Re: Ruby annotation from Glenn Adams on 2009-07-02 (public-tt@w3.org from July 2009)

From: Glenn Adams <gadams@xfsi.com>
Date: Thu, 2 Jul 2009 14:58:00 +0800
To: Gur Shlapobersky <Gur@captionsinc.com>
Cc: Sean Hayes <Sean.Hayes@microsoft.com>, public-tt@w3.org
Message-ID: <94ad087a0907012358r3ac23a2uf28443b4d0d6c6d6@mail.gmail.com>
keep in mind that window (i.e., region) sizes cannot change shape in DFXP
except under authorial control, e.g., via animation; the same applies
regarding changing from horizontal to vertical, this can only be done under
authorial control; given this circumstance, the author that is using ruby
would alter the position of the ruby region as well;
in any case, I agree that what Sean proposes (and I have also proposed the
same previously in the absence of direct support for ruby), is not the best
solution, but it is *a* solution that can be used with what is presently
defined;

regards, glenn

On Thu, Jul 2, 2009 at 2:12 PM, Gur Shlapobersky <Gur@captionsinc.com>wrote:

>  I did consider this option but using visual design to create the rubies
> is not a perfect solution.
>
>
>
> Rubies can move around when the text layout changes. For example, consider
> a one line of text that is presented in a window and the last word has a
> ruby above it. If the window changes shape and needs to be re-rendered, but
> this time wrapping the text into a two line subtitle, the ruby will need to
> go *under* the last word of the second line. Other rules apply when
> changing horizontal text to vertical.
>
>
>
> Applying the ruby positioning logic by inferring a ruby is a ruby by
> font-size and position is not possible.
>
>
>
> Including both is possible like you have shown, although we will need to
> add in the <ruby> span a reference to the <p> element that it should
> substitute.
>
>
>
> We will wait to the next version.
>
>
>
>
>
> Gur S.
>
>
>
> *From:* Sean Hayes [mailto:Sean.Hayes@microsoft.com]
> *Sent:* Wednesday, July 01, 2009 16:09
> *To:* Gur Shlapobersky; Glenn Adams
>
> *Cc:* public-tt@w3.org
> *Subject:* RE: Ruby annotation
>
>
>
> The TTWG recognises the importance of Ruby as well as
> horizontal-in-vertical text and other formatting issues for Asian scripts.
> We did debate adding direct support for Ruby on a number of occasions, but
> in the end we took the view that the visual effect of Ruby could be
> simulated using the existing features of Timed Text, and the semantics could
> be incorporated using a foreign namespace, thus we decided to not add Ruby
> markup in a timed text namespace.
>
>
>
> As Glenn points out, one can include W3C ruby markup in a foreign
> namespace, but a standard dfxp renderer would not do anything with it. To
> achieve the semantic notation, and also have the rendering, one could write
> something like the following:
>
>
>
> <?xml version="1.0" encoding="utf-8" ?>
>
> <tt xml:lang="en" xmlns="http://www.w3.org/2006/10/ttaf1" xmlns:tts="
> http://www.w3.org/2006/10/ttaf1#style" xmlns:ttm="
> http://www.w3.org/2006/10/ttaf1#metadata">
>
>   <head>
>
>     <metadata>
>
>       <ttm:title>Ruby Test</ttm:title>
>
>       <ttm:desc>Example of how to apply ruby</ttm:desc>
>
>       <ttm:copyright>Copyright (C) 2007 W3C (MIT, ERIM, Keio). All Rights
> Reserved.</ttm:copyright>
>
>     </metadata>
>
>     <styling>
>
>       <style xml:id="base" tts:color="yellow" tts:fontSize="14px"
> tts:fontFamily="MS Gothic" tts:textAlign="center" />
>
>       <style xml:id="textStyle" style="base" tts:fontSize="32px" />
>
>       <style xml:id="rubyStyle" style="base" tts:fontSize="18px" />
>
>     </styling>
>
>     <layout>
>
>       <region xml:id="r1">
>
>         <style tts:origin="0px 30px" />
>
>         <style tts:extent="440px 32px" />
>
>         <style tts:zIndex="1" />
>
>       </region>
>
>       <region xml:id="r2">
>
>         <style tts:origin="80px 12px" />
>
>         <style tts:extent="60px 22px" />
>
>         <style tts:zIndex="1" />
>
>       </region>
>
>     </layout>
>
>   </head>
>
>   <body >
>
>      <div>
>
>       <p region="r2" style="rubyStyle">また</p>
>
>       <p region="r1" style="textStyle">頭を<span>
>
>         <ruby xmlns="http://www.w3.org/TR/ruby">
>
>           <rb>また</rb>
>
>           <rt>股</rt>
>
>         </ruby>
>
>       </span>股に突つ込んで祈るわ</p>
>
>     </div>
>
>   </body>
>
> </tt>
>
>
>
> Which should produce something like:
>
> [image: cid:image001.png@01C9FA95.E806CA40]
>
>
>
> This approach does have its limitations, and so if Timed Text starts to see
> a lot of adoption for Asian scripts, then this may well be something we will
> revisit should we be chartered to look at a version 2.
>
>
>
> Sean Hayes
>
> *From:* public-tt-request@w3.org [mailto:public-tt-request@w3.org] *On
> Behalf Of *Gur Shlapobersky
> *Sent:* 01 July 2009 4:56 PM
> *To:* Glenn Adams
> *Cc:* public-tt@w3.org
> *Subject:* RE: Ruby annotation
>
>
>
> Thanks for the explanation.
>
> Yes, I would like the TTWG to consider adding ruby annotation to the
> standard. The need for ruby annotations is common in Japanese subtitles and
> other languages.
>
>
>
> Sincerely,
>
>
>
> Gur S.
>
>
>
> *From:* Glenn Adams [mailto:gadams@xfsi.com]
> *Sent:* Wednesday, July 01, 2009 02:00
> *To:* Gur Shlapobersky
> *Cc:* public-tt@w3.org
> *Subject:* Re: Ruby annotation
>
>
>
>
>
> Assuming for a moment (without checking) that you are using the W3C ruby
> vocabulary as defined by the W3C ruby specification, then, since DFXP
> permits the arbitrary use of elements and attributes in foreign (i.e., non
> TT) namespaces, this is certainly legitimate. However, a compliant DFXP
> processor should be expected to ignore such foreign vocabulary unless the
> author includes a required extension designator in a profile element, where
> that extension designator would mandate semantic support for the foreign
> vocabulary (in this case, the W3C ruby vocabulary).
>
>
>
> For example, one might have the following:
>
>
>
> <tt xmlns="http://www.w3.org/2006/10/ttaf1" >
>
> <head>
>
> <profile ttp:xmlns="http://www.w3.org/2006/10/ttaf1"
> use="dfxp-presentation">
>
> <!-- require support for a "ruby" extension as defined by the following
> third party -->
>
> <extensions xml:base="http://www.example.org/ttaf1/extension/">
>
> <extension>#ruby</extension>
>
> </extensions>
>
> </profile>
>
> ...
>
> </head>
>
> <body>
>
> <div>
>
> <p>
>
> <ruby xmlns="http://www.w3.org/TR/ruby">
>
> ...
>
> </ruby>
>
> </p>
>
> </div>
>
> </body>
>
> </tt>
>
>
>
> Note well that all extension designations are 3rd party defined and
> interpreted, and there are no standard extension designations defined (yet)
> by the TTWG for use with DFXP. Consequently, such a use of ruby is
> effectively a private extension between two third parties, the author and
> the processor.
>
>
>
> While it is possible that the TTWG could define some standard designations
> for such extensions, we have had no requests to do so. Is that what you are
> requesting here?
>
>
>
> Regards,
>
> Glenn
>
>
>
> On Wed, Jul 1, 2009 at 2:55 AM, Gur Shlapobersky <Gur@captionsinc.com>
> wrote:
>
> Is this the proper syntax for ruby annotation as part of a <p> element?
>
> Should the TT Full profile assume support for ruby annotation?
>
>
>
> <p>
>
>   <rub:ruby xmlns:rub="http://www.w3.org/TR/ruby">
>
>     <rub:rb>WWW</rub:rb>
>
>       <rub:rt>World Wide Web</rub:rt>
>
>   </rub:ruby>
>
> </p>
>
>
>
> * http://www.w3.org/TR/ruby/
>
>
>
> Thanks,
>
>
>
> Gur Shlapobersky
>
> Software Development Manager
>
>
>
> Captions, Inc.
>
> 640 South Glenwood Place.
>
> Burbank, CA 91506
>
> 818.260.2722
>
> www.captionsinc.com
>
>
>
>
>
Attachments

image/png attachment: image001.png
Received on Thursday, 2 July 2009 06:58:47 UTC