[rfc-i] Proper status for pre-IETF RFCs currently with "unknown"

John C Klensin john+rfc at jck.com
Sun Nov 6 06:31:32 PST 2011

--On Tuesday, November 01, 2011 04:14:01 +0800 Fred Baker
<fred at cisco.com> wrote...

> On Nov 1, 2011, at 2:59 AM, Andrew G. Malis wrote:
>> There are a whole bunch of pre-IETF RFCs who's status is
>> currently "unknown", including some of mine. I started a bit
>> of a discussion on the IETF list by sending a request to the
>> IESG to change my pre-IETF RFCs to "Historic" status. I've
> The problem is that it's not obvious. RFCs 768, 791-793, and
> perhaps 896 should probably be "standard", many should be
> "historic", and some (consider RFC 970 for example) are white
> papers that probably deserve to be left as "informational" or
> "experimental". Consider the many telnet RFCs...
> As I recall, the reason these were left "unknown" way back in
> the musty dusts of history was that nobody wanted to take the
> time to sort through them and decide. There isn't a blanket
> rule that really makes sense.

I think a little more than that, actually, but that is probably
most of it.  The complex dependencies that SM and others have
pointed out are another piece of the puzzle.

I've been involved in a few side-conversations about this and
have a different proposal.  It is a little bit radical, but, I
think, much cleaner and easier.

As background and using today's vocabulary, the IETF/IESG have
no obvious authority to create categories or even assign them
for anything but the IETF Stream.  Strictly speaking, the IESG's
statements about assignment of Historic classifications applies
only to IETF Stream documents.   It would be rude at best for
the IESG to try to reclassify Independent Stream, IAB Stream, or
IRTF Stream documentsl I suggest that principle extends to
documents created before there was an IETF and even longer
before the IETF started publishing (or requesting the RFC Editor
to publish) "standards track" documents in the RFC Series.  

So I propose that we ask the RFC Editor to create a new
category, called, e.g., "ARPANET".  The list of categories in
RFC 2026 need not be changed (unless the IETF Stream wants to
use it for IETF Stream documents) because, seen as above, the
categories in 2026 are relevant to the IETF Stream.  We then
reclassify all of the ARPANET-period Network Working Group
documents, or at least those that are listed as "UNKNOWN", into
that category.  Unlike "Historic", it has no implications of
"deprecated" or "useless" (whether that is true in practice or
not).  There are, consequently, no implications about
dependencies from later documents.  

For new IETF Stream documents, the IESG would have to figure out
how to handle references to documents of category "ARPANET", but
treating them the same way Information documents are treated
would be a good approximation. 

> Are you volunteering for the historical exercise?

IMO, the exercise associated with the above is a lot less
complicated (and errors would be less serious) than with the
Historical one, but a support a small collection of virtual and
actual greybeards to go through and classify things rather


More information about the rfc-interest mailing list