16 Jan
2019
16 Jan
'19
3:49 a.m.
I've been on a Data General Aviion restoration binge lately and
re-familiarizing myself with DG/UX. In my case 5.4R3.1 running on a
MC88100 based AV/300 and MC88110 dual core AV/5500. The more I
experience, the more I am impressed. There are a few things about the
system that seem impressive.
- Despite coming from a System V core, there is a lot of BSD influx -
especially on the networking side. This is a personal taste issue as
other ports have tried to mix the best of both worlds. But after a
prior month-long Sun/Solaris restoration binge of similar era hardware
(Super/Hyper/Ultra SPARC) and software (SunOS 4 through Solaris 9),
DG/UX is a welcome and refreshing change! Especially out of the box.
- It has a system of file security that seems unique for that era - at
least in my experience - of explicit and implicit directory tags with
inheritance. There is even a high security extended version of the OS.
- It has a built-in logical volume manager supporting multiple virtual
to physical disk mappings, striping, mirroring, and even archiving -
something several entire sub-industries were created for in other ports.
I am guessing this contributed to EMC's purchase of Data General for
the Clariion disk storage product lines.
- It leveraged open-source tools early. The default m88k compiler
installed with the system is GNU C 2.xx.
- It was among the earliest of operating systems to support NUMA aware
affinity on MP versions of the MC88110. (IRIX, Solaris, BSD, Linux, and
Windows support all came much later).
- Many others.
It does have it's quirks. However I get the overall impression the
folks working at DG were on their game and were a leader in the industry
in many areas. It is unfortunate a) the fate of the Motorola 88K was
tied to Data General's place in the UNIX world, and b) by the time they
migrated to IA86, enterprise business was more interested in Microsoft
NT & SQL server or Linux than an expensive vendor's UNIX port.
That being said, I don't see DG/UX mentioned much in UNIX history. In
fact, I am researching an exhibit I'm putting together for the Vintage
Computer Festival Southeast 7.0, and DG/UX isn't mentioned on any of the
'UNIX Family Tree' diagrams I can find so far. It doesn't even make
Wikipedia's 'UNIX Variants' page. It's own Wikipedia page is also
rather sparse. Like John Snow in season 1, there is a junk of missing
and plot impacting history here - centered around the people involved.
To a lesser degree, IRIX is also a red-headed step-child. It's omitted
from half the lists I can find. It just seems the importance, even if
it's an importance by being the 'first' rather than # of users, of these
ports are pretty significant.
Just curious of others' thoughts. And I wondering if anyone has
first-hand knowledge of Data General's efforts or knows of others that
can illuminate the shadows of what I'm discovering is a pretty exciting
corner of the UNIX world.
Thanks,
-Alan H.
16 Jan
16 Jan
4:07 a.m.
In my opinion, the popularity of a UNIX platform is tightly tied to
the availability of the platform at university.
If DG was available to tinker on, to run ROFF, to write small programs
for other reasons, to introspect and familiarise yourself with, Then
for those students, it became the logical choice.
If they ignored the tertiary education market, sold into industry,
they could have established a huge loyal fanbase in E.G. Finance and
Insurance. Or in the decision support systems in Oil. Or shoe makers
inventory control. But if you don't have a cohort of students who
recognize your brand, your flavour of UNIX, and you then face these
students flexing muscles at purchase time, Instead of "lets buy the
upgrade option from DG" you get "why don't we buy Sun, and then get
cheap kids to run it"
TL;DR DG did not have significant presence in the tertiary education
systems I played in (York, Leeds, UCL, UQ) and by my visibility,
almost any tertiary education facility I have seen. They didn't feed
the beast.
4:47 a.m.
On Tue, 15 Jan 2019 at 23:08, George Michaelson <ggm(a)algebras.org> wrote:
In my opinion, the popularity of a UNIX platform is
tightly tied to
the availability of the platform at university.
That's a very good point. But going too far in that direction may have
been a problem too. My understanding is that Omron's Luna 88K line was
very closely tied to the education market. It ran a customized version of
Mach, so in some sense I suppose they were tied to CMU from the get-go, and
my understanding is that they courted the education market heavily.
Oberlin College was given, outright, a four processor 88K Luna. Today I'm
not sure you could find a running Luna if you wanted to.
-Henry
6:05 a.m.
On Tue, Jan 15, 2019, 9:48 PM Henry Bent <henry.r.bent(a)gmail.com wrote:
On Tue, 15 Jan 2019 at 23:08, George Michaelson <ggm(a)algebras.org> wrote:
https://dmesgd.nycbug.org/index.cgi?do=view&id=4501
Claims to be running OpenBSD as of a few months ago. There are also reports
of NetBSD as well. There appear to be maybe 6 different machines listed
here.
Not sure this really disproves your point though.
Warner
In my opinion, the popularity of a UNIX platform
is tightly tied to
the availability of the platform at university.
That's a very good point. But going too far in that direction may have
been a problem too. My understanding is that Omron's Luna 88K line was
very closely tied to the education market. It ran a customized version of
Mach, so in some sense I suppose they were tied to CMU from the get-go, and
my understanding is that they courted the education market heavily.
Oberlin College was given, outright, a four processor 88K Luna. Today I'm
not sure you could find a running Luna if you wanted to.
2:24 p.m.
On Tue, Jan 15, 2019 at 11:08 PM George Michaelson <ggm(a)algebras.org> wrote:
In my opinion, the popularity of a UNIX platform is
tightly tied to
the availability of the platform at university.
If DG was available to tinker on, to run ROFF, to write small programs
for other reasons, to introspect and familiarise yourself with, Then
for those students, it became the logical choice.
If they ignored the tertiary education market, sold into industry,
they could have established a huge loyal fanbase in E.G. Finance and
Insurance. Or in the decision support systems in Oil. Or shoe makers
inventory control. But if you don't have a cohort of students who
recognize your brand, your flavour of UNIX, and you then face these
students flexing muscles at purchase time, Instead of "lets buy the
upgrade option from DG" you get "why don't we buy Sun, and then get
cheap kids to run it"
TL;DR DG did not have significant presence in the tertiary education
systems I played in (York, Leeds, UCL, UQ) and by my visibility,
almost any tertiary education facility I have seen. They didn't feed
the beast.
Interesting. When I was in high school in central Pennsylvania and begging,
borrowing (and yeah a little stealing) computer time from Penn State
systems, there was a CS professor who'd made his bones building something
called UREP: Unix RSCS Emulation Program. I can't remember the fellow's
name; something "Roberts" maybe. He was known for being somewhat acerbic
(he'd call students "stupid" in class, was known to be nasty on USENET,
etc) and he wasn't a healthy man. He died of a heart attack when I was in
my late teens. Anyway....
What's notable about that, to me, was that he wrote UREP for DG/UX and was
known to be fond of Data General machines. This let him talk to the
university's mainframe, which was run by the computer center, ran VM, and
was the major compute engine on campus at the time outside of specially
purchased machines supporting research. There was a Cray somewhere on
campus, for example, but that was purchased out of research funds and
wasn't generally accessible. It also let Unix machines participate on
BITNET, which was a big deal locally at the time (probably because of the
close association with mainframes). But this was well before the AViiON
series; probably around the time of the Eagle. So maybe just after the
"Soul of a New Machine" era in DG's history.
Anyway, the point is that they did have a footprint in the academic market.
I suspect their lack of success had more to do with them as a company and
their foibles in the market than anything else. Like many of the "Route
128" minicomputer companies of the early 70s, I get the impression that
they ran themselves into the ground chasing the minicomputer market and
missing the shift to microprocessors, workstations and PCs. By the time
they could try and turn things around with the storage kit, they were a bit
player in the server market. The storage thing only set them apart and kept
them afloat long enough to get bought out.
- Dan C.
(PS: I worked for a startup in NYC in the very late 1990s and early 2000;
one of those "dot com" companies [all the stories are true, though in my
defense I had no idea just how much drama was happening around me at the
time]. We picked up some kind of engineering director guy via some merger
with another dot com startup-y sort of thing based in Boston and that guy
had come from Data General. Of course, he wanted to move everything to
Boston/Cambridge and thought us New Yorkers were a bunch of dullards. It
stuck out to me because I don't think I've ever worked with an emptier
suit, though I've seen a few that gave him a run for his money.... If DG
management was anything like him, no wonder they died an inglorious death.)
2:40 p.m.
On 1/16/2019 6:24 AM, Dan Cross wrote:
On Tue, Jan 15, 2019 at 11:08 PM George Michaelson
<ggm(a)algebras.org
<mailto:ggm@algebras.org>> wrote:
In my opinion, the popularity of a UNIX platform is tightly tied to
the availability of the platform at university.
I was in the CS Department at UC Berkeley for most of the 90s.
Looking at the presence of various Unix workstations was like
looking at the rings of a large tree. At various times, various
Unix workstation vendors would donate hardware in order to help
promote their version of Unix and/or their hardware. There were
the Sun years, the HP years, the DEC years, and the Intel-based PC
years.
Jon Forrest
2:40 p.m.
I’ve heard and personally seen a lot of technical arrogance and
incompetence out of the Masshole area. Was DEC inflicted? In
“Showstopper” Cutler fled to the west coast to get away from this kind of
thing.
On Wed, Jan 16, 2019 at 2:26 PM Dan Cross <crossd(a)gmail.com> wrote:
On Tue, Jan 15, 2019 at 11:08 PM George Michaelson
<ggm(a)algebras.org>
wrote:
In my opinion, the popularity of a UNIX platform
is tightly tied to
the availability of the platform at university.
If DG was available to tinker on, to run ROFF, to write small programs
for other reasons, to introspect and familiarise yourself with, Then
for those students, it became the logical choice.
If they ignored the tertiary education market, sold into industry,
they could have established a huge loyal fanbase in E.G. Finance and
Insurance. Or in the decision support systems in Oil. Or shoe makers
inventory control. But if you don't have a cohort of students who
recognize your brand, your flavour of UNIX, and you then face these
students flexing muscles at purchase time, Instead of "lets buy the
upgrade option from DG" you get "why don't we buy Sun, and then get
cheap kids to run it"
TL;DR DG did not have significant presence in the tertiary education
systems I played in (York, Leeds, UCL, UQ) and by my visibility,
almost any tertiary education facility I have seen. They didn't feed
the beast.
Interesting. When I was in high school in central Pennsylvania and
begging, borrowing (and yeah a little stealing) computer time from Penn
State systems, there was a CS professor who'd made his bones building
something called UREP: Unix RSCS Emulation Program. I can't remember the
fellow's name; something "Roberts" maybe. He was known for being somewhat
acerbic (he'd call students "stupid" in class, was known to be nasty on
USENET, etc) and he wasn't a healthy man. He died of a heart attack when I
was in my late teens. Anyway....
What's notable about that, to me, was that he wrote UREP for DG/UX and was
known to be fond of Data General machines. This let him talk to the
university's mainframe, which was run by the computer center, ran VM, and
was the major compute engine on campus at the time outside of specially
purchased machines supporting research. There was a Cray somewhere on
campus, for example, but that was purchased out of research funds and
wasn't generally accessible. It also let Unix machines participate on
BITNET, which was a big deal locally at the time (probably because of the
close association with mainframes). But this was well before the AViiON
series; probably around the time of the Eagle. So maybe just after the
"Soul of a New Machine" era in DG's history.
Anyway, the point is that they did have a footprint in the academic
market. I suspect their lack of success had more to do with them as a
company and their foibles in the market than anything else. Like many of
the "Route 128" minicomputer companies of the early 70s, I get the
impression that they ran themselves into the ground chasing the
minicomputer market and missing the shift to microprocessors, workstations
and PCs. By the time they could try and turn things around with the storage
kit, they were a bit player in the server market. The storage thing only
set them apart and kept them afloat long enough to get bought out.
- Dan C.
(PS: I worked for a startup in NYC in the very late 1990s and early 2000;
one of those "dot com" companies [all the stories are true, though in my
defense I had no idea just how much drama was happening around me at the
time]. We picked up some kind of engineering director guy via some merger
with another dot com startup-y sort of thing based in Boston and that guy
had come from Data General. Of course, he wanted to move everything to
Boston/Cambridge and thought us New Yorkers were a bunch of dullards. It
stuck out to me because I don't think I've ever worked with an emptier
suit, though I've seen a few that gave him a run for his money.... If DG
management was anything like him, no wonder they died an inglorious death.)
2:51 p.m.
On 1/16/19 9:24 AM, Dan Cross wrote:
Anyway, the point is that they did have a footprint in
the academic market.
I suspect their lack of success had more to do with them as a company and
their foibles in the market than anything else.
We (CWRU) got a donated MV10000 from DG in the mid-80s. It was pretty
easy to get an account on it, as opposed to the "official" VAX, but
everyone hated working on it. It was just painful, especially if you
tried to get free software working.
DG tried, but, at least in our case, they just weren't up to it.
--
``The lyf so short, the craft so long to lerne.'' - Chaucer
``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, UTech, CWRU chet(a)case.edu http://tiswww.cwru.edu/~chet/
2:58 p.m.
From what I've heard, "Technical
arrogance" and "Cutler" in the same
paragraph is redundant. :-)
I work with a bunch of former Microsofties, most of whom were in the
Windows kernel group. None of them have nice things to say about him. They
acknowledged that Gates could be an insensitive jerk, but at least there
was a method behind his madness (he'd be trying to get the the bottom of
something he wanted to understand and he wanted to be sure other people
understood it too, for example). Cutler on the other hand was just cruel.
One guy said his mother was visiting and was standing in the lobby of a
Microsoft building while Cutler berated someone for some minor fault. "She
learned new words and phrases she hadn't known could exist."
Ironically, despite being totally unwilling to move to Boston in 2000, I
find myself in exile here in the outer provinces of Massachusetts now. It's
kind of amazing to see how many survivors of the minicomputer heydays seem
to be waiting for DEC, Data General, etc, to rise from the ashes and take
over the datacenter again. "Lisp Machines will rise again!"
- Dan C.
On Wed, Jan 16, 2019 at 9:40 AM Kevin Bowling <kevin.bowling(a)kev009.com>
wrote:
I’ve heard and personally seen a lot of technical
arrogance and
incompetence out of the Masshole area. Was DEC inflicted? In
“Showstopper” Cutler fled to the west coast to get away from this kind of
thing.
On Wed, Jan 16, 2019 at 2:26 PM Dan Cross <crossd(a)gmail.com> wrote:
On Tue, Jan 15, 2019 at 11:08 PM George
Michaelson <ggm(a)algebras.org>
wrote:
In my opinion, the popularity of a UNIX platform
is tightly tied to
the availability of the platform at university.
If DG was available to tinker on, to run ROFF, to write small programs
for other reasons, to introspect and familiarise yourself with, Then
for those students, it became the logical choice.
If they ignored the tertiary education market, sold into industry,
they could have established a huge loyal fanbase in E.G. Finance and
Insurance. Or in the decision support systems in Oil. Or shoe makers
inventory control. But if you don't have a cohort of students who
recognize your brand, your flavour of UNIX, and you then face these
students flexing muscles at purchase time, Instead of "lets buy the
upgrade option from DG" you get "why don't we buy Sun, and then get
cheap kids to run it"
TL;DR DG did not have significant presence in the tertiary education
systems I played in (York, Leeds, UCL, UQ) and by my visibility,
almost any tertiary education facility I have seen. They didn't feed
the beast.
Interesting. When I was in high school in central Pennsylvania and
begging, borrowing (and yeah a little stealing) computer time from Penn
State systems, there was a CS professor who'd made his bones building
something called UREP: Unix RSCS Emulation Program. I can't remember the
fellow's name; something "Roberts" maybe. He was known for being somewhat
acerbic (he'd call students "stupid" in class, was known to be nasty on
USENET, etc) and he wasn't a healthy man. He died of a heart attack when I
was in my late teens. Anyway....
What's notable about that, to me, was that he wrote UREP for DG/UX and
was known to be fond of Data General machines. This let him talk to the
university's mainframe, which was run by the computer center, ran VM, and
was the major compute engine on campus at the time outside of specially
purchased machines supporting research. There was a Cray somewhere on
campus, for example, but that was purchased out of research funds and
wasn't generally accessible. It also let Unix machines participate on
BITNET, which was a big deal locally at the time (probably because of the
close association with mainframes). But this was well before the AViiON
series; probably around the time of the Eagle. So maybe just after the
"Soul of a New Machine" era in DG's history.
Anyway, the point is that they did have a footprint in the academic
market. I suspect their lack of success had more to do with them as a
company and their foibles in the market than anything else. Like many of
the "Route 128" minicomputer companies of the early 70s, I get the
impression that they ran themselves into the ground chasing the
minicomputer market and missing the shift to microprocessors, workstations
and PCs. By the time they could try and turn things around with the storage
kit, they were a bit player in the server market. The storage thing only
set them apart and kept them afloat long enough to get bought out.
- Dan C.
(PS: I worked for a startup in NYC in the very late 1990s and early 2000;
one of those "dot com" companies [all the stories are true, though in my
defense I had no idea just how much drama was happening around me at the
time]. We picked up some kind of engineering director guy via some merger
with another dot com startup-y sort of thing based in Boston and that guy
had come from Data General. Of course, he wanted to move everything to
Boston/Cambridge and thought us New Yorkers were a bunch of dullards. It
stuck out to me because I don't think I've ever worked with an emptier
suit, though I've seen a few that gave him a run for his money.... If DG
management was anything like him, no wonder they died an inglorious death.)
3:05 p.m.
Dan Cross <crossd(a)gmail.com> writes:
I can't remember the fellow's name;
something "Roberts" maybe. [...]
What's notable about that, to me, was that he wrote UREP for DG/UX and
was known to be fond of Data General machines.
Robert Michael Owens -- in the utzoo Usenet archives there are very
early posts from him as psuvax!bear. He passed away in 1997; see the
biography here: http://sips2016.rice.edu/bob-owens/
A 1983 post to net.unix-wizards from Bruce Crabill says:
The Unix version of the RSCS emulator is called UREP
(Unix RSCS
Emulation Program) and is available from Pennsylvania State
University. It was written by Robert Owens and Joseph Boykin. More
information on it can be obtained from Robert Owens (BITNET:
OWENS@PSUVAX1). The VMS version was written by Craig Watkins (BITNET:
CRW@PSUVMS1). I have talked to both versions from BITNET and found
them to be very good emulations of RSCS (which is IBM's Remote
Spooling Communications Subsystem/Networking).
--
Adam Sampson <ats(a)offog.org> <http://offog.org/>
3:44 p.m.
Let me see if I can illuminate from a little I was aware. I know some of
the histories from my time at Locus when DG was a customer of ours.
Although, I'm going to show my old f*rt, cultural illiteracy (not a TV
person) by having no idea who 'John Snow' is, so I can not comment on that
reference ;-)
I think you are right that some of the histories tend to tided to people's
experience at their college or universities and I'm not sure how well DG
sold in those markets. Plus they bet on Moto, who fumbled with the 68000
replacement (88k), compared to IBM's PPC and later Intel's remarkable
recovery with the 386. Since they joined the losing side of the chip war,
I suspect that also hurt them, even though the system was actually well
done.
At one time I had access to the sources of DG/UX and yes you are correct it
was very clean and easy to understand (and fairly well documented). At
the time, we (LCC) had the sources to DG/UX, Ultrix, Tru64, OSF/1, SunOS,
Solaris, HP-UX, AIX, Prime, as well as all of the AT&T releases. (The
management at Locus used to say we were the Swiss of the UNIX industry -
we sold arms to all the sides of the war).
As for DG/UX, their kernel seemed like it was a rewrite (I never knew how
much of it was the SW folks in NC (who had been building their failed
"Fountainhead" system that the Soul-Of-The-New-Machine book talks about,
and how much was the MA folks that did Clarion later on). It was
definitely the sanest of all of the UNIX kernels we had access and had the
least amount of cruft in it of the commercial systems. The locking scheme
was the one of the cleanest, I ever saw (Stellar's Stellix was the only one
that was as good, IMO). The memory system was really impressive. I
remember when we were doing the TNC distributed FS work that what would
become the TruCluster FS for DEC at the same time. The DG/UX version was
the simplest (and the HP/UX version the most twisted confused).
I'm now mixing up the differences in my mind, but I think I remember DG/UX
had some sort of file system stacking scheme at the inode level. I'm not
sure if it ever shipped, but we worked on a Union file system for them;
that I remember was very cool (Plan9ish in splicing file system namespaces
together). I seem to remember it was all made possible because of the
lower level memory scheme. But I might be mixing that up with one of the
systems we were hacking (it was definately not OSF/1 or its child Tru64).
As you said, the user space API was basically System V, but DG did support
a lot of BSDism also; so bringing networking code from 4.2/4.3 was not
terrible; although since it was not fish or fowl, you had to be careful.
The big issue when it came out, is that SunOS (and later Solaris) had
become the defacto system at most universities, where much 'free software'
was being produced. Since it was more System V user API at the heart
(which was good for DG's targetted ISVs), it did suffer from the porting
issues. Since it was an 88K and SunOS had been 68K, the Endian issues of
the free SW was less of a problem, but not only was not a VAX, but it was
not BSD.
Clem
ᐧ
3:50 p.m.
On Wed, Jan 16, 2019 at 9:41 AM Kevin Bowling <kevin.bowling(a)kev009.com>
wrote:
In “Showstopper” Cutler fled to the west coast to get
away from this kind
of thing.
Be careful of popularized history. DC went to the left coast for reasons
that had nothing to do with MA culture. As people that knew him and worked
with him can attest, he was in fact as a chain smoker and ex-Marine (sorry
Dan), he was and is much more 'east-coast' than 'mellow left coaster.'
[He
kept his cigarrettes rolled up in his tee shirt just like James Dean BTW].
There is even a ESPN highlight real of him rolling his car in an NASCAR
race, which Ihave lost the URL too (I took a quick look and could not find
it).
ᐧ
17 Jan
17 Jan
6:53 a.m.
New subject: UREP - Unix RSCS Emulation Program
Dan Cross <crossd(a)gmail.com> wrote:
Interesting. When I was in high school in central
Pennsylvania and begging,
borrowing (and yeah a little stealing) computer time from Penn State
systems, there was a CS professor who'd made his bones building something
called UREP: Unix RSCS Emulation Program. ...
What's notable about that, to me, was that he wrote UREP for DG/UX and was
known to be fond of Data General machines. This let him talk to the
university's mainframe, which was run by the computer center, ran VM, and
was the major compute engine on campus at the time outside of specially
purchased machines supporting research. There was a Cray somewhere on
campus, for example, but that was purchased out of research funds and
wasn't generally accessible. It also let Unix machines participate on
BITNET, which was a big deal locally at the time (probably because of the
close association with mainframes). ....
In the mid-1980s I was a sys admin at the Emory U Computing Center and
we ran UREP on our vaxen in order to be able to send and receive BITNET
mail. It was kind of cute to watch your messages traveling the world,
as you got an interactive message back for each site it traversed on
its way to its final destination.
BUT, the code was miserable. I had to make some changes to it (don't
remember why), but EVERY time I had to dive into it, I hated it. I
used to say that I felt like I needed a shower afterwards.
Arnold
1930
days inactive
1931
days old
13 comments
12 participants
participants (12)
-
Adam Sampson -
alan＠alanlee.org -
arnold＠skeeve.com -
Chet Ramey -
Clem Cole -
Dan Cross -
George Michaelson -
Henry Bent -
Jon Forrest -
Kevin Bowling -
Lars Brinkhoff -
Warner Losh