Re: [xml2rfc] [irsg] UPDATE regarding <u> Re: AUTH48: RFC-to-be 9340 <draft-irtf-qirg-principles-11> for your review

[Carsten Bormann <[e-mail]>]

Hi Sandy,

this is an interesting discussion.

I wasn’t realizing that any specific version of xml2rfc creates a repertoire of characters directly supported in PDF generation, and that using characters outside of that repertoire requires some development effort (font selection/qualification, font integration, new xml2rfc release).

I believe that authoring tools (such as kramdown-rfc’s echars — explain characters [1]) should be able to inform authors that this might be the case well before the document reaches the RPC, so they can make an informed decision which out of a potential choice of specific characters they should be using and can, if necessary, alert the RPC that new characters may be coming up.

This would work best if a machine-processable form of the current repertoire were available somewhere, e.g., as a result of releasing a new version of xml2rfc.

(It seems to me that, as demonstrated by rfc-to-be 9340, the IRTF stream is the most likely one to benefit from the use of characters in wide use for a specific domain, hence my keeping of irsg in the CC.  Please remove if we dive into xml2rfc specifics.)

Grüße, Carsten

[1]: https://mailarchive.ietf.org/arch/msg/rswg/y2tKjY-RRxwcYjAZoHAzuGtmJC0


> On 2023-03-03, at 00:17, Sandy Ginoza <[e-mail]> wrote:
> 
> Hi Spencer,
> 
>> On Mar 1, 2023, at 5:02 PM, Spencer Dawkins at IETF <[e-mail]> wrote:
>> 
>> Hi, Sandy, 
>> 
>> On Wed, Mar 1, 2023 at 6:16 PM Sandy Ginoza <[e-mail]> wrote:
>> Greetings authors,
>> 
>> As you know, this document will be the first to include math symbols within flowing text (i.e., within <t>).  We ran into an issue with the fonts for ➔ in the PDF.  Pease note that we intend to convert this to → (U+2192).  In addition, this requires an update to xml2rfc — we will publish this RFC once a new release is available. 
>> 
>> Please let us know if you have any questions.
>> 
>> I'm not an author (on this draft), but I do have a question. 
>> 
>> You now have more experience with documents that include math symbols within flowing text than I do, and perhaps others. 
>> 
>> Is there a web page that will contain a list of characters that the RFC editors have needed to change, so that we can make our communities aware of this, before our documents get to you?
> 
> Not currently - we have had very few documents with non-ASCII characters that caused issues in this way.  
> 
> We’re discussing this issue with the RPC Advisory Team.  I hope we can develop guidelines the RPC can follow and point to from the Style Guide.
> 
> Thanks,
> Sandy 
> 
>> 
>> Best,
>> 
>> Spencer 
>