Work at the Mill: The story of Digital Equipment Corporation | Hacker News Reader

Work at the Mill: The story of Digital Equipment Corporation

abortretry.fail

170 pointsrbanffy95 comments

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.

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.)

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.

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 :-)

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.

Basically all my cool college professors worked at DEC. Their impact in the Greater Boston area is huge and lingers. Don't forget Wang!

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.

https://www.rcsri.org/collection/

A walk-through I filmed of the warehouse: https://www.youtube.com/watch?v=LL8QBvXvGjQ

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.

The rent those computers our for movies.

That's unfortunate. Movie people are notorious for destroying rented props. I know several (not just 1 or 2) people who rented rare to unique computers and computer parts (because who doesn't want to see your computer in a film), had them returned destroyed (painted, parts stripped off/glued on, burned or water damaged or obviously dropped from a good height), and were told 'collect the insurance...that's what it's for'. Not a single fuck given whether or not they even could be replaced.

I think is a big part of their budget and a lot of the old machines in their warehouse aren't currently running. Mainframe type stuff and comodore 64s were mentioned.

But it must suck to have something valuable destroyed, especially if it's still functioning. I will heed this warning.

It was a while ago but I seemed to remember ber them saying the machines came back cleaner than before.

As someone who was gifted and old MicroVax (a replaced by PCs, old cad/cam machine), many of those old machines are fun to have around but aren't super useful for day to day computing. Thank goodness for emulation.

Yeah, you're probably right; someone running it all as a business has likely figured out how to manage risk. So to be fair, my experience here is with individuals who rented out their toys. I don't think any of the (too) many classic computer forums I'm part of don't have horror stories about the film industry. And if there's say, one Sage console or one 360/195 front panel (soooo many blinkenlights) in the world, it doesn't matter if it's working or not if it's jacked up.

P.S.: The use of per production LLCs shut down asap to avoid all sorts of annoying 'rules' and 'regulations' is another fun topic.

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.

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.

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.

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.

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.

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.

As the x86 ISA teaches us, mistakes are forever.

x86 has quite a few features disabled. I'm pretty sure there are CPUs that don't implement BOUND and several that turned off SGX after shipping with it.

BCD instructions disappeared along the way.

> many junior developers love mode switches

Today we call them “feature flags”. Definitely an anti-pattern.

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?

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.

This is legit cool. Makes me want to fire up the 3d printer.

If you ignore the white portion, it almost looks like a laptop.

I miss when hardware had designs this radical

tbh if y'all sat in front of it I don't think you'd find it that compelling in reality

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.

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.

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".

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.

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.

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.

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.

> "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.

> 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.

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.

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.

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...

> 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.

https://users.cs.utah.edu/~elb/folklore/mel.html

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.

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.

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...

[2]: https://en.wikipedia.org/wiki/Vacuum_tube

[3]: https://www.skf.com/group

> 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.

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.

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.

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.

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.

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.

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.

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.

I think the unit was more or less the shape/size of a VT-240. Is that right?

It's heavier as noted, but yeah that's a pretty good description.

I was a kid when I saw one, in Brazil, at a computer convention. It was, probably "Feira Internacional de Informática" in Brazil, in the early 1980s. The DEC team was incredibly friendly, even to two kids (I went there with a cousin) who would, maybe, be their customers, 20 years later.

I became one. My cousin followed another career.

> 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.

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)

Nice shot of nostalgia seeing pictures of the mill in Maynard. I taught C++ classes there back in the early 90's. It was fairly early in my professional career and just being in that building gave me a feeling of having arrived in the place I had been trying to get to for so long.

One of the things I'm most proud of as a coder, is writing a "realtime" raytracer (could only do balls) on a cluster of DEC Alphas - what machines. Also of course, learned the Internet on a VAX with VMS (that orange glow!) around 1990. I will always have a soft spot for DEC.

I loved visiting the Mill back in the day when I was part of the crews buying DECSystem-20's for Columbia University and later the Fairchild AI Lab.

What a great place to work, and hanging out with the OS developers like Dan Murphy was just icing on the cake.

It's a shame this company didn't survive.

My college was a DEC shop in Massachusetts, and I learned how to program in labs full of dumb X terminals with creative names connected to Alpha servers running Digital Unix. There were still a few Ultrix workstations around. Never used VMS aside from the class registration system; no VAX for us students.

Kind of miss those days.

Great article. My father was a software engineer at DEC in the mid-80s through the 90s and we often had DEC equipment at home. I grew up playing FLIGHT on a MicroVAX, learning BASIC programming on a Rainbow, and creating crazy answering machine greetings on a DECtalk.

DEC is a cool name. Digital Equipment, serious stuff. Sadly the era of three punchy letter names is over. What have we got now?

“Apple” (random word, even worse, it is a type of plant or something), “Facebook” (like a type of yearbook?) “Google” (sounds like they make candy or something), or “Intel” (who knows? -tel is just what companies ended their names with in the 80’s and 90’s).

DEC. What do they make? Equipment. What kind? Digital.

IBM would be good if the middle word wasn’t “business.” Business? Eh. Lame. The business is an aside, it is here to justify the machines. E, because Equipment is solid and physical.

ARM? Looks good at first. But then you find out… the A stands for Acorn. What the hell are we doing here, gardening? No. Terrible. We want D, for Digital, because we’re in the figure, the era of ones and zeros, not acorns.

AMD? Actually that name is OK as well.

But who doesn't make digital equipment these days? Even tractors and other farming equipment is chock full of digital equipment. And what kind of device isn't or has some sort of advanced micro device in it? Those companies might as just well merge into one megacorp named Stuff, for all it really means, if we're going off of company names being useful.

something something computer science; naming things.

You wouldn't like my old employer[0], then.

[0] https://softwareag.com

oh man, that's awesome!

Although Acorn designed the ARM chip in 1983 as the Acorn RISC Machine, in 1990 it was changed to Advanced RISC Machine.

https://en.wikipedia.org/wiki/Arm_Holdings#Name

Bah, they can change the official name, but we know the truth.

Anyway, they don’t sell machines. They sell designs.

They can’t do that, no matter how much corporations try.

they did it to RAID and the Inexpensive did completely disappear

What’s it been replaced with?

redundant array of independent disks

and when you look at the prices of them (the setups for hotswapping) you'll go with that

I quite like that Acorn still sort-of has that reference living on in the modern world.

Acorn is a local company for me - a load of my colleagues / contacts through the local tech scene worked there.

It's also a company whose products I originally saw in school when I was about 8 years old - they were the fancy computers, the ones kids like me weren't allowed to mess with.

Are you suggesting DRM would be a better name than ARM?

ART, for Advanced RISC Technology ;-)

We all knew RISC technology was going to change everything.

Acorn took that acronym as well with an Acorn RISC Technologies division near the end: https://chrisacorns.computinghistory.org.uk/docs/Acorn/ART/A...

Maybe DRE (Digital RISC Equipment)? Of course, if you're using the whole ARM a DRE might get uncomfortable.

Eh, they don’t even sell machines anyway? They sell chip designs. Maybe they should be called something like “RISC Intellectual Property.” Then they can start selling a flagship chip, call the One. The ad copy writes itself practically.

Announce your presence: RIP One.

Chicago Bridge and Iron. It's not in Chicago, doesn't make bridges, and doesn't use iron.

Ok, you got me to look: https://en.wikipedia.org/wiki/Chicago_Bridge_%26_Iron_Compan...

Indeed, DEC was a cool name. I also loved the simple yet recognizable digital logo.

Intel - I think tel would invoke tele like telephone or television so maybe a phone company or television studio. In an alternate world: "Get on the air with Intel - the intelligent television company!"

IBM - Incredibly Bitchin Machines

Ignoring the three letter bit, Intel ought to be a pretty good machine by your measures: Integrated electronics. ARM reformed itself, and is Advanced RISC Machine. I would argue that DEC and ARM are much more punchy than IBM and AMD. At least they are pronounced as single syllables, while the latter are read/spoken as TLAs. If your going to go to that much effort to pronounce it, you may as well suggest that Microsoft is a good name.

But if you really want a punchy name, something that will DEC them with the thing at the end of your ARM, well, you may as well go with a four letter acronym: FIST. I have yet to figure out what the F should stand for, but the rest is clearly Integrated Semiconductor Technologies.

Fully Integrated Semiconductor Technologies?

That is a good point about pronouncing the name as a single word instead of three letters. Although, I still think DEC is better than ARM. Ending on the hard C adds to the oomph.

Fairchild Integrated Semiconductor Technologies

https://en.m.wikipedia.org/wiki/Fairchild_Semiconductor

What do you think of IDR, for Intergalactic Digital Research (which later dropped the 'I') ? https://en.wikipedia.org/wiki/Digital_Research

TMC (Thinking Machines Corporation) was ahead of its time. https://en.wikipedia.org/wiki/Thinking_Machines_Corporation

A plausible theory is that "IBM" has been coined in such a way so that each letter was superior to the corresponding letter from "NCR", its main competitor in the past.

National => International

Cash => Business

So each word in IBM was more generic and more widely encompassing than its correspondent in NCR.

HAL would like a word…

Silicon Graphics was a great company name. With a great (initial) logo.

https://en.wikipedia.org/wiki/Silicon_Graphics#/media/File:S...

Interesting, the Mill has a number of various startups as tenants now. I think Monster used to have it's main office there.

There was a brew pub in the building on the South side of the mill pond for a few years that I used to go to quite a bit, and when my son was a toddler and we were looking for a daycare place, there was one in the basement of one of the Mill buildings. But it felt weird and closed in so we opted for somewhere else. The closest I get these days is the Indian restaurant across the street.

Yes it’s been a startup office space for the last quarter century. I did a few startups in various spaces there. And Monster was HQd there in their heyday as well. There were many artifacts of DEC still around the campus.

I visited the Mill in the mid-1980s. There was still a smell of lanolin in the stairwells.

My first programming job was to write Fortran using a PC that fed into a VAX VMS. The stuff I wrote was sold to the USAF. It feels so long ago, there were no packages to import. All I had was whatever Fortran 77 had and I was ever mindful of running out of memory. I remember writing my programs on loose leaf paper by hand and then laboriously keying it in. I don’t remember there was any compilation; it ran whenever the VMS decided it ran. I would read the log file I filled with stupid error messages to myself because I was bored. One error code would be DUCK. Another would be DUCKDUCK. The third would be GO. My boss would read and shake his head wryly.

Reply on news.ycombinator.com