[ih] This Review is for Everyone

[Craig Partridge]

On Sun, Mar 22, 2026 at 7:26 PM Noel Chiappa via Internet-history <
internet-history at elists.isoc.org> wrote:

> The move to effectively sideline Dave by splitting it in two was an
> elegant,
> and somewhat gentle, solution to a real problem. I am reminded of a
> similarly
> elegant solution to a similar problem at Bletchley, during WW2, by Hugh
> Alexander (described in Hodges, "Enigma", at the bottom of pg. 227). My
> memory dropped a bit, though, because I thought the shift supervisor
> involved
> was Turing, but I see it was someone else.
>

Speculating here, but I would guess Dave [Mills -- too many important Daves
in the early Internet*] might have, in retrospect, felt it was a good thing
too.  It freed Dave to focus on desperately important problems that he was
uniquely positioned to address.  Notably:

   - Dave's Fuzzballs were the NSFNET routers in 1986/1987 and so regular
   software engineering attention from Dave was required to simply keep the
   (under resourced) NSFNET backbone working.
   - EGP was breaking and the community needed to understand if that was
   breakage because EGP was critically flawed or simply needed fixing.  Dave
   led the effort to fix EGP and, in my view, it was his SIGCOMM '87 paper
   with Hans-Werner Braun that said "this isn't gonna work".  It was another
   18 months (and substantial arguments about what was and was not feasible)
   until BGP came along, but Dave M. and HWB were able to refine the problem
   definition substantially.
   - AND he got to push NTP to a mature state (probably his greatest
   contribution)

It was these contributions that made Dave an ACM and IEEE fellow and member
of the National Academy of Engineering and a person for whom a New York
Times obituary was appropriate.

Craig

PS: Digging a bit as I wrote this I discovered the name of the IAB Task
force that split into IETF and INarc -- it was GADS (Gateway Algorithms and
Data Structures Task Force).

PPS: Too many Daves:  Dave Clark, Dave Crocker, Dave Farber, Dave Mills,
Dave Sincoskie, etc.  In the early 1990s there was a major research project
called DAWN and the joke was it was an acronym for "Dave's Awesome and
Wonderful Network".


-- 
*****
Craig Partridge's email account for professional society activities and
mailing lists.