on the rest of the text? Enlarged proportionally with the initial? Something else?
> 
> I don't think this needs to be taken care of. Ruby are usually less
> than 50% when base text is large even outside the initial letter, such
> use is often seen in comics and in headings, but CSS Ruby Level 1
> leave this up to UA[1]. I think the same applies to initial letter
> case.
> 
> [1] http://dev.w3.org/csswg/css-ruby/#ruby-size

I am not sure I follow. Even if we don't specify the exact size the ruby should have, I think it would be worth having the spec say whether the ruby on the initial letter should be the same size as the ruby over the base text, or if it should be sized based on the size of the initial letter. No?

 - Florian
Received on Thursday, 29 January 2015 13:50:24 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:51:57 UTC