[Info-vax] Rethinking DECNET ?

[Johnny Billquist]

On 2014-09-05 23:11, johnwallace4 at yahoo.co.uk wrote:
> On Thursday, 4 September 2014 20:03:54 UTC+1, Bob Koehler  wrote:
>> In article <lu9kdk$lo8$1 at Iltempo.Update.UU.SE>, Johnny Billquist <bqt at softjar.se> writes:
>>
>>>
>>
>>> That is true. In fact, I don't think any other company even did
>>
>>> implement an OSI compliant network stack at all. Which is a good sign it
>>
>>> was not going to be the future of networking...
>>
>>
>>
>>     HP and Sun both shipped ISO/OSI network products.  Or would have if
>>
>>     anybody ordered a copy.
>
> OSI products were also shipped by a number of industrial automation
> vendors, and their equivalents from the design end of things. I'm
> slightly puzzled by the claim that DEC were first to ship a full OSI
> stack. I'd probably be happier with "first to ship a fully integrated
> OSI stack", but long before that happened there was VOTS, OSAK,
> DEC/MAP and DEC/OSAP, and FTAM, and the X.whatever email stuff, and
> probably more.

I definitely remember DEC saying, and at the time as far as I can 
remember, also acknowledged by others, that they were the first to 
implement the OSI model. I think that others did parts of it, but noone 
else had done the full monty.
But this is now long ago enough that I don't remember all the details 
anymore, and would have to go dig for facts, which at the moment I don't 
have the motivation for...

[...]

> It made a great deal of sense back then. Industrial automation was
> a mess of incompatible protocols frequently running over incompatible
> hardware. Much the same applied to application integration at the design
> end (including basic stuff such as in-house or inter-company email, never
> mind complicated stuff like actual file transfer).

Well, industrial automation (especially back in the 80s) did not really 
have the computer resources for full fledged OSI implementations.

> What happened?
>
> I think the executive summary is that some years later. Microsoft
> happened, they claimed GOSIP/POSIX compliance witn NT's alleged POSIX
> subsystem (or claimed grandfather rights for Win32 or otherwise got
> themselves into the bidding process) and the rest is history (even
> if Exchange did use X.400 for quite a while).

I would say that what happened was that already in the 80s, TCP/IP 
proved that it already existed, it already worked, and was already 
available on multiple platforms. In addition, it was actually a project 
funded by the US Department of Defence, and of course they were using it.

OSI, in contrast, was a committee deisgn, with very few full 
implementations around (did really any other that DEC do it?), 
interoperability was still not proven, it was much more heavyweight that 
TCP/IP, and in the end, people took what worked.

Microsoft was not even on the horizon back in those days.

	Johnny