[ih] early competition and networking

[John R. Levine]

On Mon, 15 Apr 2024, John Day wrote:

> To some degree, ‘back in the day’ was pretty recently. ;-) Long after the fundamental problem was solved. As I just noted to Dave, the original problem was the existence and expectation of many heterogenous networks of different technologies, some multi-access, some switched or relayed, some circuit-switched, virtual circuit switched, datagram, etc.
>
> The problem was how to interconnect them.  A solution was arrived at. From what I can tell, somewhere before 1976 or so.
>
> The question is precisely what was it? Or perhaps more to the point, what did people think it was?

i don''t think there was or is a single answer.

One answer was layering, so you can have consistent behavior at upper 
levels while having different lower levels.  We take this for granted now, 
e.g., IP packets in my office are flowing over twisted pair, wifi, 5G, and 
fiber, but I think it's only obvious in retrospect.

The other was gateways, translate traffic from one form to another.  Some 
of the translations are conceptually simple, like turning phone numbers 
from North American en-bloc signalling to CCITT compelled signalling. 
Some is very baroque like the mail gateways Dave worked on.

There may be others but at the moment I can't think of any.

R's,
John

>> On Apr 14, 2024, at 23:11, John Levine via Internet-history <internet-history at elists.isoc.org> wrote:
>>
>> It appears that Dave Crocker via Internet-history <dcrocker at bbiw.net> said:
>>> My understanding is that the design model then and now is a homogeneous
>>> service, on top a variety of heterogeneous services. Hiding the
>>> differences by providing a common service on top of them.  That the
>>> common service was originally specified as one layer and evolved into
>>> two does not change the meta-design approach.
>>
>> Back in the day we moved a fair amount of mail via UUCP, a store and
>> forward dialup scheme quite different from SMTP. The format of the
>> mail messages was the same, and once we had the uucp mapping project
>> to figure out the routing, the addresses looked pretty much the same,
>> too, e.g. my address was johnl at ima.uucp. The sendmail mail program,
>> which is still widely used today, has a complex internal language to
>> describe how to route various addresses while keeping everything above
>> the same.
>>
>> The ability to use the same higher level software and just futz with the
>> lower levels was and is quite powerful.
>>
>> R's,
>> John
>> --
>> Internet-history mailing list
>> Internet-history at elists.isoc.org
>> https://elists.isoc.org/mailman/listinfo/internet-history
>
>

Regards,
John Levine, johnl at taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly