Re: The IETF, Standards process, and the impact on the RFC series document production

[Keith Moore <[e-mail]>]

On 10/4/19 10:40 PM, Brian E Carpenter wrote:

>>> Implementing our standards involves a treasure hunt
>>> to find how many RFC have to be read before understanding the whole
>>> picture.
>> This is problematic, but inherent, not so much in the RFC series, but in
>> the way we've chosen to do our standards process.
> Bingo! Changing the RFC series*will not*  fix this problem.

If I understand what problem is being described, I'm not sure changing 
the standards process would address it either.   (Though I do suspect 
that some changes are in order for other reasons.)

Most standards that are used on a large scale will need to be amended 
and/or extended over time.    Attempting to get everything right the 
first time can easily doom a standard to failure for any of multiple 
reasons (e.g. irrelevance, or too much complexity, or too little ability 
to adapt to changing conditions).

In a sense, the extensions and amendments are signs of success. If 
nobody used the standard, there would probably not be sufficient energy 
to make the changes.

Keith