[rfc-i] v3imp #4 Ruby text

Sean Leonard dev+ietf at seantek.com
Wed Jan 28 08:44:19 PST 2015

On 1/23/2015 8:39 AM, Tony Hansen wrote:
> to the design team:
> On 1/23/15 4:04 AM, Sean Leonard wrote:
>> Improvement Need
>> #4 Ruby text
>> This improvement calls for support for ruby text, also known as 
>> interlinear annotation. If you are not familiar, check out 
>> <http://en.wikipedia.org/wiki/Ruby_character>.
>> Support can be in markup form {<ruby> <rt> <rp> -- see HTML5} or by 
>> supporting the raw Unicode code points {U+FFF9 U+FFFA U+FFFB}.
>> Personally I think the Unicode code points are sufficient for the 
>> canonical format; a formatter can convert these codes into 
>> appropriate markup (e.g., HTML5 <ruby>). However as our own Martin J. 
>> Dürst is the co-author of UTR-20 
>> <http://www.unicode.org/reports/tr20/#Interlinear>, the markup 
>> position may win out.
>> My biggest concern is that introducing <ruby> markup into the v3 
>> format may significantly complicate canonical processing, since 
>> interlinear annotation is applicable to structured text fields (e.g., 
>> author names, document title, references), not just unstructured 
>> spec-text. HTML5 really goes quite overboard with the ruby elements; 
>> I believe it is unreasonable to require support for that level of 
>> complexity from all xml2rfc-related tools. Furthermore, putting ruby 
>> into artwork *should* be allowed—this requires no additional work for 
>> the v3 vocabulary since the Unicode code points are already allocated.
> Since V3 is Unicode-based, and Unicode is allowed in all the places 
> Sean seems to be concerned about, I think this is covered.

+1000. That is my point.

> This only thing I'm not sure about is the use of ruby in artwork -- 
> would that just be ruby annotations on words found within the artwork, 
> or is Sean thinking it would be used in another fashion? If the 
> former, I think it's okay.
It's the former.

> The only real question I have is whether there needs to be an explicit 
> statement about ruby being supported or not.

Since V3 is Unicode-based, all reasonable Unicode characters should be 
acceptable in all contexts, subject to the editorial process.

This statement and the request for ruby/interlinear annotation are 
consistent with draft-flanagan-nonascii-03. In a follow-up e-mail I will 
cite the requested specific examples.


More information about the rfc-interest mailing list