[ih] This Review is for Everyone

[Greg Skinner]

On Mar 22, 2026, at 11:23 AM, Craig Partridge <craig at tereschau.net> wrote:
> 
> 
> On Sun, Mar 22, 2026 at 11:51 AM Greg Skinner via Internet-history <internet-history at elists.isoc.org <mailto:internet-history at elists.isoc.org>> wrote:
>> 
>> I have seen references to an “Internet Engineering Task Force” that existed prior to the first meeting of the IETF [1] in articles such as this one about Barry Leiner. [2] A quote from the article:
>> 
>> A number of Task Forces were created under the IAB including one called "Internet Engineering", which ultimately gained responsibility for managing the other task forces; these were later renamed "working groups" except for those that were clearly associated with research, which were assembled into "research groups" of the Internet Research Task Force. By agreement with the IAB in the early 1990s, the hands-on responsibility for the Internet standards was passed from the IAB to the Internet Engineering Task Force (IETF) leadership. The IAB continued to play an important role in overseeing certain aspects of the work of the IETF and in developing independent views on critical architectural matters concerning the Internet. For many years, Leiner had the ultimate responsibility for the development of this part of the Internet organizational structure, which he oversaw during the mid-1980s.
> 
> Not quite as I remember it, as someone getting into that community around 1984/1985.
> 
> There were several Task Forces created under the IAB c. 1983,  roughly a Task Force for every IAB member not the chair.
> These included the End2End TF, the Privacy and Security TF,  and (I may get the name wrong), the Internet Architecture TF (INarc).  INarc split in 1986 into IETF, centered on right-now Internet challenges, and a more experiment/future-focused task force (may still have been called INarc, I can't recall -- Dave Mills chaired it and it persisted until c. 1991).  I was not privy to the dynamics that led to the split other than hearing rumors that there were too many operational problems that required attention and having them co-exist in a task force that was also trying to be looking multiple years in the future was awkward.
> 
> (Useful to remember that IAB "Task Forces" at the time were typically about 10-15 people, participating by invitation only, who met a few times a year.  So imagine you're in 1986: the Internet desperately needs vendor-independent Intra-AS routing protocols, a working inter-AS routing protocol [EGP is struggling], a network management solution beyond Telnetting to each router, a mechanism to address congestion storms, standard profiles for routers [aka Router requirements], etc... and that's all on the desk of 10 people who are also trying to figure out how to deal with impending 100Mbps, soon gigabit networks, and whatever interesting problems that future portends. Creating IETF [which was initially invitation only] and then opening its membership was a logical solution.)
> 
> As the pressing Internet challenges grew, and because about half the IAB TFs weren't regularly meeting, the IETF scope grew and c. 1989, the major Internet standards efforts were 95% IETF plus about 5% End2End and Privacy and Security.  In some cases, the efforts were collaborative: Host Requirements had a core of End2End folks (the chair of E2E chaired Host Requirements) but was in IETF, similarly the TCP extensions for high performance (which were developed in E2E and brought to IETF), and, I think, IPsec had a similar dynamic with Privacy and Security.
> 
> Then the Kobe restructuring happened and there were just two Task Forces: IETF and a new Internet Research Task Force (IRTF) and the non-IETF IAB TFs that still functioned became part of the IRTF.
> 
> Craig
> 
> --
> *****
> Craig Partridge's email account for professional society activities and mailing lists.

There was some discussion on the tcp-ip list during December 1985 indicating that a split was in the works. [1] Mike Muuss had complained to the list about the core gateways running out of space. The ensuing discussion included a response by Dave Mills that the issue would be brought up at the GADS meeting in January 1986, which is identifed in (current) IETF literature as the first IETF.

--gregbo

[1] https://github.com/matthewgream/www-securitydigest-org/blob/master/tcp-ip/archive/1985/12.txt.gz