How Linus Torvalds Invented Today's Work From Home Paradigm In 1991
from the way-of-the-penguin dept
Working from home is beginning to move from being a necessary but temporary way of achieving social distancing in offices, to a radical shift in how many companies will operate. Until now, most of the evidence of that change has been anecdotal. But a Twitter thread by Chris Herd, who is CEO of FirstbaseHQ, which "lets you supply, finance and manage all the physical equipment your remote teams need to do great work at home", provides some fascinating statistics on the scale of the shift to working from home. Herd says he has talked to around 1000 companies over the last six months about their plans for remote work. One trend is that corporate headquarters are "finished", he says: companies will cut their commercial office space by 40 to 60%, with people working from home for two to four days each week. Some 30% of the companies Herd talked to say that they intend to get rid of offices completely, and move fully to remote working.
Some of the reasons for this shift are obvious. Things like increasing worker satisfaction by avoiding stressful daily commuting, and enabling them to participate in family life during the entire day through flexible working patterns. Slashing office costs is a major factor for the companies, but also cited is the reduction in the pollution generated by traditional office working. However, the main driver for a shift to remote working may be surprising:
The first reason they are going remote-first is simple -- it lets them hire more talented people
Rather than hiring the best person in a 30-mile radius of the office, they can hire the best person in the world for every role
Traditional ways of running a company have made it hard to bring about this change. But there is one sphere whose stunning success is built on this very shift. The world of free software and open source has embraced distributed teams working at home for nearly 30 years. This has allowed projects to select people on the basis of their skills, rather on their availability for a local office. It also means that people can work on what they are best at, and most interested in, rather than on what their local team needs them to do. As a result, open source software has gone from a bit of coding fun in the bedroom of a Finnish student, Linus Torvalds, to the dominant form of software in every field, with the lone exception of the desktop. Its success has also inspired a range of related movements, such as open access, open data, open science and many more.
What's remarkable is that Linus did not set out to create this new kind of global, distributed software development methodology. It simply evolved from the time he placed his first, rough version of the Linux kernel on an FTP server in Finland, and invited people to download it freely. The crucial step was his willingness to accept suggestions to improve the code from anyone, provided they were good ones. That encouraged people to join the project, because they knew that there was no traditional business hierarchy based on seniority, just a meritocracy, where their suggestions would be accepted if their work was demonstrably better than the existing code. The companies that will thrive most from today's epochal shift to working from home will be those that are willing to implement similar ideas to those of Linus from 30 years ago, transposed to a general business context.
Follow me @glynmoody on Twitter, Diaspora, or Mastodon.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: distributed, linus torvalds, open source, pandemic, work from home
Reader Comments
Subscribe: RSS
View by: Time | Thread
Is this finally a case of
Herd Community?
[ link to this | view in chronology ]
Re: Is this finally a case of
And, as it happens, a case of GNU/Hurd Immunity.
[ link to this | view in chronology ]
You might be thinking union-controlled business hierarchies. In the tech world, unions are not a problem. But there is still the other hierarchy (or kleptocracy, if you will): it's based on MBA-ness.[1] Success involves more nobbling of your peers and toadying to your overlord than cultivating of your own work units. The good part of this is that most cultivation does more damage to the effectiveness of the work unit; the bad part is, that humans often would have responded well to good cultivation. Obviously, people who are both able and willing to accomplish useful work will prefer a meritocracy to a kleptocracy (whether or not overlaid on a union-based senility scale over the actual work-units.)
[1] I've long been defining "M.B.A. program" as "a process for inducing sociopathy in favorably predisposed individuals".
[ link to this | view in chronology ]
Torvalds just ported from source code made by real programmers.
No idea why you wish to aggrandize Torvalds -- especially when he merely imitated Unix, which was then and is now inferior to MS-DOS (which is FAR from perfect).
So he programmed in his bedroom? WOW. So did THOUSANDS in the 80s.
And if you think "open source" isn't generally CRAP, just read the comments any week at Distrowatch: about half of all distros can't get to work at all, rest have many problems caused (in my opinion) by attempting "cool" and "eye candy" instead of seeing that basic functions actually work.
[ link to this | view in chronology ]
Re: Torvalds just ported from source code made by real programme
I always love it when ignorant dickheads use open protocols via their open source browser to navigate to a website built on an open source database, web server on an open source operating system to talk shit about how open source is useless.
[ link to this | view in chronology ]
Re: Re: Torvalds just ported from source code made by real progr
It's almost like it's an open source of open sores for this opinesaurus[1].
[1] Opinesaurus - one whose system of cognition (such as it is), remains trapped in a bygone era and considers buggy whips to be an advanced technology.
[ link to this | view in chronology ]
Re: Re: Re: Torvalds just ported from source code made by real p
I do like poking fun at them. I bet he has no idea how many open source components he had to use in order to post that nonsense, but I'd be surprised if the number was in the single digits.
[ link to this | view in chronology ]
Re: Re: Re: Re: Torvalds just ported from source code made by re
and he probably sucks at golf
[ link to this | view in chronology ]
Re: Torvalds just ported from source code made by real programme
You're joking, right? Unix is so stable that Steve Jobs used it as the kernel for OS X and the current Macintosh Operating System is still #10 so as to show how reliable it is. MS-DOS was so bad its only benefit is compatibility because the massive legacy of programs that use it is why it was supported for so long.
[ link to this | view in chronology ]
Re: Re: Torvalds just ported from source code made by real progr
I think he means inferior in the way a back-hoe is inferior to a shovel, since you only need rudimentary skills to operate a shovel.
[ link to this | view in chronology ]
Re: Re: Torvalds just ported from source code made by real progr
He's right - MS-DOS is superior to UNIX... if your needs don't involve all those pesky security features and optimised performance.
Although the question then becomes why he's commenting on a site hosted on a current UNIX clone to argue that an OS that ceased to exist 20 years ago is superior. Maybe it's the same reason he's using an open sourced browser to state that open source software doesn't work.
[ link to this | view in chronology ]
Re: Re: Torvalds just ported from source code made by real progr
History: MS-DOS exists due to fact Digital Research CP/M creator would not bend to Microsoft. IBM PC-DOS BIOS was recreated so the compatible market sell boxes with non-IBM BIOS with MS-DOS and pass compatibility.
CP/M is a better small model operating system than MS-DOS but M$ killed it. There is much more insanity around PC computer than I care say.
Unix: Mac OS 9 was Apple's look and feel of Xerox Alto. AT&T Unix based code could run, I have never installed it. Mac osx is based on BSD/Darwin open source code - and much friction on the lack of giveback.
Currently: iOS is slowly making osx evolve macOS. MacOS 10 is intel and new macOS 11 is for apple ARM Macs (run iOS apps in future).
The war on general purpose computers is in scorched earth mode. The work by Torvalds is going to be last Unix-like OS standing.
[ link to this | view in chronology ]
Re: Re: Re: Torvalds just ported from source code made by real p
"History: MS-DOS exists due to fact Digital Research CP/M creator would not bend to Microsoft"
Hmmm... that's part of the story. The version I'm aware of being most accurate is that company founder Gary Kildall was flying a plane on the day in question, having just started a vacation not knowing IBM negotiators were en route and couldn't be reliably reached by the people negotiating with IBM. The problem being that IBM demanded a non-disclosure agreement before they would go into any details about what the meeting was for, and the staff in the meeting were not sure if it was a good idea to sign such a thing without details. Because Microsoft (who at that point were best known for a BASIC interpreted) had promised a working OS quickly, they went to 86-DOS, a CP/M clone, rather than wait to be able to negotiate again with the original founder.
I mention these distinctions because they tie into so many issues we're dealing with to the modern day - MS were interested in getting something done quickly, not going to the best (a trend they have long continued), they depended on 3rd party implementations of APIs and reverse engineering of BIOS code to sell their product (things they fight against when their competition uses them), and it's lawyers rather than engineers who mould some of their important decisions.
[ link to this | view in chronology ]
Re: Re: Torvalds just ported from source code made by real progr
I can not agree more on the fact the unix is so stable. My macbook (used unix as kernel) can run without turning off by months, but my windows PC can't since it will run very slowly if I run it for 1 week without turning off.
[ link to this | view in chronology ]
Re: Torvalds just ported from source code made by real programme
As usual, Baghdad Bob, you keep shooting your own arguments in the neck;
"...especially when he merely imitated Unix, which was then and is now inferior to MS-DOS (which is FAR from perfect..."
It's pretty obvious that if he had imitated Unix...Novell would have owned his ass in one brief burst of that copyright you glorify so often. So self-evidently, your assertion is wrong.
And then you try to compare MS-DOS, a broken piece of garbage even it's original inventors now try to pretend doesn't exist was superior. Let me know why today not a single machine runs on MS-DOS while just about every other server in existence runs some form of linux.
"And if you think "open source" isn't generally CRAP..."
He said, ignorant that 95% of what he uses in daily life relies exclusively on Open Source.
[ link to this | view in chronology ]
Re: Re: Torvalds just ported from source code made by real progr
"And then you try to compare MS-DOS, a broken piece of garbage even it's original inventors now try to pretend doesn't exist was superior"
Now, now - the original MS-DOS was a rebranded version of QDOS, a clone of CP/M, and I'm sure the original creators of those have been pretending to forget it due to the bad deal they got for a long time...
Fun fact: QDOS was short for "Quick and Dirty Operating System", which while Microsoft didn't originate it, that does seem to have been their business model until recently.
"He said, ignorant that 95% of what he uses in daily life relies exclusively on Open Source."
I'd love him to identify the browser he's using, given that you really have to go out of your way at the moment to use a browser that isn't open source.
[ link to this | view in chronology ]
Re: Re: Re: Torvalds just ported from source code made by real p
"Fun fact: QDOS was short for "Quick and Dirty Operating System", which while Microsoft didn't originate it, that does seem to have been their business model until recently."
So the "quick and Dirty Operating System" gave rise to "Makes Systems Do Obnoxious Shit"? Well, the legacy seems fully preserved, I must say.
"I'd love him to identify the browser he's using, given that you really have to go out of your way at the moment to use a browser that isn't open source."
Given that we can put a timestamp on old Bobmail's onset of recursive spurts of insane diarrhea on this forum and Torrentfreak alike and that we can measure his willingness to learn and adapt...I'd say odds are good he's still on IE 6.
[ link to this | view in chronology ]
Re: Re: Re: Re: Torvalds just ported from source code made by re
Nah, using IE6 would require actual thought and effort, intellectual honesty and sticking to the steady conviction of his stated beliefs. If he's on any remotely new computer, he would have to work to use something that old. That's doubtful, although being restricted to an ancient XP machine might explain his sudden disappearance over the last few months.
More likely, he thinks he's being cocky by using Edge, but is too ignorant to realise they switched to using the open source Chromium engine on the back end.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Torvalds just ported from source code made b
"That's doubtful, although being restricted to an ancient XP machine might explain his sudden disappearance over the last few months."
Given his current outburst about MS-DOS, are we sure he isn't still chugging away on an old x286?
"More likely, he thinks he's being cocky by using Edge, but is too ignorant to realise they switched to using the open source Chromium engine on the back end."
Well, he's often enough demonstrated a multi-dysfunctional case of dunning-kruger in IT (and business, law, history and basic math) so it's likely he has absolutely no clue beyond his firm beliefs and ad hoc convictions about whatever system he's using. Your guess is as good as any, I guess.
It truly wouldn't surprise me to see old Bobmail/Baghdad Bob/blue try to wax eloquently about the benefits of proprietary systems in full ignorance that MS latest offers are all rooted on their new-found love of Open Source.
It's kinda his unique tell to consistently manage to make a claim which on casual perusal turns out to be the exact opposite of what he claims.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Torvalds just ported from source code ma
"in full ignorance that MS latest offers are all rooted on their new-found love of Open Source"
A "love" that has come around purely as a reaction to market forces that got them pummelled in the free market, which they mainly survived due to a combination of lock-in and lack of interoperability that kept people using their products until they changes their way - and even that was as much due to major mistakes that had nothing to do with open source (their horrendous performance in the mobile market, the toxic reaction to the Metro interface, etc.) that made non-proprietary solutions so much more attractive.
Seriously, when the major holdout from embracing FOSS was forced to do so for economic reasons, it's hilarious to see someone continue to cling to the claim that it's not viable, especially when so clearly using it himself to make those comments.
[ link to this | view in chronology ]
Re: Torvalds just ported from source code made by real programme
...and he has no idea how important open source is.
Even M$ recognizes the importance of open source. But since they can't make any money on open source they have to figure a way to imitate it.
For myself, I love open source. Guess what? I haven't paid the M$ tax to run this computer. Why? Because it's running on Linux. If ever there was an open source, Linux is it.
[ link to this | view in chronology ]
I've used Multics. It used to be said that operating systems programmers, having not understood Multics, were forever doomed to (failing) attempts to copy it.
To be fair, MS-DOS is not a failed attempt to copy Multics. It's only a failed attempt to copy a dumpster copy of DR-DOS. Which didn't really try to be an operating system at all.
The needle touching this matter is called "incommensurability".
[ link to this | view in chronology ]
Linus didn’t invent this.
I started working from home in 1979. It wasn’t a big deal, I didn’t invent working from home either. I used to go in once a week and drop off a floppy of source code, and pick up my check.
[ link to this | view in chronology ]
Re: Linus didn’t invent this.
I'm pretty sure that Linus is being credited with demonstrating the effectiveness of remote team collaboration, not with the idea of working alone in your basement all week.
[ link to this | view in chronology ]
Re: Re: Linus didn’t invent this.
Kind of like how Al Gore didn't invent the internet (nor did he ever claim to), but rather demonstrated the effectiveness of the internet by passing legislation that led to the creation of the first internet browser.
[ link to this | view in chronology ]
Re: Linus didn’t invent this.
If Congress/Parliament not give copyright control for software in 1970s, open source start in 1970s instead 1990s. Copyright delay modern open source 20 years.
[ link to this | view in chronology ]
Re: Re: Linus didn’t invent this.
Not quite. The idea of Copyleft and Free Software was around the time that copyright became opt-out instead of opt-in:
https://en.wikipedia.org/wiki/Copyleft
From the article:
[ link to this | view in chronology ]
At first I thought this was going to be about how Linus Torvalds invented today's work from home paradigm in the same way that
Shiva Ayyadurai "invented" email. I should have known better since the headline did not have scare quotes.
[ link to this | view in chronology ]
Or just the cheapest.
I'm as happy as anybody not to have to commute every morning, but it's not difficult to see the downsides of companies looking at opportunities to outsource more jobs than they already have.
[ link to this | view in chronology ]
So, is Richard Stallman chopped liver?
Let's not forget that, while Linus was responsible for the kernel, some 90% of the codebase of the GNU/Linux system predates him, dating back as far as 1983 (though it's less clear when it became a work-from-home project).
Now if you want to credit Linus for Git/Github, that's a different story, and probably the one you should have written.
[ link to this | view in chronology ]
Re: So, is Richard Stallman chopped liver?
Yes, RMS is chopped liver.
He used the cleaver on himself, repeatedly, and like many in traditional business hierarchies he was focused on getting press for other people's work. See: GNU/Linux.
[ link to this | view in chronology ]
Remote working alone is not going to allow companies to recruit the best applicants. First companies need to do better at sorting applicants. As but one example, I've seen technical jobs in which one of the requirements listed is knowing how to use microsoft office. I have a PhD in physics, I've never used Office, but learning how to use some low level tool like that is a no brainer.
In general, a lot of the software I've used required either reading the source code and/or reading some handwritten documentation that was a few pages, very sparse and the rest was trial and error. On the other hand, I once worked at a job as a systems programmer in a unix environment that I was not qualified for because the only programming language I knew was Fortran along with a half dozen scripting languages and i really don't like programming. I'm not sure why they ever thought I was a programmer, but I got the invite via a usenet forum for some reason.
I literally bought K&R and wrote "Hello World," before starting work along with a few books on the Unix operating system and kept K&R at my desk to figure out how to do whatever I needed to do. I did managed to do just fine, even though I was a bit slower than my co-workers who were light years ahead of me. I'm sure there were a lot more people that were more qualified than I was they could have hired. I only left because I hated the job and I asked my boss if I needed to have a lobotomy and become a bean counter to get a raise and an office with a door.
The moral of the story is that unless the application process itself changes, no one will get the best applicants for the job. The idea that working remotely will do anything at all to get the most qualified applicants is dumb. All that will do is allow companies to hire people who tick the right boxes in resume that's scanned for key words based on a job ad that includes some requirements (like MS Office) that any idiot could meet just by starting Office and spending a few days using it.
I just started my own business to avoid that shit after not getting jobs I could easily do, but didn't tick the right boxes for some of the mickey mouse skills listed in the ad. Having access to a larger pool of applicants doen't change anything if the sorting process doesn't select on the right things. Also, my ex gf worked at home one or two days/week, but she was basically subject to being surveiled to ensure she was "working," via keyboard, audio and video. That's worse than going into work.
[ link to this | view in chronology ]
Re: Your failure is not a precedent
But an exception. Just like Fortran is today.
Maybe it's for the best, hiring programmers that like to program was always a better solution in my experience.
[ link to this | view in chronology ]
Re: Re: Your failure is not a precedent
Yes, surprisingly - fortran is still in use today.
[ link to this | view in chronology ]
Re:
Also problem from time zone, culture, language, company culture, best people have work already.
[ link to this | view in chronology ]
1976 Working from home
My father "telecommuted" from 1976 until 1979, and to some greater or lesser extent, from 1980 on, while developing some of General Electric's time sharing software (for the youth out there, "Time Sharing" = the OG Cloud that you think is such an amazing recent development). The phone company in England had installed a dedicated line, and G.E. provided a massive, 132 column, dot matrix terminal, complete with external cassette drive, and a ticker tape thingy that spat out long, yellow, perforated coils of paper, which were far too interesting for a young boy to resist. After we moved to the States, the giant beast was switched out for a TI Silent 700, complete with the accoustic coupler. That thing, connected to my the GE network via its 110 BAUD coupler, is what I first learned BASIC on, along with Fortran, etc.
Since the mid '90s, all I have wanted to do is meet Linus, shake his hand, and thank him for freeing me from the soul crushing curse of that abominable crossbreeding of a sloppy reimplementation of CP/M and a bucket of raw meat torn from the still living body of OS/2, but I cannot give him credit for inventing the paradigm of working from home via remote access.
[ link to this | view in chronology ]
Re: 1976 Working from home
"I cannot give him credit for inventing the paradigm of working from home via remote access."
I don't think people are saying he invented that per se, but rather than he popularised and proved the efficiency of team collaboration with no centralised physical location.
Whatever you think of the specific claims, there's no doubt that without the Linux project, we would be in a much different place overall to where we are today.
[ link to this | view in chronology ]
Re: Re: 1976 Working from home
"Whatever you think of the specific claims, there's no doubt that without the Linux project, we would be in a much different place overall to where we are today."
Yes. I can enthusiastically and emphatically agree with that. I can't readily think of any one person or industry that doesn't, in some way, owe Mr. Torvalds some sort of a "tip of the hat", regardless of whether or not they know it, want to admit it, or actively deny it. Personally, if I were to ever meet him, I would have a very hard time resisting the urge to hug the stuffing out of him.
[ link to this | view in chronology ]