Re: Query about procedure: Tombstones?

[Ralph Droms <rdroms at cisco.com>]

I had two problems...

First, draft-ietf-dhc-agentopt-radius-03.txt (of which I am a co-author) was
expired prematurely and replaced by tombstone
draft-ietf-dhc-agentopt-radius-04.txt.  I read the tombstone to mean that
draft-ietf-dhc-agentopt-radius-04.txt had expired.  I couldn't remember ever
submitting draft-ietf-dhc-agentopt-radius-04.txt, and I couldn't find any
record of submitting draft-ietf-dhc-agentopt-radius-04.txt (after searching
for a while and contacting the co-author).  I sent e-mail to internet-drafts
to check on the problem, and we discovered
draft-ietf-dhc-agentopt-radius-03.txt had been expired prematurely.  Had I
know draft-ietf-dhc-agentopt-radius-04.txt meant that
draft-ietf-dhc-agentopt-radius-03.txt was expired, I wouldn't have spent so
much time checking my records to understand where
draft-ietf-dhc-agentopt-radius-04.txt came from.

Second, draft-ietf-dhc-failover-12.txt expired (without my noticing) and I
referred to draft-ietf-dhc-failover-12.txt in a WG last call.  I got a note
pointing out that draft-ietf-dhc-failover-12.txt didn't exist and
draft-ietf-dhc-failover-13.txt was the tombstone.  Since I hadn't submitted
draft-ietf-dhc-failover-13.txt (I'm a co-author), I had to check with the
current editor to see if he had submitted draft-ietf-dhc-failover-13.txt, so
I could find out what changes he had made.  He looked in his records and
couldn't figure out where draft-ietf-dhc-failover-13.txt came from.  Again,
we wouldn't have looked around for the source of
draft-ietf-dhc-failover-13.txt if we had known it was the tombstone.

Not a big deal in either case (and I apologize for the long-winded
explanation).  But the lack of an announcement caused us
both some amount of pain...

- Ralph

At 01:50 PM 12/9/2003 -0800, Harald Tveit Alvestrand wrote:
> Ralph,
>
> --On 9. desember 2003 15:19 -0500 Ralph Droms <rdroms at cisco.com> wrote:
>
> > Was this change to the tombstone procedure announced and discussed?  I
> > just
> > discovered the new procedure the hard way by tripping over conflicts
> > between
> > a reference to a draft-n that was replaced by tombstone-n+1.
>
> what was the nature of the "trip-up"?
>
> The change was instituted following complaints that we were losing 
> information, and made it hard to keep track of all versions of drafts - 
> some subpoena requested access to some recently expired drafts, and the 
> I-D directories were only producing tombstones, the drafts having been 
> overwritten.
>
> So we did a brief internal discussion and couldn't see any downside, and 
> Just Did It. Guess we didn't check with everybody ... but then again, at 
> the time, it seemed like an issue not worth spending all that much time on....
>
>                   Harald