[ih] First Eurociscos [was Ethernet, was Why TCP]

[Brian E Carpenter]

On 06/09/2016 10:28, Dave Crocker wrote:
> On 9/5/2016 2:26 PM, Brian E Carpenter wrote:
>> The game was over by 1991, although not everybody had realised it. The ITU didn't
>> really fold until 1995, but ISO certainly got there sooner.
> 
> 
> That's probably a reasonable date to cite for declaring its failure as 
> being clearly and publicly visible, but I'd claim it was undeniably over 
> some years before that.

It depended who you were and where you sat. For me the moment was when I
presented a talk called "Is OSI too late?" at the RARE* conference in May
1989 - although the talk gave a mixed answer to that qustion, the audience
reaction was conclusively "Yes". In January 1990, a committee of three and
a half (B.Carpenter, L.Backstrom, G.Pujolle, assisted by P.Kirstein)
reported to the RARE Council of Administration recommending that
"RARE recognises TCP/IP as complementary to OSI and preferable to
proprietary protocols for immediate use ... RARE recognises RIPE as an
appropriate body for current TCP/IP  coordination activities..."
and basically that's what happened. OSI efforts dragged on for a while,
however.

*The Europe-wide association of research networks at that time.

   Brian

> 
> I think I've noted this already, but around 1987 my department was 
> developing both various TCP/IP stacks as well as an OSI stack.  As work 
> progressed, we started asking our customer base about the kind of 
> products they might need to assist in the transition from using Internet 
> technologies to using OSI.
> 
> We were unprepared for how consistent the response was.  They had 
> serious interest only in going from OSI to TCP.  There was zero interest 
> in the other direction.
> 
> In my view, that was a game-over moment.
> 
> d/
>