[ih] More Topology, Packet Radio

[Noel Chiappa]

    > From: Barbara Denny

    > The PE code (some or all?) could also have been written by Holly Nelson,

Ah, I knew Holly slightly; she came to some TCP/IP-related meeting at MIT-LCS
(maybe she'd be listed in the attendee sections of the early TCP/IP meeting
minutes IENs). Couldn't remember her name, though!

My fairly strong recollection is that she worked on the PE at a later stage in
its life - after the version that MIT got and tweaked. (Later: yes, see
below).) I don't recall the origin of that impression; maybe it was
conversations I had with her during her visit.

I'm pretty sure Jim Mathis wrote the version I had; it uses his Macro-11
'structure definition' macro technique (pretty impressive, to have structure
definitions in an assembler program :-).

    > SRI technical reports usually included everyone who worked on an effort
    > so I don't think the PE could have been written by a person not listed
    > on the report.

I recall seeing an SRI TR on the PE. I looked online, and found it;
TR-1080-140-1; there are two copies: DTIC has a scan of an original:

  https://apps.dtic.mil/sti/pdfs/ADA156186.pdf

and there's a cleaned-up, reformatted version here:

  https://www.warthman.com/images/SRI%20ARPANET-IMP.book.pdf

Two interesting things I see in it. It describes it as using an ACC 1822
interface; the one we had used an SRI interface (a manual for which is also
available throught DTIC). Also, it says it's written in BLISS-11; the one I
have is entirely in MACRO-11. I think that version must be the one that Holly
worked on; I think the MACRO-11 one must be Jim's.


    > From: Lawrence Stewart

    >> The problem was that the DH interface _also_ had ground ("the cable
    >> shields should be very solidly connected to the host's signal ground")

I really need to learn to turn my brain on before posting. It is, of course,
not true that tying the shields to ground at the host provides a ground from
the IMP; but I'm kind of right, anyway, in the case where a 'host' 1822
interface is used to emulate an IMP.

That's because in a 'host' 1822 interface, the cable shield is connected to
the host's ground - so if there's a 'host' 1822 interface on both ends, the
cable shields will be connected to the host's ground at each end, tying the
two hosts' grounds together.

    > I found both the 1975 and the 1978 versions of the BBN-1822 report and
    > both say "Ground Isolation is provided by the IMP" but the circuits in
    > Appendix D do not have this.

Yeah, I'd noticed that.

    > The 1975 version says the signals are transformer coupled and the 1978
    > version says optically isolated.

Yeah; either way the ground is isolated, though.

	Noel