Re: Lawyer's advice on boilerplate in RFCs and I-Ds

todd glassey wrote:

> let me re-ask it - how many RFC's have had BIS 
variants produced?

Something like "3066bis", "2821bis",
"2822upd", or "son-of-1036" are
informal names for Internet Drafts (or memos not published
as I-D).

I-Ds have an "offical" name, it starts with draft-
and ends with a
version number (or a version number + file extension like
.txt), e.g.
draft-klensin-rfc2821bis-01

Some I-Ds are approved by the IESG to be published as RFC. 
At this
time the ID still has its "official" I-D name, but
it's actually an
approved RFC waiting for its number.

Authors and RFC-editor under the oversight of a responsible
document
shepherd will fix some editorial issues like replacing
"RFC xxxx" by
the number (assigned by the RFC editor), so that last I-D
can still
be edited before it's eventually published as RFC.

After publication RFCs don't change, typos etc. can be
submitted as
errata, but the master copy and its RFC number are frozen
forever.

As an example the I-D informally called 3066bis is now RFC
4646, and
the only things that can be changed about it are the
"status" (now a
BCP, or rather a part of a BCP) and any errata published
separately.

Of course new I-Ds like "4646bis" are free to
claim that they would
obsolete RFC 4646, but if and when that happens this
"4646bis" would
get its own RFC number.

The numbers are not used strictly sequentially, there can be
gaps
without published RFC, and RFC x with x < y can be
published after
RFC y.

> I have also noticed other BCP documents as well that
get reved...

BCP numbers, STD numbers, and FYI numbers are a different
issue:

The BCP known as "BCP 47" used to be RFC 3066 for
some years, now
BCP 47 is RFC 4646 + RFC 4647 (a BCP with two RFCs).  These
BCP
numbers actually stand for sets of published RFCs, and these
sets
can be modified.  I can't tell if they've ever been modified
to an
empty set so far, check it out.

Frank



_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg

--On Wednesday, March 21, 2007 19:37 +0100 Frank Ellermann 
<nobodyxyzzy.claranet.de> wrote:

> todd glassey wrote:
>
>> let me re-ask it - how many RFC's have had BIS 
variants
>> produced?
>
> Something like "3066bis",
"2821bis", "2822upd", or
> "son-of-1036" are informal names for Internet
Drafts (or memos
> not published as I-D).
>
> I-Ds have an "offical" name, it starts with
draft- and ends
> with a version number (or a version number + file
extension
> like .txt), e.g. draft-klensin-rfc2821bis-01

But this is completely arbitrary stuff, often just author 
preference.    Taking the example you cite, had my mood been

different on the day I submitted the initial version, it
could 
have easily have been
   draft-klensin-rfc821ter-00
   draft-klensin-smtpupd-00 (normally, that would have been
the 
I-D that became 2821, but _that_ I-D had a
draft-ietf-drums... 
name)
   or any number of less mnemonic things.

This clearly does not have any IPR implications are all,
since 
the IETF doesn't own nearly enough trademark rights to
prevent 
posting of an I-D with "rfc" and some digits
imbedded in its 
name.  I hope we never get into enough silly states that the

IESG has to impose specific approval conditions for posting

something named draft-AUTHORNAME-rfcNNNNbis but, even if
they 
did, that would be a rule about file names for posting, not
an 
IPR issue.

So the point was?

      john


_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg