Work at the Mill: The story of Digital Equipment Corporation
99 comments
·February 10, 2025ben7799
cbm-vic-20
If you're in New England, it's definitely worth a visit to the Rhode Island Computer Museum in Warwick, RI. They have a staggering amount of DEC stuff, including rare stuff like one of the nine remaining PDP-9 systems, a PDP-12, most models of the PDP-11 series, and a PDP "Straight" 8, the best looking computer ever made. And a bunch of later VAX machines, DECSystems, etc.
https://www.ricomputermuseum.org/collections-gallery/equipme...
The Retro-Computing Society of RI, a few miles away in Providence, also has a bunch of DEC stuff, also including a PDP-12. These might be the closest working PDP-12s in existence.
dmd
A walk-through I filmed of the warehouse: https://www.youtube.com/watch?v=LL8QBvXvGjQ
acomjean
Its a great museum. They had the ibm displaywriter with the 10" floppies we had in our house (my mom used it for technical translations).
https://sites.google.com/a/ricomputermuseum.org/ricm-learnin...
The warehouse (Fraiser archives?) is fun, but off site from the museum. My partner noted that driving over there seemed like a start to a bad movie....
I remember the guy giving the tour telling us that he started by collecting cars, but computers seemed more manageable...
The rent those computers our for movies.
ghaff
Beyond DEC, there were many minicomputer and related companies in Massachusetts in their heyday. They weren't all on Route 128 (including DEC) but that was the shorthand.
I VERY briefly worked for EMC after they acquired Data General which was a DEC spinoff.
lttlrck
The first time I saw the WWW was on an Apollo workstation running NCSA Mosaic.
This was at Portsmouth Uni in the UK. Those were all replaced with HP PA-RISC workstations after HP bought Apollo.
Apollo was founded in Chelmsford MA where I, coincidentally, bought my first house :-)
ben7799
I visited HP at Apollo drive in Chelmsford as an intern in college.
Then I later worked in the building when Cisco took over that building/park.
icedchai
DEC influence was definitely all over the north east! The first multi-user system I dialed into was a VAX/VMS box. I went to a college that had almost all DEC systems - mostly DECstations (MIPS) and Alphas. I later worked at software company full of tons of ex-Digital folk, running most of their software on Alphas running VMS and Tru64 (aka Digital Unix.)
2OEH8eoCRo0
Basically all my cool college professors worked at DEC. Their impact in the Greater Boston area is huge and lingers. Don't forget Wang!
acrophiliac
If you're interested in DEC history and haven't read "The Soul of a New Machine" by Tracy Kidder, you owe it to yourself to get a copy. A Pulitzer Prize winner and IMO one of the best tech books ever written. It is a captivating book that chronicles the creation of Data General's Eclipse MV/8000 computer from the engineers' point of view.
abraae
One insight from that great book I've always liked was when the CEO tells the team "no mode switch". His directive was about backwards compatibility with the old instruction set. With hindsight the team realise this was a useful constraint that led to a better product.
In my experience many junior developers love mode switches, because they aren't sufficiently scared of complexity. "Yeah, lets add a setting to turn on or off advanced mode. Then the power users and the noobs will all be happy!".
After some time in the trenches it becomes clear to most that mode switches lead to combinatorial complexity and should be avoided at all costs.
cafard
The old Nova instruction set, which lived on into the 16-bit and eventually 32-bit Eclipse instruction set, did have some curiosities: only four registers, and as I recall you couldn't do indirect byte addressing. The book does say that the architect on West's team had a much more VAX-like instruction set in mind for a project he didn't get to do.
That said, I was once at a customer's site that had an original "blue and white" MV/8000 (the MV series in general were brown-clad). If I remember correctly what the admin said, they boot up the new 32-bit OS, AOS/VS, for part of the day, and an older 16-bit OS, either RDOS or AOS, for the rest. I wish I remembered this more clearly.
And of course the VAX-style instruction set that West's architect envied fell out of style, didn't it?
AdamN
It's also why thinking critically about the initial 'mode' is important. The fewer mistakes that have to be embedded into the distant future, the better.
rbanffy
As the x86 ISA teaches us, mistakes are forever.
rbanffy
> many junior developers love mode switches
Today we call them “feature flags”. Definitely an anti-pattern.
ghaff
A huge number of DEC kernel engineers also ended up working on the Linux kernel as Red Hat employees in Westford, MA.
I dotted lined into Tom West for a time at Data General and, later, had a >decade stint at Red Hat where I got hired on by a senior ex-DEC person.
hnthrowaway0315
I feel that was a piece of history very much worthy talking about, plus it is very inspirational for younger engineers who want to be real engineers.
I for one look up to people who can design, implement and debug complex systems, especially computers components and low level computer software.
partomniscient
Steven Levy's Hackers: Heroes of the Computer Revolution is also worth a read, particularly the early part of the book which covers in detail stuff like the guys at MIT writing and playing Spacewar! on the PDP-1 in 1961. They even built their own joysticks.
neilv
The photo of the VT05 in the article doesn't do it justice:
http://www.columbia.edu/cu/computinghistory/vt05.html
https://terminals-wiki.org/wiki/index.php/File:DEC_VT05_1217...
This was before my time, but I imagine that the looks said the future is now, and you're piloting a spaceship as you code.
mark_undoio
I've love to have one of these to interact through! But every time I see these vintage TTYs I simultaneously think "that's so cool, I have to have it" and "I have no space for that on my desk".
DrillShopper
I miss when hardware had designs this radical
null
KerrAvon
tbh if y'all sat in front of it I don't think you'd find it that compelling in reality
neilv
Not as impressive from seated operator position. But I'd guess, back then, with often multiple people crowded around the too-few terminals, on top of all the unexplored potential of the fascinating new machine, and everyone craving hands on their keyboard, the extra design touches someone cared to make on the chassis are looking extra cool.
rbanffy
On these you’d hit the screen frame that extends to the bottom of the keyboard. Looks much better than it actually works.
OTOH, is love to have a 3278-2 on my desk. I’d even make an effort to make zOS into a daily driver.
pryelluw
This is legit cool. Makes me want to fire up the 3d printer.
layer8
If you ignore the white portion, it almost looks like a laptop.
jamesy0ung
It's interesting how we've come full circle. Some of the engineers who worked at DEC later worked at PA Semi and Apple after the acquisition, helping develop Apple Silicon. DEC's Alpha workstations running Unix were among the most powerful systems of their time, and now, decades later, Apple Silicon has brought us back to high-performance RISC Unix workstations — just with a different flavor of ARM instead of Alpha.
adrian_b
Also many of the engineers that have developed AMD Athlon (launched in 1999), the first AMD CPU that has exceeded the performance of the contemporaneous Intel CPUs, have come from the DEC Alpha designers.
rbanffy
I love to remind myself that Apple is the heir of the RISC Unix workstation market. Well, it’s actually NeXT, but this is a minor detail.
jksmith
Great thread. Love DEC/VAX history. But as soon as I saw a pic of Robert Palmer with his hair product and 2k suits in 1990, I knew then end was near. I mean yeah Ellison at Oracle was similar at the time, but this was DEC!, not just a database engine.
I bought a rack-mount Alpha back in the early 90's running Symbolics on it. Probably all I need to say about that.
russellbeattie
> "CompuServe had some PDP-10s operating until at least 2007"
This sentence sent me down an internet rabbit hole searching for some sort of citation, as it seems absurd on the face of it. I couldn't really find an original source, but lots of comments and references about how important PDP-10s were to CompuServe. They even used clones after the platform was discontinued. It seems CompuServe kept using and updating their minis until the end.
Can you imagine maintaining a PDP-10 in the mid 2000s? Surreal. I know it's non-trivial to port software off an old computer system, but this is a machine with a clock ticking in microseconds. I can't imagine it'd take more than a week or so to reimplement any sort of business logic it had.
abraae
> this is a machine with a clock ticking in microseconds. I can't imagine it'd take more than a week or so to reimplement any sort of business logic it had.
You're kidding right? In the old days people had to do insane things with their code to make it run fast enough. I'm thinking for example of programmers working with then-slow disk and drum storage who had to play elaborate timing tricks with their code based on when they expected data on the disk to rotate into position under the read heads. Just a few lines of assembler code, "ticking in microseconds", could conceal a terrifyingly fragile concoction of code. You could spend hours or days trying to understand it before you would dare to make a change.
Sure if you had an accurate, up to date, human-readable description of the system's "business logic" you'd be off to a flying start to migrate it to a new platform.
But the inalienable truth of old mainframe systems is that such a description no longer exists, if it ever did. The system itself - in COBOL, assembler or whatever - is the true description of the business logic. That's why IBM still sells mainframes compatible with their ancestors from over half a century ago. No-one dares to change many of these systems.
russellbeattie
OK. You're right, "weeks" was a severe understatement. But we're talking about 30 years and CompuServe wasn't a bank or nuclear power plant or anything. Extracting the business logic should have been a project that was done and gone by the 2000s.
Bluecobra
> I'm thinking for example of programmers working with then-slow disk and drum storage who had to play elaborate timing tricks with their code based on when they expected data on the disk to rotate into position under the read heads.
There was an epic story I once read that revolved around this and it’s going to drive me crazy until I find it.
Edit: Found it! It’s from “The Story of Mel, a Real Programmer”:
> Mel never wrote time-delay loops, either, even when the balky Flexowriter required a delay between output characters to work right. He just located instructions on the drum so each successive one was just past the read head when it was needed; the drum had to execute another complete revolution to find the next instruction.
euroderf
This makes me think of that algorithmic black box that the IRS has. It is spoken of in reverent tones tinged with dread. I wonder if anyone's tried to throw an AI at it yet to figure out what the heck is really going on in it? And maybe decompile it to render it amenable to modification and recompilation.
Bluecobra
It might be easier for DOGE to just close shop and implement a flat tax than try to untangle that ball of wall of wax...
kjs3
They weren't the giant DEC-made PDP-10s, AFAIK. They were System Concept clones of the PDP-10 (SC30, SC40). Relatively small, modern (for the time) semiconductors, SCSI I/O, Ethernet & FDDI network.
There were also PDP-10 clones called TOAD from a company called XKL that were sold into the 80s and maybe 90s.
Both were much, much faster than the original PDP-10s.
drewg123
I have a huge soft spot in my heart for DEC. They hold the same place in my heart that Sun does for a lot of people. One of my first *nix accounts was on MIPS DECstations running ULTRIX as an undergrad.
My first job was as a *nix sysadmin for a small academic department running DECstations with ULTRIX. This was in 1993, and I helped them transition to DEC Alphas. One of the first things I did was upgrade the RAM in a 3000/500X ("hot pink flamingo". The cost of the RAM kit was more than my annual salary, so I paid extra close attention to the EST precautions.
I moved on to being research staff in a department doing OS research, and was paid to help port FreeBSD to the DEC Alpha. One of my fondest memories is getting a UP1000 from API (Alpha Processor Inc, who made Alpha boards with AMD chipsets) to port FreeBSD to. (Alpha was a bit like older Arm, and each board was a bit different w/o a generic way to discover hardware).
And to cap it all off, I got to work with Dick Sites (one of the Alpha architects) at Google. When I first met him, he autographed my old copy of the DEC Alpha Architecture Reference manual. He was the nicest guy ever, and even offered to give me a ride to the airport one day when I had a really early flight. I love it when my heros are amazing people.
unwind
It's really hilarious that the founder of DEC married a girl from Finland whose last name was Valve [1][2].
Also, the linked article states:
Before enrolling in graduate school, Olsen took a summer off and traveled to Finland. His parents’ neighbor had had a Finnish girl, Eeva-Liisa Aulikki, who was a visiting student, stay with them. Olsen liked her and he failed in his initial pursuit. Yet, Olsen was a determined man. He got a job at a ball bearing factory in Goteberg.
This is just ... strangely done. The ball bearing factory was SKF's [3] in Gothenburg Sweden. Finnish girls come from Finland typically, so this part is just strange and glitchy.
[1]: https://hightechhistory.wordpress.com/2009/03/26/aulikki-ols...
allturtles
> This is just ... strangely done. The ball bearing factory was SKF's [3] in Gothenburg Sweden. Finnish girls come from Finland typically, so this part is just strange and glitchy.
Well it's a lot closer to visit Finland when you are already in Sweden than from the U.S. This was probably circa 1950 or earlier, so cheap jet travel didn't exist yet. Maybe they met up in Stockholm on weekends.
cafard
Yes, well, the article describes the Navy moving Olsen from Great Lakes Naval Training Station to Chicago. Great Lakes Naval Training Station is and, as far as I know, always has been in Chicago. The non-computer stuff maybe didn't get the same level of care.
null
rbanffy
The very Finish girl I met was in São Paulo on a yearly ball by the Hungarian culture association. Sadly, for me, she was into my older cousin.
choult
Nice to see a mention of Reading UK in there - for those interested, DEC's first UK office was upstairs at what is now a Brewdog bar.
These days, I can frequently be found in said space on Wednesdays, practicing my improv skills.
wglb
My experience does include some exposure to several of the projects. The first was at Mark Williams Company (MWC), where the first version of the Unix-like operating system Coherent was first done on a PDP-11/45. The first thing to be done was to build a C compiler from scratch and was accomplished by one of the 10x programmers that I knew, David G. Conroy (DGC). For a time, Intel used the MWC 8086 compiler repackaged as their own. The C compiler was also sold for MS-DOS and some really odd machines. One of those was for Rediffusion who wanted to a compiler for their bespoke architecture. It was simple enough that I wrote an emulator for the language as a basis to test the compiler.
Another computer mentioned was the Rainbow. DGC ported the compiler there, and in the process, wrote MicroEMACS, which is now known as MG on Linux and Unix. (Apparently this is the editor that Linus uses.)
We also had a VAX (I think it was a 730, very low-powered) to port the 8086 compiler to be hosted on VMS.
After MWC, DGC went to DEC and worked on DecTalk, an early text-to-speech device. The story goes that one member of the team, whose voice was used as the basis for the sounds, lost his voice. Thus his voice lived on only through this device.
When I was at Sycor in the late 70s, our computer was a PDP-10 (or Dec-10) running Tops 10. We wrote a cross compiler for 8085 in Bliss-36. (The assembly language for the PDP-10 was probably the most attractive I have come across.)
DGC also worked on the Alpha team (https://www.computerhistory.org/collections/catalog/X1720.99...) . He is now at Apple, and he has a side hobby of building Alpha boards out of FPGA (http://fpgaretrocomputing.org/).
When I was at Datalogics (1990-1992), who trained probably. half the world in SGML, and provided software for database publishing, legal loose leaf publishing (ask me about footnotes) and whose software was printing a substantial fraction of the red herring prospectuses at the time. They used a VAX running VMS. I had a Dec Station running some flavor of Unix. This was my first use of Emacs. As opposed to MicroEMACS.
kjs3
DecTalk, an early text-to-speech device. The story goes that one member of the team, whose voice was used as the basis for the sounds, lost his voice.
Dr. Dennis Klatt. He developed cancer (throat of some sort, I'd assume) and did lose his voice but died within a year, AFAIK.
Perhaps more memorable than the very notable DECTalk, he built original voice synthesizer for, and was the voice of, Stephen Hawkings.
rbanffy
DECtalk, IIRC, was much higher quality than Stephen Hawking’s voice synth. A real shame to hear this story about the owner of that voice.
kjs3
Well...a DECtalk also weighs about 15 lbs and is packed with components. So it was probably a weight/quality tradeoff considering it needed to be mounted on a wheelchair.
I understand Hawkings last version use a Raspberry PI, with the original voice.
musicale
> DEC VT100 at the Living Computer Museum, image by Jason Scott
LCM RIP.
I wish I'd managed to visit it before it shut down.
riedel
Interesting read. Our lab was founded inside a campus engineering center of digital. So we still have some memorabilia (like a 'cyber cafe' sign from digital) lying around. Actually digital research in Germany was sold to SAP, which didn't have research at the time (SAP actually pretty much scrapped their research division pretty much about 10 year since they seemingly never grew really fond of it)
I'm in my late 40s and have been a software engineer in New England since graduating college. We moved to Massachusetts in 1986, my father went to work at a startup company that was across the street from a DEC building and was in the DEC ecosystem.
The pervasiveness of DEC here is amazing even after all these years. My father in law also spent almost his entire career at DEC and retired as an HP employee. One of my mentors in my late 20s (older Gen X) started his career there. I've known many other people who worked at Digital.
My only real exposure was being like 10-12 years old and getting trucked to the office during "crunch time" and I got to sit at the console on a VAX at one point and got to play early text based games on it. This was weirdly right in the machine room at the operating console as there was some rework project in the room that my father had to be present for. Later on at my first internship in the mid 1990s I had a QA job and I had to test software and I remember having to install on both Ultrix and DEC Unix and I got to use some of the Alpha machines.
There are still all these buildings and artifacts all over the place from DEC all over MA. I bank with DCU. When I was in middle school my best friend and fellow computer nerd's father worked at another Massachusetts institution.. EMC.
I never got pushed into software but it seemed to be pre-ordained... we always had computers and I was given free rein to play with them starting in the early-mid-80s but not really allowed to have video games and we always had business type machines and not the more fun consumer oriented stuff.