[rfc-i] RFC folklore
sm at resistor.net
Thu Apr 26 09:27:24 PDT 2012
[following up rfc-interest]
At 07:10 26-04-2012, Phillip Hallam-Baker wrote:
>I really dislike the fact that a lot of the Internet specification has
>turned into folklore. There is what is written down and then there are
>the additional rules that are not written down anywhere or if they are
>are written poorly.
"Out of scope" is sometimes used as a colourable device in a
RFC. The assumptions are not explained. That could have be done in
a companion document. I was working on an old specification. It
contained material which explained the decisions.
There's some folklore in the SDO circle which are considered as
facts. Some RFCs are poorly written. It can be a problem when other
work builds upon them. I am aware of very few authors who will take
the blame when their RFC is misinterpreted.
>The problem with the current approach is that someone wants to do
>something, they get told 'no' and so they do it anyway but in a way
>that nobody else can then build on top of because its badly documented
>and makes no allowance for extension. Things like grabbing the
>unprefixed TXT record for anti-spam alone.
There is the Independent Stream. It's there for a purpose. I don't
see anything wrong with publishing a critical review of the approach.
More information about the rfc-interest