W3C home > Mailing lists > Public > public-html@w3.org > January 2018

Re: Generalizing Markup for Interlinear Text Presentation

From: Chaals McCathie Nevile <chaals@yandex.ru>
Date: Thu, 18 Jan 2018 12:28:56 +0100
To: public-html@w3.org
Message-ID: <op.zc1biimrftbnq3@desktop-kurf4r9.pnp.gw>
On Wed, 17 Jan 2018 18:59:36 +0100, Daniel Yacob <yacob@geez.org> wrote:

> Florian,
>
> I am grateful for the response, thank you and better late than never

e too, because it reminded me that I wanted to respond...

> The primary limitations that I have encountered is support to present  
> more than one line of >interlinear text bound to and aligned with one or  
> more characters in the base text. The individual >lines must also be  
> independent from one another and will have separate semantics.  Nested  
> Ruby >markup at first appears to be a layout solution but the  
> overlapping markup problem intervenes in >some layout use cases.

Yep. There is a reasonably massive body of text that has multiple  
interlinear annotations, for all sorts of purposes. Hand-annotated  
administrative documents, multilingual documents or documents with  
multilingual annotations, ...

> CSS work arounds are available to create the desired visual effects, but  
> are really only suitable >for a single browser and version.

Which is to say they are not suitable for the Web :)

> The most successful work around leads to duplication of the base text  
> for each interlinear line >needed. A standard for multiline interlinear  
> layout requirements should address the portability and >duplication  
> issues.

Yep.

> The pages indexed in the link below provide a description of the Zaima  
> practice, a document model, >use-cases, and browser test pages.  If time  
> is limited I would direct you to the model and use->cases pages first.  
> Note that the HTML+CSS sample demonstrations are very, very sensitive to  
> >browser, browser version and font choices and have been optimized for  
> Chrome 61 and higher:
>
>  http://w3c.github.io/elreq/zaima/
> The model document attempts to present a general solution for Chant  
> semantics, then develops markup >that is specific to Zaima documents.  I  
> am not familiar with the special layout requirements of >other chant  
> traditions but simple inspection indicates that many requirements are  
> common.

Thank you for this. It is inter alia a helpful introduction to use cases I  
know nothing about.

I doubt that we want to try the approach of making a lot of specific  
extensions for specific types of content like chant - that generally leads  
to a lot more complexity than is manageable, and gets broken quickly.

However, I think it would be useful to consider a way to solve this class  
of problems.

There have been a lot of efforts on annotation of Web content that allows  
something more flexible than the strict tree structure forced by the DOM,  
but finding a sweet spot between making it simple enough to use so it gets  
traction in content, simple enough to implement and maintain that people  
do, and general enough to stop an explonential expansion of HTML, is a  
challenge...

It is probably worth opening something in the Web Patform Incubator, to  
follow this up. My sense is that if we don't have a fairly generic  
solution, we are unlikely to get very far.

I would also suggest prodding the Music notation community group since 1.  
This is music, and 2. they have similar sorts of issues even though their  
focus is on "traditional" (by which I think they mean modern) Western  
music notation, as far as I understand.

cheers

Chaals

> I'm sparing some details now but am happy to respond to questions.
>
> thank you,
>
> -Daniel
>
>
> On Tue, Jan 16, 2018 at 10:50 PM Florian Rivoal <florian@rivoal.net>  
> wrote:
>> Hi,
>>
>>
>>
>> Sorry for the very late reply.
>>
>>
>>
>> Other than the fact that ruby is called ruby and not a more general  
>> term, could you give some >>details about what is missing from the  
>> existing technologies (markup and css) to be able to >>adequately  
>> represent the kind of interlinear text you are thinking about?
>>
>>
>>
>> —Florian
>>
>>
>>
>>> On Nov 28, 2017, at 12:35, Daniel Yacob <yacob@geez.org> wrote:
>>
>>>
>>
>>> Greetings,
>>
>>>
>>
>>> HTML markup support for interlinear text is limited to Ruby annotation  
>>> use cases. Arguably Ruby >>is the most common and best understood  
>>> interlinear writing practice. There are other practices >>that have  
>>> layout needs in common with Ruby but then have their own additional  
>>> presentation >>requirements. While there may be many uses cases for  
>>> interlinear text, I'm personally only >>familiar with staffless chant  
>>> notation (e.g. Znamenny, Syric, Byzantine, Ethiopic, Hebrew, etc).
>>
>>>
>>
>>> Could the W3C WG for HTML take on the bigger problem of tag support  
>>> for interlinear text >>layout?  Ruby might then be treated as a  
>>> special case of interlinear text and chant as another.  >>Both could  
>>> apply a common base tag set, then more specialized tags as needed.
>>
>>>
>>
>>> I realize that this approach throws a wrench into established Ruby  
>>> markup, which could be >>grandfathered in and deprecated over time, or  
>>> not.  My main thrust here is to raise the need for >>broader  
>>> interlinear text layout support under HTML and make myself available  
>>> to discuss >>requirements with interested parties.
>>
>>>
>>
>>> thank you,
>>
>>>
>>
>>> -Daniel
>>
>>
>>
>



-- 
Chaals is Charles McCathie Nevile
find more at http://yandex.com
Received on Thursday, 18 January 2018 11:29:32 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 18 January 2018 11:29:32 UTC