Re: [Cellar] math in IETF specifications

[Adam Roach <[e-mail]>]

On 12/2/15 1:22 PM, Dave Rice wrote:
> Dear IETF lurkers,
>
> The current version of the FFV1 specification includes a lot of latex 
> style mathematical equations to express various aspects of the 
> encoding format. These can be seen in the markdown in locations like 
> this 
> https://github.com/FFmpeg/FFV1/blob/master/ffv1.md#range-binary-values and 
> in rendered form in HTML, 
> http://www.ffmpeg.org/~michael/ffv1-markdown/ffv1.html#range-coding-mode 
> <http://www.ffmpeg.org/%7Emichael/ffv1-markdown/ffv1.html#range-coding-mode>, 
> and PDF, http://www.ffmpeg.org/~michael/ffv1-markdown/ffv1.pdf 
> <http://www.ffmpeg.org/%7Emichael/ffv1-markdown/ffv1.pdf>.
>
> I've been investigating to see how other RFC documents integrate latex 
> style mathematical equations but have not found any? If there a 
> recommended manner to include math? Should it be translated into a 
> narrative? Is there a recommended way to render math in a plain text RFC?
>

There's an effort underway to modernize the RFC format, and the new 
format will allow for significantly better formatting of things like 
mathematical formulae. See:

https://tools.ietf.org/html/draft-flanagan-rfc-framework

and

http://www.rfc-editor.org/rse/format-faq/

It seems quite likely that this work will be reaching completion in 
about the same timeframe as CELLAR is ready to publish its final output.

In the meanwhile, there is precedent for publishing both .txt and .pdf 
versions of documents, with the .txt versions referring to the .pdf for 
information that cannot be rendered in text.

Another approach that has been taken is expression of this kind of 
information in algorithmic formats, using pseudocode or actual 
programming languages.

/a