On Fri, Jan 15, 2016 at 1:00 AM, Eliot Lear <lear@xxxxxxxxx> wrote:
> So... {useless history discussion follows}
>
> On 1/15/16 1:00 AM, John C Klensin wrote:
>>
>> --On Thursday, January 14, 2016 15:48 +0100 Eliot Lear
>> <lear@xxxxxxxxx> wrote:
>>
>
>> Whatever debate you are referring to, the IETF wasn't having it.
>> We didn't retire the "OSI Integration" Area (with Erik Huizer
>> and Dave Piscitello as the last two co-chairs) until circa IETF
>> 25 in November 1992 and my recollection is that it was clear
>> until not long before that that the plan was to get Internet
>> applications and upper layers running over an OSI substrate.
>
> TCP/IP over OSI first...
>> By the time the area was finally shut down (and most or all of
>> the remaining WGs folded into Applications, with Eric moving
>> over), the goal was clearly to figure out which OSI applications
>> were in use and/or likely to be worth anything going forward and
>> get them to work over TCP and/or IP.
>
> OSI over TCP/IP later...
>
> GOSIP wars in the industry in between. Ross Callon probably remembers more.
>
> And also in between, there was an amusing discussion about what people
> actually put on their business cards re X.400 (Chuck Hedrick from
> Rutgers @ Hawaii - I was in the room; it was quite funny, which is why I
> remember it).
I wasn't part of the IETF end of the OSI discussion but either your
dates are off or they are hilarious.
Thing is that the first time I was in a meeting at which an OSI
transition plan was discussed was July or so of 1992. And it was very
much pre-planning since the only vendor that was pushing OSI
networking was Digital and DECnet Phase V wasn't due for several
years.
So we were starting efforts at the very time IETF was shutting them down.
Exactly 12 months later, July 1993, OSI was effectively dead. The Web
content was all on the Internet and the Web was growing at 10,000% a
month. The only thing OSI could do would be to provide a slower, less
reliable Web through a proxy.