Being able to install a UNIX-type operating system on my personal computer, one that was free, was a life-changing experience for me.
At the time there were prohibitively expensive UNIX operating systems on the market, many of which required even more expensive proprietary hardware to run on, and then you'd have to fork out even more money for a compiler.
An enormous thanks to Linus and the GNU team for changing all of that and making this accessible to pretty much anyone crazy enough to try and install it on their computer.
You wonder what the innovation curve looks like when people are interested then can't afford the 20k per year for unix4. Someone probably would have gotten there (GUN/hurd is only 20 years away, probably because of many people working on linux instead). A lot of the innovation in tech is a result of tinkering or being able to run on a low cost platform to make the margins pencil out. This wasn't much of an option under the commercial OSs that existed before Linus started.
Ah, google groups, the only reliable way to download more than a megabyte of data to render 1kb of text.
Somewhere, I still have a copy of this message, from when it appeared in my newsreader. It would be an interesting exercise in archaeology to see if modern linux has the tools to mount that old filesystem... guess the rest of today's productivity will have to take a back seat.
I'm pretty sure a filesystem from back then should be mountable-- the filesystem Linus implemented for the first iterations of Linux was the MINIX filesystem, and there's still a kernel driver for it [1].
I don't have a filesystem from back then to actually try it, though.
Ah, memories of my Slackware (kernel 1.2.8) CD set. Whatever version of LILO it shipped with munged the lilo.conf, and hence MBR, every time you ran it.
That may be true for desktops, but on a server, you can still run a very clean system. My virtual server has 16 processes running on it, six of them "systemd-something".
Well done! I am mostly referring to the 'bracket' kernel thread processes, which were not there back in the day. I must admit some laziness here, as I could research these in just a few minutes.
I avoid Gnome and KDE for that reason: they involve too many processes.
Sure enough, my first shot at installing Slackware also resulted in LILO trashing my MBR.
I was attempting a dual-boot setup with Windows 95 at the time and, being too inexperienced to recover from the mishap, lost everything in the process. (Although one could argue that nothing of value was lost; Windows 95 was practically unusable.)
I blame both Slack and LILO for traumatizing an impressionable young mind: some 20 years on, I still hesitate whenever I have to deal with a bootloader.
Thankfully, nowadays we can boot by straight-up running the kernel as a UEFI executable[1]. No more MBR/VBR/bootloader song and dance for me.
Someone pointed out to me recently that the ubiquitous icon for 'save' is the depiction of a floppy drive, which is essentially a skeuomorph (https://en.wikipedia.org/wiki/Skeuomorph).
I love all the parentheses (including nested parentheses!) in that introduction post from Linus, and how he's careful to make it clear that it's just a hobby and not professional.
Linux is amazing! From 86% of the world's smartphones to pretty much all of the top500 supercomputers to millions of Chromebooks in schools to tens of millions of critical servers to umpteen embedded uses I've never even heard of!
Linus must be considered one of the greatest project managers ever (seriously!)...results don't lie. His manner is almost identical to Leslie Groves, who managed both the construction of the Pentagon and then the Manhattan Project (greatest engineering project in history). Some tasks seem to demand people who don't need to be liked.
It is just incredible that we can use something as awesome as Linux in a free and open manner.
Thanks Linus! I don't care if you're a jerk, you deliver like a freaking boss!
> Thanks Linus! I don't care if you're a jerk, you deliver like a freaking boss!
When you put it into perspective like that, he's not a jerk, just someone who has had a statistically insignificant number of bad days.
I'd like to see any other person on earth deal with their project being pulled in that many directions and act like a saint. I'd have told you all to go and get fucked long ago.
Politeness is a good thing to strive for much of the time, but I also think that no-bullshit directness is just as important, if not more so.
It becomes a self-fulfilling prophecy. The less direct we become as a people, the more sensitive people become to it, the more offense they take to it, the less direct we become, and so forth.
Basically, Linus is a valiant anti-bullshit warrior.
Most people don't remember (or weren't around to see) what it was like before Linux. IMHO Linus's biggest contribution to the world was the modern open source development methodology.
In my 4th year at university I did a project on Mach and was very excited to continue working with it through the HURD. At the time, getting up to date source code for the HURD required sending an email to the team and requesting it. I dutifully did so and was greeted with a reply asking for my CV. I sent it in (as bare as you might imagine it would be as a university student) and was denied access since they only wanted experienced kernel developers to work on the project. I've often regretted not saving that email.
But this is the way it was back then. You waited until something was released to play with it, or you contacted the team to see if they would grant you access to the latest development code.
Linus changed all that by giving ubiquitous access to the code and taking patches from anyone. It was a huge revelation. Keep in mind that this coincided with more open access to the internet, so his attitude was facilitated by the fact that you didn't need to rely on UUCP to slowly distribute code. More and more, people had access to FTP (and soon the web).
The reason for Linux's success, IMHO, is down to that. Everyone flocked to Linux because it was obviously unencumbered by the BSD legal issues and Linus would look at any patch coming his way, regardless of who sent it. This is now de rigour -- to the point where I'm sure the vast majority of people using free software today believe that it's always been that way.
For me, RMS invented the concept of free software. Linus showed how to actually make it work. There were other good projects at the time, but from my perspective nothing came close to Linux.
It's a fair point. I often wonder what would have happened without the potential BSD lawsuits hanging over everything. I remember at the time being torn between installing Linux on my box or BSD. When mmap was finally implemented on Linux, I decided (rather unhappily IIRC) to go with it thinking it was the "safer" choice.
One thing to keep in mind, though, wrt OpenBSD is that by the time it arrived on the scene, Linux was well and truly established. The better comparison is NetBSD, which showed up in 1993, but even then Linux was in the fabled 0.99 version (and rapidly running out of letters). It was really Linus's actions before that time that cemented Linux as a legitimate contender -- he had attracted a really large number of very talented programmers. NetBSD, if memory serves, was suffering from a fair amount of internal infighting which eventually ended up with Theo de Raadt being ousted.
Correct me if I'm wrong, but my understanding is that companies maintain patchsets that they periodically update to apply to newer mainline versions. A lot of it ends up merged into mainline eventually. They're not really forks.
> Thanks Linus! I don't care if you're a jerk, you deliver like a freaking boss!
He's not a jerk. People just like to post writings of his, out of context.
My favourite out-of-context post of his was here on HN where it was the typical "oo, isn't Linus sweary!" porn. It was post ~150 of a mailthread, and he was blowing up at a user. I went back through the thread, and this was post #12 of Linus's, and most of his previous comments were politely trying to help that same user understand a concept. The user was being dickish and probably intentionally obstructive.
> In 1981, the spin-off company Symbolics had hired away nearly all of the hackers from the AI Lab, and the depopulated community was unable to maintain itself.
Been a full time Linux user for over a year now. Windows 10 pushed me off the Microsoft treadmill. The future is Linux, everything else will be an historical oddity.
I remember when they showed commercials during the Super Bowl about Linux. My 12-year-old geek self couldn't believe it! I had a bunch of older people at my parent's party asking me about Linux. Felt cool to be able to show off my home built computer running slackware to a bunch of adults who often chastised me for being in the basement too much :)
Remember getting Linux root and boot diskette images via FTP mail (https://en.wikipedia.org/wiki/FTPmail) and downloading them using UUCP on a "superfast" 9600 baud dialup connection. Thx Linus!
66 comments
[ 2.9 ms ] story [ 123 ms ] threadAt the time there were prohibitively expensive UNIX operating systems on the market, many of which required even more expensive proprietary hardware to run on, and then you'd have to fork out even more money for a compiler.
An enormous thanks to Linus and the GNU team for changing all of that and making this accessible to pretty much anyone crazy enough to try and install it on their computer.
Somewhere, I still have a copy of this message, from when it appeared in my newsreader. It would be an interesting exercise in archaeology to see if modern linux has the tools to mount that old filesystem... guess the rest of today's productivity will have to take a back seat.
I don't have a filesystem from back then to actually try it, though.
[1] http://lxr.free-electrons.com/source/fs/minix/
Incidentally I can't overemphasize how far basic sysadmin skills will get you.
Trial by fire. Good times.
Also I knew the purpose of every process running on my box. Now I am just as bewildered as I am by Windows.
I avoid Gnome and KDE for that reason: they involve too many processes.
I was attempting a dual-boot setup with Windows 95 at the time and, being too inexperienced to recover from the mishap, lost everything in the process. (Although one could argue that nothing of value was lost; Windows 95 was practically unusable.)
I blame both Slack and LILO for traumatizing an impressionable young mind: some 20 years on, I still hesitate whenever I have to deal with a bootloader.
Thankfully, nowadays we can boot by straight-up running the kernel as a UEFI executable[1]. No more MBR/VBR/bootloader song and dance for me.
[1] https://www.kernel.org/doc/Documentation/efi-stub.txt
It wouldn't have done me much good, though, as I didn't have a CD drive. I had to try to scroung up enough floppies to write out the "A" and "N" sets.
Someone pointed out to me recently that the ubiquitous icon for 'save' is the depiction of a floppy drive, which is essentially a skeuomorph (https://en.wikipedia.org/wiki/Skeuomorph).
Congrats Torvalds, thanks for changing the world!
And thus Linus dug a hole out of which Linux has only recently begun to clamber.
Ha.
"Thanks for creating Linux." 24/06/2011 John
which you can graft to happen before that commit. Or just look at separately if you're lazy ;)
Linus must be considered one of the greatest project managers ever (seriously!)...results don't lie. His manner is almost identical to Leslie Groves, who managed both the construction of the Pentagon and then the Manhattan Project (greatest engineering project in history). Some tasks seem to demand people who don't need to be liked.
It is just incredible that we can use something as awesome as Linux in a free and open manner.
Thanks Linus! I don't care if you're a jerk, you deliver like a freaking boss!
When you put it into perspective like that, he's not a jerk, just someone who has had a statistically insignificant number of bad days.
I'd like to see any other person on earth deal with their project being pulled in that many directions and act like a saint. I'd have told you all to go and get fucked long ago.
It becomes a self-fulfilling prophecy. The less direct we become as a people, the more sensitive people become to it, the more offense they take to it, the less direct we become, and so forth.
Basically, Linus is a valiant anti-bullshit warrior.
In my 4th year at university I did a project on Mach and was very excited to continue working with it through the HURD. At the time, getting up to date source code for the HURD required sending an email to the team and requesting it. I dutifully did so and was greeted with a reply asking for my CV. I sent it in (as bare as you might imagine it would be as a university student) and was denied access since they only wanted experienced kernel developers to work on the project. I've often regretted not saving that email.
But this is the way it was back then. You waited until something was released to play with it, or you contacted the team to see if they would grant you access to the latest development code.
Linus changed all that by giving ubiquitous access to the code and taking patches from anyone. It was a huge revelation. Keep in mind that this coincided with more open access to the internet, so his attitude was facilitated by the fact that you didn't need to rely on UUCP to slowly distribute code. More and more, people had access to FTP (and soon the web).
The reason for Linux's success, IMHO, is down to that. Everyone flocked to Linux because it was obviously unencumbered by the BSD legal issues and Linus would look at any patch coming his way, regardless of who sent it. This is now de rigour -- to the point where I'm sure the vast majority of people using free software today believe that it's always been that way.
For me, RMS invented the concept of free software. Linus showed how to actually make it work. There were other good projects at the time, but from my perspective nothing came close to Linux.
OpenBSD made a pretty big contribution as well.
http://www.openbsd.org/papers/anoncvs-paper.pdf
http://www.openbsd.org/papers/anoncvs-slides.pdf
One thing to keep in mind, though, wrt OpenBSD is that by the time it arrived on the scene, Linux was well and truly established. The better comparison is NetBSD, which showed up in 1993, but even then Linux was in the fabled 0.99 version (and rapidly running out of letters). It was really Linus's actions before that time that cemented Linux as a legitimate contender -- he had attracted a really large number of very talented programmers. NetBSD, if memory serves, was suffering from a fair amount of internal infighting which eventually ended up with Theo de Raadt being ousted.
He's not a jerk. People just like to post writings of his, out of context.
My favourite out-of-context post of his was here on HN where it was the typical "oo, isn't Linus sweary!" porn. It was post ~150 of a mailthread, and he was blowing up at a user. I went back through the thread, and this was post #12 of Linus's, and most of his previous comments were politely trying to help that same user understand a concept. The user was being dickish and probably intentionally obstructive.
RMS and his grudge against the brain drain at MIT single-handedly changed the free software movement forever.
I haven't really heard about this part of his motivation. What was going on at MIT that made him feel that there was a brain drain in progress?
> In 1981, the spin-off company Symbolics had hired away nearly all of the hackers from the AI Lab, and the depopulated community was unable to maintain itself.
- Theo de Raadt
:)
Damn. That's a lot of floppies.
Well, I'll be damned. It works! Let's see if I can't get this Apache web server (what silly names! Tee hee!) to go.
Ha! That works better than (whatever I was using - I forget).
(still using it)