Oral History of Linus Torvalds
CHM Ref: X4147.2008 © 2008 Computer History Museum Page 20 of 41
Booch: The 0.12 release was also interesting in that you released it under the GPL license I understand.
Tell me about that.
Torvalds: Well, the GPL, I think that actually happened-- I think 12 was released with the old license.
But in the release, that was the timeframe where really people in the U.S. were starting to-- like at Unix
users' groups they were starting to do disk duplication services where, I mean, you said, "Hey, I found this
ploy on the internet." And people literally wanted to fit on three or four disks. You've the kernel. You've
got some utilities there. And Unix users' groups would have people who did this for them. But the original
license said, "You have to give all modifications back and you cannot charge anything, not even copying
fees." Right? And that all harks back to the fact that okay, I was a poor student and I had thought that
MINIX was way too expensive for me when it cost whatever, $160. So the source part was a big issue for
me, but the price part was literally also an issue. So the first license literally said you can't even-- even if
you, like, do disc duplication services you'd have to give out all that work for free. And people actually
started emailing me and saying, "Hey, could I please recoup my expenses?" Right? I mean they didn't
want to make money off this, but they wanted to at least not have to lose money on it. And when you get
to that point and like, where do you draw the line. Right? Your time is an expense, too. You want to
maybe get five bucks for the effort of having done this disc copying service. And I thought that was kind
of-- that sounded perfectly reasonable even to me. And I realized that the whole cost issue was really not
an issue at all, and it was unreasonable to really have that in the license at all. So I had been aware of
the GPL and I had friends who were rabid about it. I was never really all that rabid about it, but it really fit
my needs very well. And it was a well-known license. It was a well-known license especially of the one
program I really kind of looked up to, GCC. Without GCC I couldn't have done any of this. So it was kind
of also saying, okay, GCC is really important to me. GCC uses the GPL. I'll reciprocate, not by giving
code backs in GCC, but at least kind of say that okay, you've used a really nice license, I'll do the same.
Booch: Had you trademarked the term Linux by this time? That came later?
Torvalds: No. No. I mean that is the kind of paperwork crap that I don't want anything whatsoever to do
with. And the fact that I have the trademark now is just a huge pain in the ass.
Booch: Why is it a pain in the ass? People write to you for use of the trademark?
Torvalds: Well, no, we've actually set it up now, finally, so that you can basically self-certify most of the
time. Well, there's these rules about when you can use Linux without any-- needing any permission. But
it still comes up every once in a while. But it used to be a pain just because then eventually when Linux--
not that much later when Linux became more now, and I still lived in Finland, somebody else trademarked
the name and started extorting people for money, basically. And then suddenly the fact that I had not
trademarked it was-- I mean this would never have been an issue in Finland. I was still in Finland.
Trademarking, why? That's not something non-commercial now. So people ended up actually taking this
to court. And I had really nothing to do with it except for the fact that I was the only person who people
could kind of agree to trust with the trademark, because all the people involved who really hated the
trademark in the first place were all like distribution makers on a small scale at that point, and like the
Linux Journal. And they all were kind of nervous about each other, but they weren't nervous about me, so
I ended up getting the trademark just because of that. But it was just a horrible experience. And…
Oral History of Linus Torvalds
CHM Ref: X4147.2008 © 2008 Computer History Museum Page 21 of 41
Booch: You mentioned the Christmas release in December '91, January '92. But then events really
started unfolding then because you had the .95 version, the first Linux newsgroup and the first
commercial distribution. How did that unfold?
Torvalds: Well, I actually don't even know how the commercial distributions unfolded. I mean they
became legal with the change in license, and they were pretty small at the time. I forget which one may
have been the first commercial. It might have been SLS which…
Booch: Do you remember who created that?
Torvalds: No.
Booch: That was the soft-lining Linux System?
Torvalds: Soft lining. I was wondering what was SLS? What did it even stand for? I couldn't remember.
So I really didn't have anything to do with it. More importantly I didn't want to have anything to do with it.
I was perfectly happy with people doing it. In fact I was really pleasantly surprised by the fact that, okay,
apparently it's doing so well that people wanted to pay for the-- I mean, it was mostly floppy disc copying
services at the time. And the only thing I cared about was that I wanted all the improvements to be-- to
come back. They obviously did. Right? And the thing I cared about more was the fact-- and the thing
that caused us to jump number again was not only did the license change after-- or with .12. but some
people started putting X to Linux. And that wasn't something I hadn't really foreseen, because I'd actually
not used X very much. We had xterm also at the university, but I'd mostly interacted with Unix actually
over my regular serial line terminal. So I was like, the fact that we had graphics was something really
unexpected to me, right, and not something I had ever really planned. So then people put an X to Linux,
again mostly by actually building on the Linux capabilities. And that was such a huge step that I thought
okay, we're getting close to a 1.0 release so I'll call it 0.95 instead of 0.13 which would have been, I think,
the natural progression. And it turns out that I was a bit too optimistic.
Booch: In the sense that it wasn't quite ready for prime time?
Torvalds:
Yes, in that sense. And we actually ended up…
END OF TAPE 2
Booch: Events in 92 where commercialization really began and I have in my notes Adam Richter was a
gentlemen who did one of the first commercial releases.
Torvalds: Yes.
Booch: Tell me about him and how you encountered him and vice versa.
Torvalds: This is again, one of the fun things about Linux has been how
I did not-- the whole Adam
Richter thi
ng and making a distribution, he just did it, and it ended up being called Yggdrasil. I don’t