Re: AUTH48 and "Guidance for NIST Staff on Using Inclusive Language in Documentary Standards"

[John Scudder <[e-mail]>]

Hi SM,

> On Jun 5, 2025, at 6:45 AM, S Moonesamy <[e-mail]> wrote:
> 
>  1. The I-D is reviewed by an Area Director.
> 
>  2. The community is given some time to review the I-D.
> 
>  3. The I-D is reviewed by the Internet Engineering Steering Group (IESG).
> 
> Let's assume that an author wrote something which might be against
> some rule.  It would only get through if the Area Director, the
> community and the Internet Engineering Steering Group were not paying
> attention.  Q was asked to go and find out whether the script might
> have missed some words.  

I don’t see where Q says that [1]. It’s not in the reply from Jean [2] either, which quotes the standard question that the RPC asks for all documents during AUTH48 and therefore is presumably verbatim what Q was asked:

```
   Please review the "Inclusive Language" portion of the online
   Style Guide
   <https://www.rfc-editor.org/styleguide/part2/#inclusive_language>
   and let us know if any changes are needed.  Updates of this nature
   typically result in more precise language, which is helpful for
   readers.
```

For that matter, I don’t know what “the script” you refer to might be. (Given the context of our discussion, the most obvious reading of “the script” is that you imagine there’s some automation, à la idnits, that’s supposed to do this work.)

> I don't find that very efficient.

Insofar as the question you’ve represented isn’t the one Q was asked [3], I don’t think it’s possible to come to any conclusion about “efficiency”.

If it would be correct to rewrite your last sentence as “… find out whether procedural steps 1-3 might have missed …” then at least I can understand what you’re saying. I still disagree with your conclusion that it’s “inefficient”, though, for two reasons:

First, every one of the parties in your list is a fallabile human (or group of humans). Things that can be automated, should be, and then one might complain about “inefficiency” if the authors are asked to repeat the work. What we have left, can't be automated (or hasn’t been). Making judgement calls about terminology is inherently non-automatable [4].

Second, none of the parties in your list has been tasked with ensuring the stylistic integrity of the document as a primary responsibility. Many people make a good-faith attempt, but in the end, we have a professional staff for that: the RFC Editor. Indeed, more than once over the last fiew years, the argument has been made (on this list, even) that Area Directors are doing too much editorial review, and should back off and leave it to the professionals. In particular, as AD I heard more than once that it wasn’t my job to address readability and style issues in the documents I reviewed, other than the most grossly disqualifying ones. 

In summary, I don’t think there’s something broken here [5], so let’s not fix it.

—John

[1] https://mailarchive.ietf.org/arch/msg/ietf/nPIakKEL2JT7KmPSluo1_XonN5A/, “… as part of this the RFC editor asks me to review my text for possible issues around inclusive and respectful language…"
[2] https://mailarchive.ietf.org/arch/msg/ietf/7OkHUQwLVP7M73VmvhUOBzyYjTs/
[3] I would say it’s a strawman, except that carries with it an implication of bad faith argument, and I don’t intend that. 
[4] Let us not give in to the temptation to take a sharp turn into a discussion about fobbing this off on an LLM.
[5] That is to say, nothing in this tangent is broken. As for the point Q kicked this discussion off with, that we reference a withdrawn document, that’s a different story. Wayback Machine notwithstanding, I agree that it would be desirable to reference some other, non-withdrawn, document.