19 Oct
2003
19 Oct
'03
11:12 a.m.
Peter Salus is quoted as saying
...
When the VAX was ``pre-announced,'' the Unix architects at Bell Labs had
become
disillusioned with DEC, they didn't like VMS and they thought that the VAX had
an ``offensively fat instruction set.'' Anyway, Steve Johnson and Dennis
Ritchie were working on their Unix port to the Interdata. (Which Steve referred
to as the ``Intersnail.'')
We were far from disillusioned, either with the company
or the design; see my contemporary transcription of Ossanna's
notes of the preannouncement presentation.
http://cm.bell-labs.com/cm/cs/who/dmr/vax.html
But it is true that our own attention was focussed on the Interdata
work at the time; not only was it underway, but for stretching
portability it looked useful to work on an architecture that
was Not "culturally compatible" with the PDP-11.
So DEC approached Charlie Roberts at AT&T in
Holmdel, NJ. Tom London, John
Reiser and Ken Swanson were interested; they got a VAX in early 1978. In three
months they ported Version 7 to the VAX. Roberts told me: ``We got the machine
in January, they had it running in April, and by August it really worked.''
>>
and indeed left the Vax port to Reiser and London. (VMS
didn't figure into the equation.)
A later poster, nao, asked about London and Reiser's memo
about their work on what became 32V (TM-78-1353-4).
This seems to be in the company archives, but not in scanned form.
I've ordered a paper copy, but the mechanism sometimes
is a black hole.
Dennis