It's interesting and heartwarming to see how similar the spirit of many successful software projects was. Creative collaboration, open play, extremely high trust, by people who really intrinsically love what they do.
It goes against so much of the MBA-worldview and bigcorp offices.
Unix, GNU, Linux, early Python, early Rockstar Games etc.
Yes, but that requires admitting that the employees are flesh and bones humans with human social relationships between them and this stuff can't be tabulated and accounted and bean-counted. And it's rather outright seen as a risk factor, makes people less fungible, there's a risk that they gang up against the management, or there could be legal risk if they build too much of an insider culture etc.
Highly aligned motivated and talented people with shared core values (like the pioneer ethos of hackers, tech optimism etc, and cultural references etc) can achieve so much. And they tend to work odd hours and overtime because they want to make the thing, and not because of hoping to get rich.
Now I have two things to say about this.
First, this type of passion is special and a manager can't hope to simply force it, and if it's demanded then all you get will be talkers who know the lingo of being so super passionate. The thing must be worth being passionate about and the team has to be aligned with the goal and have latitude to shape the project. This kind of work bes happens when the higher up kinda forget the team and they just intrinsically do it. It can't be planned.
Second, some people are very opposed to these hight effort, almost obsessed teams because they see it as unfair ideals or unfair competition, because obviously someone with a family who has to pick up the kids at 4 pm can't do this cracked push overnight and sleeping at the office etc. But greatness simy cannot be made from steady, fixed-pace 9-5 jobs with work life balance and atomized employees. And that's okay, for most things we don't need greatness just okayness. But often people still can't stand that there are such great teams and want to drag them down in one way or another.
> But often people still can't stand that there are such great teams and want to drag them down in one way or another.
This seems like an unnecessary attribution of malicious intentions. The obvious explanation for why people often oppose a culture of long hours is that long hours suck for anyone with a life outside of work. You explain this yourself with the example of someone who has kids.
Yes, but it shouldn't be mandatory for "regular" jobs. People often dislike the idea that there are some ambitious, agentic, talented, motivated people who really put in the hours and their job is simultaneously their hobby etc.
Again, I'm not for the cringe corporate pretending to be one big family smiling and singing. But clearly there are really strong, aligned teams that want to accomplish something and have the corresponding freedom to creatively shape the path forward. And such teams are just vastly more effective and can do stuff in a weekend that a traditional committee-based process would not get done in a year.
But again, this is a fraction of a percent of people who are built for that kind of work. But let some people just be weird and misfits! Somehow the same standards are not applied to athletes who train insane hours or musicians on tour.
My point is, let's be grateful to have such exceptional people instead of complaining about them or how they set an unrealistic standard or whatever.
If these exceptional people are making it necessary for me to work very long hours too, then it’s fair enough for me to complain about that, no? Very few of the people you are talking about are curing cancer (so to speak), so why are the rest of us required to be grateful to them?
Let’s look at it another way. Someone who’s willing to work an 80 hour week for the same pay that I get is roughly equivalent to someone who’ll do my job for half the pay (leaving aside the dubious productivity benefits of long hours). Should I be grateful for the existence of such a person? We do not usually romanticize people who are willing to do professional jobs for low compensation. Why romanticize people who work crazy hours? If I voluntarily took a 50% pay cut, would you wax lyrical about how the world needs more exceptional people like me?
> And such teams are just vastly more effective and can do stuff in a weekend that a traditional committee-based process would not get done in a year.
Here you’re conflating two different things. Small agile teams that have the freedom to work without bureaucratic overhead are great, but there is no inherent need for them to work crazy hours. If anything, long hours are often a symptom of an environment where people are judged on how long they stay in the office rather than on the quality of their work.
This is beypnd jobs and corporations and is about humanity and civilization. A small minority of people in every era actually pushed on any kind of frontier. That's ok. We need people who just execute the mundane tasks too, that's most of us.
But giving space to exceptional people is not even that expensive. The overtime doesn't even have to be documented. Just leave them alone. There aren't enough of them to truly outcompete regular people.
For most people focusing on family and having a mediocre career is optimal and there really isn't anything wrong with that. But you won't get top achievement out of that. Nor will you get that from forcing people to be in the office. You get it from a rare special alignment of stars where somehow you get an aligned high trust team. Just don't trample that flower, that's all.
Regarding cancer. Lots of cancer research, simulations and drug discovery use GPUs that were only developed because there was a gaming industry which was kickstarted by a bunch of outlier nerds like John Carmack and the popularity of those 3d games enabled economies of scale for making specialized hardware, GPUs. And then other tinkerers who likely also pulled lots of all nighters developed GPGPU, general purpose computations on GPUs which was not the original purpose of GPUs at all. Seeing the research success of such uses, NVIDIA developed Cuda and made GPUs more convenient in non graphics use cases. You just never know. Let the outlier people do their thing and pursue their passion. There will never be a world where all the average employees can have this kind of output. Somebody has to dream a vision, the implementation tasks don't fall out of the sky. Tall poppy syndrome is very bad for societies.
The vast majority of people working very long hours aren't pushing any frontiers. They are just trying to become partners in a law firm or get promoted at McKinsey. (Or, if we look outside our middle class bubble for a moment, they are working in kitchens and cleaning apartments for below minimum wage.)
I'm not convinced that a significant number of people actually do object to other people working long hours. It's only a problem if unreasonable expectations become the norm – in which case you don't have to attribute malicious motives to people ('tall poppy syndrome') to explain their objections. They object simply because they don't want to spend all their time working, which is easy enough to comprehend.
I'm convinced that you can't ever get much done outside high trust environments, you just can't. The bureaucracy eventually takes over, managers generally wall things off, keep things secret, erode trust, new people can't even navigate the bureaucracy are so far from effecting change, new ideas just DOA.
Not to say big companies or big projects can't make money / be well adopted, but you want to really do change, try new things sometime this century? ... need high trust.
The Bell Telephone monopoly was among the biggest of the bigcorps of that era and had no shortage of MBAs. A better characterization might be that the Ma Bell had money to burn and computerization was a hot trend so the bean counters were willing to back speculative projects and research, not unlike VCs throwing money around.
It’s worth remembering this was Bell Labs. It was a research and development organization. These weren’t line maintenance people and telephone operators free to create things during their work. They did a lot of applied research and a lot of pure research at Bell Labs back in the day.
Lasers, fiber optics, underseas cables, communication satellites, transistors, discovery of the cosmic microwave background, and more came out of that organization. It was largely supported by the consent deal between the US government and AT&T that allowed them a telephone monopoly so long as their non-telephone research and inventions didn’t get marketed separately. So they’d create things that helped the company and the rest of the world, and then just release those things to the rest of the world.
There's a danger of romanticizing the past, but I think there was just less metric based pressure on these people than today. In the 70s, the whole consultant-led transformation of all corporations wasn't fully complete yet.
The article even makes reference to the use-or-lose approach to managing a departmental budget: And history shows that it happened partly because the department paying for it “had extra money, and if they didn’t spend it, they’d lose it the next year…”
As for Rockstar North/DMA specifically: It was a bunch of nerds making games in Scotland. From having reverse engineered gta3 and vice city and therefore knowing the code of these games quite intimately, i can tell that even at that time (i don't know what exactly was meant by "early") they were still a fairly small bunch of very talented people building the best game together that they could. No huge engines or design patterns, just very straightforward, well or reasonably well written code that does just what's it supposed to. All from scratch, the tooling as well. Of course that's just my interpretation (and maybe i'm projecting a bit) but i imagine it must have been a very fun project for the people involved. Doesn't reek much of management, bureaucracy and questionable practices getting in the way.
That's true but it was little more than a portable rendering API, which was of course very useful for the PS2, but probably less interesting for the PC ports. So if you want to count that, you're right, it's not totally from scratch. But it wasn't built on an actual game engine.
I just remembered this video https://youtu.be/7vWSi44ZTSw and it seemed like a chill place with nerds having fun making something. (Actually they were still DMA at this time, not Rockstar).
And as for the achievement, the product turned into a franchise with the biggest entertainment products ever made (GTA 5/6).
The early users being patent secretaries, then "administrative kind of stuff, typing in trouble tickets," and adoption spreading because people liked it, is kind of cool. That creates different kinds of pressures than a big top-down-dictated project does, maybe healthy pressures: if you're going to play with a new idea about how things should work you can't break things; you need the thing running reliably for the people using it day-to-day. One way you can have huge projects fail is by fiddling around too long without contact with reality.
Given Linux's origins--"(just a hobby, won't be big and professional like gnu)"--it's interesting that early UNIX, in this telling, was also not the big professional push to build the OS of the future so much as just some folks trying to cobble something useful together (though of course, that they were playing around in Bell Labs gave their experiment some great advantages!).
I hadn't heard about the stolen security boots. It's interesting that it was resolved by a peer-to-peer negotiated settlement for the security guards to violate official corporate policy, rather than through management.
> Thompson remembers designing the Unix filesystem on a blackboard in an office with Rudd Canaday — using a special Bell Labs phone number that took dictation and delivered a typed-up transcript the next day.
Fancy :), this just became normal for the general public in the last couple of years. I assume of course that there was a secretary at the end of the line, not AI. But it's not completely unthinkable, Bell Labs did do very impressive things in text-to-speech at least.
Yeah, something about the ephemeral nature of spoken words that makes it easier to ramble and therefore go into unexpected and more "natural" directions, compared to text which I also have the need to strictly control as I'm typing it.
Using dictation for when you really need to not go back and edit is really helpful.
Best way I have found to get past this is to break the feedback loop. Easiest was is to type with my eyes closed , or on a system with high latency like epaper or ssh over lte. It forces me to do more work in my head and only use the computer to record it.
From the first paragraph at the above page (just for starters):
>As a former subsidiary of the American Telephone and Telegraph Company (AT&T), Bell Labs and its researchers have been credited with the development of radio astronomy, the transistor, the laser, the photovoltaic cell, the charge-coupled device (CCD), information theory, the Unix operating system, and the programming languages B, C, C++, S, SNOBOL, AWK, AMPL, and others, throughout the 20th century. Eleven Nobel Prizes and five Turing Awards have been awarded for work completed at Bell Laboratories.[1]
Also see the section titled Discoveries and developments.
Well yes of course, I just meant that them having voice transcription AI even in the 60s is not that unlikely, since they did a lot specifically in speech synthesis and audio analysis, besides everything else. I don't think they had it, but they had some of the important prerequisites figured out.
I guess it makes sense, voice-related AI is highly relevant for a phone company.
EDIT: Not full voice transcription, but it turns out that they did have systems to recognise digits and small (~a dozen) vocabularies of command words.
The prototype called Audrey could recognise 0-9 spoken digits with 97% accuracy in 1952! Although it did have to be calibrated for each individual speaker.
It's interesting how so many of the early tools were designed to create "communities" (mesg, talk etc.). The semi open nature of the platform really encouraged it too. It's nice to be able to cd into someone else's home directory and look at their files.
The passwords were only if you were connecting over the network. If you were using a directly attached terminal, you didn't need one.
RMS insisted that everyone use their UNAME as their password, but he wasn't widely listened to because the whole reason PWORD came into effect was because turists were getting increasingly destructive. People weren't happy when their mail got marked read (or worse, deleted) because some random from the network had logged in as them simply because they could and did not understand what their automatic login script was doing.
That was only true in very early systems. By the time of the PDP-10, HACTRN will nag you to log in if you run most commands and the gunner would kill off your job after a relatively short interval (the exact interval differed from machine to machine).
ITS had no file permissions, but even before PWORD was installed to keep randoms from the network away there were means of keeping the turists out when the system was to be reserved for Real Work. Other parts of the system that were considered sensitive were hidden behind undocumented commands or program-level passwords - For example, the innards of INQUIR, since the INQUIR database determined who was to be excluded and who was not.
There may have been no file permissions, but there was a definite hierarchy of users that was enforced by other (generally more subtle) means.
The caption under the picture of ken and dmr both standing that attributes to the latter “their motivation was to build a system ‘around which a fellowship could form’” calls to mind quotes from The Lord of the Rings in comments at the top of perl’s source files. For example:
There's this hardcore punk album from 1981 called "This is Boston not LA." On it, there's a track called "Radio UNIX USA" by the FUs.
I can't find ANY origin stories about the title. The lyrics have nothing to do with UNIX either, weirdly enough. However, this band is from Boston, and MIT was doing UNIXy stuff at around this time.
Anyone have any clue as to the origin for this track?
The lyrics include the lines "But you got / No balls no balls no balls no balls no balls / No balls no balls no balls no balls no balls", so "Unix" is very likely a pun on "eunuchs". I'm not very familiar with US radio station naming conventions, but it seems like 4-letter call signs are common? So the origin could be as simple as converting "eunuchs" to a radio station call sign.
All US FM radio station call signs start with either W or K (depending on location, mostly); an acronym starting with U wouldn't look like a call sign at all to me
If it isn't on the Internet, it didn't happen right? Maybe we can change that...
Unix _is_ a play on "eunuchs" but that fact wouldn't have sold well during the mini computer [ https://www.britannica.com/technology/minicomputer ] wars, especially in the later 1980s [ https://youtu.be/IRpKHFfsH3A ] when Unix was exiting the exclusive world of academia and Bell Labs. This was an era when everyone for the prior thirty years had come up with mainframes and data centers were stocked full of "heavy iron" (IBM and IBM-clones like Amdahl) or at least very large "mini" computers from companies like DEC, which was so well run from the late 1950s on that its leader has been declared to be one of the greatest in the history of Corporate America and was studied at Harvard and Wharton for decades. The Unix technology "specialists" on the other hand were super-nerds: ghastly, feral, mostly pear-shaped, plain clothed technicians only BARELY tolerated in their own settings. [Ok, maybe all of them except Eric Schmidt heh.] Realize that the vast majority of American engineers in the 1980s still wore suits and ties --but not if they had anything to do with Unix. (Ken Olsen fer sure wore suits, but also drove a Ford Pinto, BTW.) When the demo dollies tasked with pushing any number of alternate hardware platforms up against IBM and DEC in the constant battle for those massive "heavy iron" budgets were asked to pitch UNIX (System 3, System V, System 7, BSD) up against bedrock OS/MVS and VMS, at first they would answer the obvious question (of what UNIX stood for) with "UNIX is not UNIX". That pretty much stuck in the period literature (COMPUTER MAGAZINE RAGS) too --no way they were going to answer "eunuchs"!
Also worth noting in this context: This was the era of "Nobody Ever Got Fired For Buying IBM" and the amount of money your company spent on "iron" was seen as a marker of its success AND YOUR PERSONAL CAREER STATUS in the tech universe, so you can imagine the type of customers and professionals that actually did buy into obscure UNIX-based hardware. This also created a lot of "friction" in the Industry that you can't easily learn about in this Future. It wasn't like today where people have "home labs" and can train themselves to go for whatever job they want using free software (even while sitting in a hellish ghetto of the poorest country on Earth). Back in the day, one was trained on what their school or employer had available (or they learned from carrying around books and imagination, or using X.25-based timeshare if they were lucky). Period. So maybe you landed a great job, but you had to use a shitty Unix computer with broken down terminals or maybe you had a shitty job but they gave you a coveted VAXstation. All your experience with Unix wouldn't buy you much in a DEC or IBM shop and vice versa. The implications this had on the layered applications of the day were profound, but mostly this created a lot of animosity between tech professionals of different backgrounds. There were constant attempts to address this, but the computer hardware manufacturers were complicit in it because it made it easier to lock their customers into one architecture or another.
Is it safe to now say that all general purpose operating systems except LINUX are nothing but husks to run LINUX (and whatever legacy ecosystem)? The most successful of all the 1980s demo dollies, The Scott McNealy, took a page out of DEC's playbook and instead of trying to go in with a massive super powerful Unix mini computer, he would pitch a few workstations running something called "SunOS" (BSD eunuchs) that "networked" over TCP/IP to effect "the system is the network" (a totally new concept then) before his company bet everything on a new chip (SPARC) that used RISC architecture to outperform the established industry players and make the guys in charge of those "heavy iron" budgets feel a bit inferior if they didn't buy-in a little. SunOS, SPARC and Solaris definitely caused a lot of disruption, but it really never had much of a chance to unseat IBM or DEC and was also slowly sinking into La Brea tar pit along with everything else (though it had a bit more life due to all the capX as the dot-com bubble was inflating around TCP/IP). IBM had already totally lost control of its maverick PC initiative (by under-estimating Billy The Kid who had also hired away DEC's top VMS engineer) and the ENTIRE market for mini computers (whether they ran OS/MVS, VMS or eunuchs like SunOS or NextStep) totally collapsed. Just the promise that a PC might be as powerful at VMS and could network as well as SunOS was sufficient to change perception and bet corporate budgets on a "computer, not a terminal, for every desk." More importantly, the resulting PC industry economies of scale meant that all of the tech workers could own a "home lab" and, in particular, allowed at least one kid growing up just outside the Soviet Union to go through the pages of Andy Tanenbaum's famous book on operating systems (that demonstrated key concepts for the reader through the creation of a eunuchs operating system Andy called MINIX). Combined with the political antics of a creepy academic communist at MIT and an irresponsible Defense backbone ISP in San Diego, the slow death of all operating systems has manifested (because LINUX ELF binaries and runtime support are now available on IBM mainframes, Windows and as of last June, macOS). Of course, there are still legacy shops, embedded systems and most of the new ELF-running operating systems still run LINUX in nested virtualization, but LINUX has pretty much taken over the game and eunuchs is el muerto.
Meanwhile, even the AIs incorrectly think that UNIX is "a playful reference on UNICS, the larger, more complex Multics 'project'" Sounds totally plausible like everything else coming out of an LLM, but we meat bags know better.
---
'Now there were these places called cities and they had the knowin' of a lot of things, they did. They had skyscrapers, videos and sonic.. Then this thing called the Pockey Clips happened and you have to understand, this is Home and there's no Tomorrow Land.' https://youtu.be/rn4aIinTJBQ
This was an incredibly interesting read and, combined with the other poster's response above, completely answered my question. Thank you so much for typing this out. What a completely different world this must have been. (I started my career in 2007.)
If you enjoyed this, you might check out "Hackers" by Steven Levy. I read it as a kid when I was first getting into programming, and revisited it recently. It really held up for me. The book traces hacker culture from the MIT AI Lab through the Homebrew Computer Club to the early game programmers, and what got me excited then and now is the pure joy of building things in collaboration with like-minded people. I've managed to spend a lot of my career in early stage startups where this is still possible under the right circumstances.
I haven't read "The Innovators" but I've heard good things about it. However, Isaacson's recent Musk biography was pretty widely acknowledged as being a hagiography, and the NY Times reporters who wrote "Character Limit" about the Twitter takeover revealed that Isaacson was giving Musk advice on Twitter UX and business decisions while supposedly writing an objective biography. I'd definitely recommend "Character Limit" if you're interested in that saga - it's thoroughly reported and includes some wild details about how chaotic and destructive the whole thing was. Reading "Hackers" as a kid, this wasn't quite the tech industry I thought I was signing up for.
This is one of those books that I read in the 80s that helped me change career directions to be a programmer in Silicon Valley and eventually get a PhD and teach programming at the university level.
The serious beards were a century earlier, when the terms "sideburns" and "mutton chops" were coined, when Dickens had a doorknocker beard, when Thomas Nast drew Uncle Sam with a goatee, and very few men were clean-shaven.
One of the early pictures on that page shows Ken Thompson didn't have a beard in the early 1970s.
It is most likely, related to how growing one nowadays is a kind of hipster thing with the trendy barber shops decorated as if they were western barber shops scattered a bit all over the globe.
One of my favorite Ken Thompson hacks is one where he demonstrated how a backdoor could be introduced into a compiler in such a way that it would be difficult to notice https://wiki.c2.com/?TheKenThompsonHack
It's interesting and heartwarming to see how similar the spirit of many successful software projects was. Creative collaboration, open play, extremely high trust, by people who really intrinsically love what they do.
It goes against so much of the MBA-worldview and bigcorp offices.
Unix, GNU, Linux, early Python, early Rockstar Games etc.
> extremely high trust
A lot of problems disappear when you have a high-trust societies, projects, companies, etc.
Yes, but that requires admitting that the employees are flesh and bones humans with human social relationships between them and this stuff can't be tabulated and accounted and bean-counted. And it's rather outright seen as a risk factor, makes people less fungible, there's a risk that they gang up against the management, or there could be legal risk if they build too much of an insider culture etc.
Highly aligned motivated and talented people with shared core values (like the pioneer ethos of hackers, tech optimism etc, and cultural references etc) can achieve so much. And they tend to work odd hours and overtime because they want to make the thing, and not because of hoping to get rich.
Now I have two things to say about this.
First, this type of passion is special and a manager can't hope to simply force it, and if it's demanded then all you get will be talkers who know the lingo of being so super passionate. The thing must be worth being passionate about and the team has to be aligned with the goal and have latitude to shape the project. This kind of work bes happens when the higher up kinda forget the team and they just intrinsically do it. It can't be planned.
Second, some people are very opposed to these hight effort, almost obsessed teams because they see it as unfair ideals or unfair competition, because obviously someone with a family who has to pick up the kids at 4 pm can't do this cracked push overnight and sleeping at the office etc. But greatness simy cannot be made from steady, fixed-pace 9-5 jobs with work life balance and atomized employees. And that's okay, for most things we don't need greatness just okayness. But often people still can't stand that there are such great teams and want to drag them down in one way or another.
> But often people still can't stand that there are such great teams and want to drag them down in one way or another.
This seems like an unnecessary attribution of malicious intentions. The obvious explanation for why people often oppose a culture of long hours is that long hours suck for anyone with a life outside of work. You explain this yourself with the example of someone who has kids.
Yes, but it shouldn't be mandatory for "regular" jobs. People often dislike the idea that there are some ambitious, agentic, talented, motivated people who really put in the hours and their job is simultaneously their hobby etc.
Again, I'm not for the cringe corporate pretending to be one big family smiling and singing. But clearly there are really strong, aligned teams that want to accomplish something and have the corresponding freedom to creatively shape the path forward. And such teams are just vastly more effective and can do stuff in a weekend that a traditional committee-based process would not get done in a year.
But again, this is a fraction of a percent of people who are built for that kind of work. But let some people just be weird and misfits! Somehow the same standards are not applied to athletes who train insane hours or musicians on tour.
My point is, let's be grateful to have such exceptional people instead of complaining about them or how they set an unrealistic standard or whatever.
If these exceptional people are making it necessary for me to work very long hours too, then it’s fair enough for me to complain about that, no? Very few of the people you are talking about are curing cancer (so to speak), so why are the rest of us required to be grateful to them?
Let’s look at it another way. Someone who’s willing to work an 80 hour week for the same pay that I get is roughly equivalent to someone who’ll do my job for half the pay (leaving aside the dubious productivity benefits of long hours). Should I be grateful for the existence of such a person? We do not usually romanticize people who are willing to do professional jobs for low compensation. Why romanticize people who work crazy hours? If I voluntarily took a 50% pay cut, would you wax lyrical about how the world needs more exceptional people like me?
> And such teams are just vastly more effective and can do stuff in a weekend that a traditional committee-based process would not get done in a year.
Here you’re conflating two different things. Small agile teams that have the freedom to work without bureaucratic overhead are great, but there is no inherent need for them to work crazy hours. If anything, long hours are often a symptom of an environment where people are judged on how long they stay in the office rather than on the quality of their work.
This is beypnd jobs and corporations and is about humanity and civilization. A small minority of people in every era actually pushed on any kind of frontier. That's ok. We need people who just execute the mundane tasks too, that's most of us.
But giving space to exceptional people is not even that expensive. The overtime doesn't even have to be documented. Just leave them alone. There aren't enough of them to truly outcompete regular people.
For most people focusing on family and having a mediocre career is optimal and there really isn't anything wrong with that. But you won't get top achievement out of that. Nor will you get that from forcing people to be in the office. You get it from a rare special alignment of stars where somehow you get an aligned high trust team. Just don't trample that flower, that's all.
Regarding cancer. Lots of cancer research, simulations and drug discovery use GPUs that were only developed because there was a gaming industry which was kickstarted by a bunch of outlier nerds like John Carmack and the popularity of those 3d games enabled economies of scale for making specialized hardware, GPUs. And then other tinkerers who likely also pulled lots of all nighters developed GPGPU, general purpose computations on GPUs which was not the original purpose of GPUs at all. Seeing the research success of such uses, NVIDIA developed Cuda and made GPUs more convenient in non graphics use cases. You just never know. Let the outlier people do their thing and pursue their passion. There will never be a world where all the average employees can have this kind of output. Somebody has to dream a vision, the implementation tasks don't fall out of the sky. Tall poppy syndrome is very bad for societies.
The vast majority of people working very long hours aren't pushing any frontiers. They are just trying to become partners in a law firm or get promoted at McKinsey. (Or, if we look outside our middle class bubble for a moment, they are working in kitchens and cleaning apartments for below minimum wage.)
I'm not convinced that a significant number of people actually do object to other people working long hours. It's only a problem if unreasonable expectations become the norm – in which case you don't have to attribute malicious motives to people ('tall poppy syndrome') to explain their objections. They object simply because they don't want to spend all their time working, which is easy enough to comprehend.
I worked at a big corp for a long time.
Then I moved to a small company.
I'm convinced that you can't ever get much done outside high trust environments, you just can't. The bureaucracy eventually takes over, managers generally wall things off, keep things secret, erode trust, new people can't even navigate the bureaucracy are so far from effecting change, new ideas just DOA.
Not to say big companies or big projects can't make money / be well adopted, but you want to really do change, try new things sometime this century? ... need high trust.
The Bell Telephone monopoly was among the biggest of the bigcorps of that era and had no shortage of MBAs. A better characterization might be that the Ma Bell had money to burn and computerization was a hot trend so the bean counters were willing to back speculative projects and research, not unlike VCs throwing money around.
It’s worth remembering this was Bell Labs. It was a research and development organization. These weren’t line maintenance people and telephone operators free to create things during their work. They did a lot of applied research and a lot of pure research at Bell Labs back in the day.
Lasers, fiber optics, underseas cables, communication satellites, transistors, discovery of the cosmic microwave background, and more came out of that organization. It was largely supported by the consent deal between the US government and AT&T that allowed them a telephone monopoly so long as their non-telephone research and inventions didn’t get marketed separately. So they’d create things that helped the company and the rest of the world, and then just release those things to the rest of the world.
There's a danger of romanticizing the past, but I think there was just less metric based pressure on these people than today. In the 70s, the whole consultant-led transformation of all corporations wasn't fully complete yet.
The article even makes reference to the use-or-lose approach to managing a departmental budget: And history shows that it happened partly because the department paying for it “had extra money, and if they didn’t spend it, they’d lose it the next year…”
Like in government departments and also in some companies that have a similar mentality.
> early Rockstar Games
I did not expect to see them in this list, can you elaborate?
As for Rockstar North/DMA specifically: It was a bunch of nerds making games in Scotland. From having reverse engineered gta3 and vice city and therefore knowing the code of these games quite intimately, i can tell that even at that time (i don't know what exactly was meant by "early") they were still a fairly small bunch of very talented people building the best game together that they could. No huge engines or design patterns, just very straightforward, well or reasonably well written code that does just what's it supposed to. All from scratch, the tooling as well. Of course that's just my interpretation (and maybe i'm projecting a bit) but i imagine it must have been a very fun project for the people involved. Doesn't reek much of management, bureaucracy and questionable practices getting in the way.
From scratch? AIUI GTA3 and Vice City were built on RenderWare.
RenderWare was the preeminent PS2 graphics library. My impression is that it's more akin to OpenGL than to Unreal?
That's true but it was little more than a portable rendering API, which was of course very useful for the PS2, but probably less interesting for the PC ports. So if you want to count that, you're right, it's not totally from scratch. But it wasn't built on an actual game engine.
I just remembered this video https://youtu.be/7vWSi44ZTSw and it seemed like a chill place with nerds having fun making something. (Actually they were still DMA at this time, not Rockstar).
And as for the achievement, the product turned into a franchise with the biggest entertainment products ever made (GTA 5/6).
I would add "Ken Olsen"-DEC and Sun Microsystems.
The early users being patent secretaries, then "administrative kind of stuff, typing in trouble tickets," and adoption spreading because people liked it, is kind of cool. That creates different kinds of pressures than a big top-down-dictated project does, maybe healthy pressures: if you're going to play with a new idea about how things should work you can't break things; you need the thing running reliably for the people using it day-to-day. One way you can have huge projects fail is by fiddling around too long without contact with reality.
Given Linux's origins--"(just a hobby, won't be big and professional like gnu)"--it's interesting that early UNIX, in this telling, was also not the big professional push to build the OS of the future so much as just some folks trying to cobble something useful together (though of course, that they were playing around in Bell Labs gave their experiment some great advantages!).
I hadn't heard about the stolen security boots. It's interesting that it was resolved by a peer-to-peer negotiated settlement for the security guards to violate official corporate policy, rather than through management.
There's an MIT lock picking guide, from 1991. Is lock picking something widespread in academia, I suppose in the tradition of Richard Feynman?
Yes. Locks convert knowledge directly into strength.
> Thompson remembers designing the Unix filesystem on a blackboard in an office with Rudd Canaday — using a special Bell Labs phone number that took dictation and delivered a typed-up transcript the next day.
Fancy :), this just became normal for the general public in the last couple of years. I assume of course that there was a secretary at the end of the line, not AI. But it's not completely unthinkable, Bell Labs did do very impressive things in text-to-speech at least.
My thoughts come more fluently as speech than in writing. With writing I'm always wanting to go back and edit, which is distracting.
Yeah, something about the ephemeral nature of spoken words that makes it easier to ramble and therefore go into unexpected and more "natural" directions, compared to text which I also have the need to strictly control as I'm typing it.
Using dictation for when you really need to not go back and edit is really helpful.
Best way I have found to get past this is to break the feedback loop. Easiest was is to type with my eyes closed , or on a system with high latency like epaper or ssh over lte. It forces me to do more work in my head and only use the computer to record it.
>But it's not completely unthinkable, Bell Labs did do very impressive things in text-to-speech at least.
They did a lot more than that.
https://en.wikipedia.org/wiki/Bell_Labs
From the first paragraph at the above page (just for starters):
>As a former subsidiary of the American Telephone and Telegraph Company (AT&T), Bell Labs and its researchers have been credited with the development of radio astronomy, the transistor, the laser, the photovoltaic cell, the charge-coupled device (CCD), information theory, the Unix operating system, and the programming languages B, C, C++, S, SNOBOL, AWK, AMPL, and others, throughout the 20th century. Eleven Nobel Prizes and five Turing Awards have been awarded for work completed at Bell Laboratories.[1]
Also see the section titled Discoveries and developments.
Well yes of course, I just meant that them having voice transcription AI even in the 60s is not that unlikely, since they did a lot specifically in speech synthesis and audio analysis, besides everything else. I don't think they had it, but they had some of the important prerequisites figured out.
I guess it makes sense, voice-related AI is highly relevant for a phone company.
EDIT: Not full voice transcription, but it turns out that they did have systems to recognise digits and small (~a dozen) vocabularies of command words.
The prototype called Audrey could recognise 0-9 spoken digits with 97% accuracy in 1952! Although it did have to be calibrated for each individual speaker.
If you want to see Ken's contributions to Go, they are all there in Git. There's some fun stuff there (no spoilers). :)
It's interesting how so many of the early tools were designed to create "communities" (mesg, talk etc.). The semi open nature of the platform really encouraged it too. It's nice to be able to cd into someone else's home directory and look at their files.
That's ITS' philosophy and design, not Unix. Every serious Unix server would have every HOME dir with 0700 perms.
ITS had no permissions and encouraged collaboration since the beginning.
Then they introduced passwords. However, Stallman insisted that everyone use the same one, you can still boot it:
https://github.com/PDP-10/its
The passwords were only if you were connecting over the network. If you were using a directly attached terminal, you didn't need one.
RMS insisted that everyone use their UNAME as their password, but he wasn't widely listened to because the whole reason PWORD came into effect was because turists were getting increasingly destructive. People weren't happy when their mail got marked read (or worse, deleted) because some random from the network had logged in as them simply because they could and did not understand what their automatic login script was doing.
Login in was just a 'gentleman' policy. Everyone would just have root permissions and help/get helped from anywhere to anywhere in the system.
That was only true in very early systems. By the time of the PDP-10, HACTRN will nag you to log in if you run most commands and the gunner would kill off your job after a relatively short interval (the exact interval differed from machine to machine).
Still more open than Unix. Also if you got your hands on DDT, it would be a non-issue.
HACTRN is the (timesharing top level job) DDT.
ITS had no file permissions, but even before PWORD was installed to keep randoms from the network away there were means of keeping the turists out when the system was to be reserved for Real Work. Other parts of the system that were considered sensitive were hidden behind undocumented commands or program-level passwords - For example, the innards of INQUIR, since the INQUIR database determined who was to be excluded and who was not.
There may have been no file permissions, but there was a definite hierarchy of users that was enforced by other (generally more subtle) means.
“It’s nice to be able to walk into someone else’s office and look through their desk.”
The caption under the picture of ken and dmr both standing that attributes to the latter “their motivation was to build a system ‘around which a fellowship could form’” calls to mind quotes from The Lord of the Rings in comments at the top of perl’s source files. For example:
https://github.com/Perl/perl5/blob/blead/perl.c#L15
Okay, so weird and maybe unrelated question.
There's this hardcore punk album from 1981 called "This is Boston not LA." On it, there's a track called "Radio UNIX USA" by the FUs.
I can't find ANY origin stories about the title. The lyrics have nothing to do with UNIX either, weirdly enough. However, this band is from Boston, and MIT was doing UNIXy stuff at around this time.
Anyone have any clue as to the origin for this track?
The lyrics include the lines "But you got / No balls no balls no balls no balls no balls / No balls no balls no balls no balls no balls", so "Unix" is very likely a pun on "eunuchs". I'm not very familiar with US radio station naming conventions, but it seems like 4-letter call signs are common? So the origin could be as simple as converting "eunuchs" to a radio station call sign.
All US FM radio station call signs start with either W or K (depending on location, mostly); an acronym starting with U wouldn't look like a call sign at all to me
Right; it would start with a W.
If it isn't on the Internet, it didn't happen right? Maybe we can change that...
Unix _is_ a play on "eunuchs" but that fact wouldn't have sold well during the mini computer [ https://www.britannica.com/technology/minicomputer ] wars, especially in the later 1980s [ https://youtu.be/IRpKHFfsH3A ] when Unix was exiting the exclusive world of academia and Bell Labs. This was an era when everyone for the prior thirty years had come up with mainframes and data centers were stocked full of "heavy iron" (IBM and IBM-clones like Amdahl) or at least very large "mini" computers from companies like DEC, which was so well run from the late 1950s on that its leader has been declared to be one of the greatest in the history of Corporate America and was studied at Harvard and Wharton for decades. The Unix technology "specialists" on the other hand were super-nerds: ghastly, feral, mostly pear-shaped, plain clothed technicians only BARELY tolerated in their own settings. [Ok, maybe all of them except Eric Schmidt heh.] Realize that the vast majority of American engineers in the 1980s still wore suits and ties --but not if they had anything to do with Unix. (Ken Olsen fer sure wore suits, but also drove a Ford Pinto, BTW.) When the demo dollies tasked with pushing any number of alternate hardware platforms up against IBM and DEC in the constant battle for those massive "heavy iron" budgets were asked to pitch UNIX (System 3, System V, System 7, BSD) up against bedrock OS/MVS and VMS, at first they would answer the obvious question (of what UNIX stood for) with "UNIX is not UNIX". That pretty much stuck in the period literature (COMPUTER MAGAZINE RAGS) too --no way they were going to answer "eunuchs"!
Also worth noting in this context: This was the era of "Nobody Ever Got Fired For Buying IBM" and the amount of money your company spent on "iron" was seen as a marker of its success AND YOUR PERSONAL CAREER STATUS in the tech universe, so you can imagine the type of customers and professionals that actually did buy into obscure UNIX-based hardware. This also created a lot of "friction" in the Industry that you can't easily learn about in this Future. It wasn't like today where people have "home labs" and can train themselves to go for whatever job they want using free software (even while sitting in a hellish ghetto of the poorest country on Earth). Back in the day, one was trained on what their school or employer had available (or they learned from carrying around books and imagination, or using X.25-based timeshare if they were lucky). Period. So maybe you landed a great job, but you had to use a shitty Unix computer with broken down terminals or maybe you had a shitty job but they gave you a coveted VAXstation. All your experience with Unix wouldn't buy you much in a DEC or IBM shop and vice versa. The implications this had on the layered applications of the day were profound, but mostly this created a lot of animosity between tech professionals of different backgrounds. There were constant attempts to address this, but the computer hardware manufacturers were complicit in it because it made it easier to lock their customers into one architecture or another.
<cue https://youtu.be/ciUfdVs-p84 >
Is it safe to now say that all general purpose operating systems except LINUX are nothing but husks to run LINUX (and whatever legacy ecosystem)? The most successful of all the 1980s demo dollies, The Scott McNealy, took a page out of DEC's playbook and instead of trying to go in with a massive super powerful Unix mini computer, he would pitch a few workstations running something called "SunOS" (BSD eunuchs) that "networked" over TCP/IP to effect "the system is the network" (a totally new concept then) before his company bet everything on a new chip (SPARC) that used RISC architecture to outperform the established industry players and make the guys in charge of those "heavy iron" budgets feel a bit inferior if they didn't buy-in a little. SunOS, SPARC and Solaris definitely caused a lot of disruption, but it really never had much of a chance to unseat IBM or DEC and was also slowly sinking into La Brea tar pit along with everything else (though it had a bit more life due to all the capX as the dot-com bubble was inflating around TCP/IP). IBM had already totally lost control of its maverick PC initiative (by under-estimating Billy The Kid who had also hired away DEC's top VMS engineer) and the ENTIRE market for mini computers (whether they ran OS/MVS, VMS or eunuchs like SunOS or NextStep) totally collapsed. Just the promise that a PC might be as powerful at VMS and could network as well as SunOS was sufficient to change perception and bet corporate budgets on a "computer, not a terminal, for every desk." More importantly, the resulting PC industry economies of scale meant that all of the tech workers could own a "home lab" and, in particular, allowed at least one kid growing up just outside the Soviet Union to go through the pages of Andy Tanenbaum's famous book on operating systems (that demonstrated key concepts for the reader through the creation of a eunuchs operating system Andy called MINIX). Combined with the political antics of a creepy academic communist at MIT and an irresponsible Defense backbone ISP in San Diego, the slow death of all operating systems has manifested (because LINUX ELF binaries and runtime support are now available on IBM mainframes, Windows and as of last June, macOS). Of course, there are still legacy shops, embedded systems and most of the new ELF-running operating systems still run LINUX in nested virtualization, but LINUX has pretty much taken over the game and eunuchs is el muerto.
Meanwhile, even the AIs incorrectly think that UNIX is "a playful reference on UNICS, the larger, more complex Multics 'project'" Sounds totally plausible like everything else coming out of an LLM, but we meat bags know better.
--- 'Now there were these places called cities and they had the knowin' of a lot of things, they did. They had skyscrapers, videos and sonic.. Then this thing called the Pockey Clips happened and you have to understand, this is Home and there's no Tomorrow Land.' https://youtu.be/rn4aIinTJBQ
This was an incredibly interesting read and, combined with the other poster's response above, completely answered my question. Thank you so much for typing this out. What a completely different world this must have been. (I started my career in 2007.)
Ken Thompson interviewed by Brian Kernighan at VCF East in 2019 > https://www.youtube.com/watch?v=EY6q5dv_B-o
Love this one, well worth the watch
"Unix: A History and a Memoir, by Brian Kernihgan" is also an excellent read.
Birth of a serious change (and leadership) always requires questioning of status quo and probably a bit of rowdy, jungle instincts.
If you enjoyed this, you might check out "Hackers" by Steven Levy. I read it as a kid when I was first getting into programming, and revisited it recently. It really held up for me. The book traces hacker culture from the MIT AI Lab through the Homebrew Computer Club to the early game programmers, and what got me excited then and now is the pure joy of building things in collaboration with like-minded people. I've managed to spend a lot of my career in early stage startups where this is still possible under the right circumstances.
The Innovators is a must read as well. Walter Isaacson is world class.
https://www.goodreads.com/book/show/21856367-the-innovators
I haven't read "The Innovators" but I've heard good things about it. However, Isaacson's recent Musk biography was pretty widely acknowledged as being a hagiography, and the NY Times reporters who wrote "Character Limit" about the Twitter takeover revealed that Isaacson was giving Musk advice on Twitter UX and business decisions while supposedly writing an objective biography. I'd definitely recommend "Character Limit" if you're interested in that saga - it's thoroughly reported and includes some wild details about how chaotic and destructive the whole thing was. Reading "Hackers" as a kid, this wasn't quite the tech industry I thought I was signing up for.
That's unfortunate but I won't read anything about musk anyway so it's going to be a dead artist situation for me.
Thanks for posting this! Seriously great book.
This is one of those books that I read in the 80s that helped me change career directions to be a programmer in Silicon Valley and eventually get a PhD and teach programming at the university level.
back in the days when beards were serious beards
The serious beards were a century earlier, when the terms "sideburns" and "mutton chops" were coined, when Dickens had a doorknocker beard, when Thomas Nast drew Uncle Sam with a goatee, and very few men were clean-shaven.
One of the early pictures on that page shows Ken Thompson didn't have a beard in the early 1970s.
That’s…sarcasm?
It is most likely, related to how growing one nowadays is a kind of hipster thing with the trendy barber shops decorated as if they were western barber shops scattered a bit all over the globe.
You don't think Ken's beard was serious?
Nope - simple observation!
We still have serious beards kicking around.
The Linux folks, Andrew Kelley etc all qualify as True Beards.
One of my favorite Ken Thompson hacks is one where he demonstrated how a backdoor could be introduced into a compiler in such a way that it would be difficult to notice https://wiki.c2.com/?TheKenThompsonHack
For those interested, LaurieWired recently published a video about this very thing.
https://youtu.be/Fu3laL5VYdM
See also his Turing Award lecture “Reflections on Trusting Trust.”
https://www.cs.cmu.edu/~rdriley/487/papers/Thompson_1984_Ref...