Are there any documented or remembered instances PDP-7 or Interdata 8/32 UNIX being installed on any such machines for use in the Bell System aside from their original hosts? Along similar lines, was the mention of PDP-7 UNIX also supporting the PDP-9 based solely on consistencies in the architecture or did this early version of UNIX actually get bootstrapped on a real PDP-9 at some point?
My understanding of the pre-3B-and-VAX landscape of UNIX in the Bell System is predominantly PDP-11 systems, but there was also work in the late 70s regarding 8086 hosts as evidenced in some BSTJ and other publications, and there is the System/370 work (Holmdel?) which I don't know enough about to say whether it technically starts before or after UNIX touches the VAX.
Thanks for any info!
- Matt G.
Hi All.
This is a bit off-topic, but people here are likely to know the answer.
V7 had a timzone function:
char *timezone(int zone, int dst);
that returned a timezone name. POSIX has a timezone variable which is
the offset in seconds from UTC.
The man pages for all of {Net,Free,Open}BSD seem to indicate that both
are available on those systems.
My question is, how? The declarations for both are given as being in <time.h>.
But don't the symbols in libc.a conflict with each other? How does a programmer
on *BSD choose which version of timezone they will get?
Feel free to reply privately.
Thanks,
Arnold
> Paul -- you left out the other "feature" -- the noise, which was still
deafening even with a model N1 and its cover.
It was indeed loud, but GE out-roared them with a blindingly fast card
reader. The machine had a supposedly gentle touch; it grabbed cards with
vacuum rather than tongs. But the make-and-break pneumatic explosions
sounded like a machine gun. A noise meter I borrowed from the Labs' tool
crib read 90db 6 feet away.
Doug
I have been working with a VAX780 sim running
Unix System V r2 VAX780 and am having strange
issues.
TERM is defined at vt100
When firing up vi at times the cursor is positioned
in the wrong place or when inserting text it over
writes areas on the screen.
I have tried vt100, vt100-am, vt100-nam and none
work as expected.
Any ideas? Thanks
-Ken
Happy holidays
--
WWL 📚
Thanks Ken! I hadn't, even considered the PDP-15. Looks like SimH supports it, that could make for an interesting project...
- Matt G.
On Monday, December 18th, 2023 at 6:12 PM, Ken Thompson <kenbob(a)gmail.com> wrote:
> the pdp-7 was run on the almost compatible
> pdp-9 and pdp-15 computers.
> i dont think that version of unix ever made
> it out of the research department.
>
> On Mon, Dec 18, 2023 at 5:54 PM segaloco via TUHS <tuhs(a)tuhs.org> wrote:
>
>> Are there any documented or remembered instances PDP-7 or Interdata 8/32 UNIX being installed on any such machines for use in the Bell System aside from their original hosts? Along similar lines, was the mention of PDP-7 UNIX also supporting the PDP-9 based solely on consistencies in the architecture or did this early version of UNIX actually get bootstrapped on a real PDP-9 at some point?
>>
>> My understanding of the pre-3B-and-VAX landscape of UNIX in the Bell System is predominantly PDP-11 systems, but there was also work in the late 70s regarding 8086 hosts as evidenced in some BSTJ and other publications, and there is the System/370 work (Holmdel?) which I don't know enough about to say whether it technically starts before or after UNIX touches the VAX.
>>
>> Thanks for any info!
>>
>> - Matt G.
> On 17 Dec 2023, at 13:02, KenUnix <ken.unix.guy(a)gmail.com> wrote:
-8<—
> I have tried vt100, vt100-am, vt100-nam and none
> work as expected.
I have a long-ago recollection that using vt100 had rendering issues with emacs, but vt102 was fine. Maybe worth a shot?
d
Somewhere between UNIX Release 3.0 and Release 4.1, a portion of the User's Manual was split off into a separate Administrator's Manual, leading to a reordering of the sections among other things. In the directories, these pieces would be placed in u_man and a_man respectively.
There may be some evidence of the manual being intact as of 4.0 or at least not completely separated. I've found consistently that references to manpages in the Documents for UNIX Release 4.0 collection follow their pre-split numbering and all refer to the User's Manual. The catch is that all references are to the UNIX User's Manual Release 3.0, so this may not point conclusively to the state of /usr/man on disk at the time. The Release 4.1 Administrator's Manual hasn't turned up yet but the User's Manual reflects the renumbering and is less the a_man pages. To complete the circle, the various Release 5.0 revisions of the documents do refer to the Administrator's Manual where appropriate.
Was the manual getting split up of any great shock or was it to be expected as the software grew? It would come to happen again between SysV and SVR2 with p_man. Out of curiosity I checked how my own manpage set is organized, it seems to be of the research order, with special files in section 4 rather than section 7 for instance. I've never studied how far reaching the different orders are.
- Matt G.
In a BSTJ article[1] it is said "The availability of a simulated UNIX operating system in DMERT allows UNIX programs from other processors to execute on the 3B20D Processor." Does this just mean C programs which are rebuilt or is there some implication DMERT's particular UNIX environment featured some sort of emulation facilities? I may be reading too much into it...
- Matt G.
[1] - https://archive.org/details/bstj62-1-303/page/n11/mode/2up
P.S. I learned I may walk past DMERT and a 3B20D most days, there's a long-operational telephone CO on my usual walk that through referencing public records I've discovered has a WECo 5ESS up in it somewhere. That's all the listing said, dunno if WECo is given as meaning an early model or just a generic name. Either way...so close yet so far, makes me ever so curious what dusty old bookshelves in that building might hold.
> From: Ken Thompson
> someone rewired someones desk lamp. i dont know how that worked out.
Sometimes electrical 'jokes' don't pan out - in a big way.
I was hacking the light switch in Jerry Saltzer's office (I don't recall
exactly what I was planning; IIRC, something mundane and lame like flipping
it upside down), and as I took it out of the box, the hot terminal touched
the side of the box (which was, properly, well grounded).
The entire 5th floor powered down.
What had happened was that the breaker for Jerry's office probably hadn't
been tripped in decades (maybe since it was put in), and it was apparently a
little sticky. Also, the floor had originally been wired back when all that
most people had in their offices, in the way of electrical load, was an
incandescent desk lamp or so. Now, most offices had, not just a couple of
terminals, most also had an Alto - greatly increased overall load. The total
draw for the whole floor was now very close to the rating of the main breaker
for the whole floor - and my slip of the hand had put it over. And that one
_wasn't_ sticky.
The worst part was that when we looked in the 5th floor electrical closet, we
couldn't find anything wrong. An electrician was summoned (luckily, or
unluckily, it was daytime; not having access to a 5th floor master, we'd gone
in while everything was unlocked - daytime), and he finally located the
breaker responsible - in an electrical closet on the 9th floor.
I got carpeted by Jerry, when he got back; I escaped without major
punishment,in part, IIRC, because I pointed out that I'd exposed a
previously-unsuspected issue. (I have this vague memory that the wiring on
the 5th floor was upgraded not long after.)
That wasn't the only historic CS building that has been abandoned. 545
Technology Square, one-time home of the Multics project, the MIT AI Lab,
and much else (including the above story) was exited by MIT some years
ago.
There, too, some history was abandoned - including the hack that allowed
people to call the elevators to their floor from their terminals. (Some
hackers had run some carefully disguised wires up into the elevator
controller - ran them along the back of structural members, carefully hidden
- and thence to the TV-11 that ran all the Knight TV bit-mapped displays
attached to the AI ITS time-sharing machine. So from a Knight TV console, if
you typed 'Escape E', it called the elevator to your floor - the code:
https://github.com/PDP-10/its/blob/master/src/system/tv.147
even has a table - at ELETAB: - giving which floor each console was on, so it
got called to the correct floor. I wonder what happened to that when the
Knight TV system was ditched? Did it get moved to another machine? Actually,
I have a dim memory that the elevator people found it, and it was removed.)
Noel
