decoration decoration
Stories

GROKLAW
When you want to know more...
decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books

Gear

Groklaw Gear

Click here to send an email to the editor of this weblog.


You won't find me on Facebook


Donate

Donate Paypal


No Legal Advice

The information on Groklaw is not intended to constitute legal advice. While Mark is a lawyer and he has asked other lawyers and law students to contribute articles, all of these articles are offered to help educate, not to provide specific legal advice. They are not your lawyers.

Here's Groklaw's comments policy.


What's New

STORIES
No new stories

COMMENTS last 48 hrs
No new comments


Sponsors

Hosting:
hosted by ibiblio

On servers donated to ibiblio by AMD.

Webmaster
Ever Wonder How the Kernel Gets Built?
Tuesday, February 03 2004 @ 05:15 PM EST

If you have ever wondered how the kernel is built, take a look at Infoworld. The article explains the process and tells how two bugs he found were quickly identified and fixed. The main impression you will form is that McBride's description of the process isn't even in the ballpark:
Decisions on what features and patches are incorporated into the official kernel are generally preceded by much debate among kernel developers, but are ultimately made by the kernel maintainer, a central authority who shoulders the brunt of day-to-day maintenance, as well as the responsibility for official kernel releases. Given the size and scope of the kernel, neither the maintainers nor Linus Torvalds himself can fully know and understand every portion of the kernel. To alleviate this, several unofficial kernel subsystem maintainers are entrusted to keep a watchful eye on their chosen sections of the code and to contribute validated patch sets to the maintainer for inclusion in the next release.

New releases of the stable kernel are vetted through a release candidate process, during which kernel patches are tested by the community. In addition to release candidate kernels, patches for stable and unstable kernels are distributed by a select few core developers, such as Alan Cox and Andrew Morton. These patches usually contain experimental code that hasn’t been officially introduced into the source by Torvalds, as well as bug fixes or hardware support likely to be incorporated into the next release.

While the kernel maintainers are responsible for the kernel under their care, Linus Torvalds still runs the show. Officially, Linus is the persistent maintainer of the current development kernel, and he hand picks the maintainer of the new stable kernel when release time approaches. Drawing on hundreds of developers, a few maintainers, and a QA team in the thousands, the Linux kernel keeps on rolling.
As you can see, it's organized, it's stable, and there is oversight. And Linus isn't doing it all singlehandedly, a picture McBride likes to paint so as to imply that things slip past Linus because it's too large a job for just one man. As you can see from the article, Linus isn't doing the work alone. Here is an interview with Andrew Morton, his right-hand man, but just one of his helpers.

They must be doing something right. As of today, according to Netcraft, SCO is running its new website on Linux. Ahem.

OS, Web Server and Hosting History for www.thescogroup.com

OS Linux
Server Apache
Last changed 3-Feb-2004

OS NetBSD/OpenBSD
Server Apache
Last changed 2-Feb-2004




  


Ever Wonder How the Kernel Gets Built? | 214 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
Ever Wonder How the Kernel Gets Built?
Authored by: Anonymous on Tuesday, February 03 2004 @ 05:33 PM EST
This got me to thinking. Does anyone know if there is a contingency plans in
place incase, God forbid, Linus or one of the other 'major players' were killed
or otherwise incompacitated? Who would take over the kernal and manage it in
the same way as Linus, or would Linux be consumed by another entity?

And I apologise now for the morbidness of this thought, but really, it should be
figured out now instead of when its too late. That is if it isn't already taken
care of.

[ Reply to This | # ]

Sounds like most large projects
Authored by: Anonymous on Tuesday, February 03 2004 @ 05:51 PM EST
Delegation of authority, review and test procedures, the works.

Worthy of any large COMMERCIAL endeavor.

No wonder so many COMPANIES trust it for their mission-critical applications.

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: Anonymous on Tuesday, February 03 2004 @ 05:54 PM EST
make menuconfig
make dep
make modules
make modules_install
make install


Wahh, Wahh, wahh....wahhhhhhhhhhhh. :P

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: caliboss on Tuesday, February 03 2004 @ 05:59 PM EST
If you want to know HOW the Linux development process came to be, read Rebel Code by Glyn Moody.

He covers the development of Linux from before the beginning, the spat with Tannenbaum (sp?), the problems with incorporating BSD/TCP code, etc. etc. It's written in a surprisingly easy-to-read manner - even for non-geeks. It would be smash hit with the Groklaw crowd - IMHO.

---
Grok the Law / Rock the World

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: Nick_UK on Tuesday, February 03 2004 @ 06:01 PM EST
Alan Cox was the #1 kernel hacker to Linus, but he is on a sabbatical (but he
still can't resist posting in the lkml).

For a free OS, the Linux kernel has a better organisation than any other
production (or Company) I know.

And there are 10,000's of people just doing little bits, let alone the core
stuff.

Nick

http://streetend.force9.co.uk/cgi-bin/wiki.pl?NickWarne

[ Reply to This | # ]

OT: The worm didn't hurt Microsoft
Authored by: toolboxnz on Tuesday, February 03 2004 @ 06:20 PM EST
http:// www.canada.com/technology/ story.html?id=6C29A5E1-9C5B-4EEC- BE98-0D3D5F0B434C

OK so it didn't hurt Microsoft. I don't have a reference to it but virus labs have indicated both the SCO and MS viruses were written by the same person. Surely they would have been clever enough to make both viruses are virulent as each other. This adds doubts to the level of DDoS that SCO "experienced".

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Developed?
Authored by: ajrs on Tuesday, February 03 2004 @ 06:23 PM EST
As some couldn't help but point out, anybody* can build the kernel. The novel thing about GNU/Linux, is it was the first time that anybody* could build an entirely free run my buisness on it system. The wonderful thing about the GPL is that anybody* can join the development team, and have their work rewarded by geting and derived work back.

*Anybody is subject to limitations of education, dedication, hardware, and natural ability. Anybody may be further restricted by free time, non disclosure and/or employee agreements, and other restrictions imposed by legal status, the DCMA, power outages, and sleep deprivation. Poster and/or Groklaw take no responsibility for you not being anybody, but are glad you are somebody.

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: dougS on Tuesday, February 03 2004 @ 06:28 PM EST
Actually they (SCO) may not be running linux ... several firewall products do,
and that's what you see from the outside.

Doug

[ Reply to This | # ]

  • What? - Authored by: Anonymous on Tuesday, February 03 2004 @ 06:41 PM EST
  • SCO using Linux - Authored by: Anonymous on Tuesday, February 03 2004 @ 10:34 PM EST
Irony: SCO is using Fedora
Authored by: converted on Tuesday, February 03 2004 @ 06:40 PM EST
After reading the canada.com article a couple posts above, I wanted to see if I could get to sco.com (no www) Well it failed. So for the hell of it I tried www.sco.com and got the same default install apache web page as on my box professing "Fedora Core Test Page" I almosted busted a gut laughing.

"I am not a demographic! I am a human being!"

[ Reply to This | # ]

Kernel knowledge :D
Authored by: inode_buddha on Tuesday, February 03 2004 @ 06:47 PM EST
God knows, I've built enough of them. It's pretty well described by other posters here, so I'd like to offer my observations about the process since I've been on the mail-list a few years.

1: It's based *completely* on trust. (Which implies honesty, IMHO) Linus trusts his top coders ("liutenants"). They have subordinates and peers that *they* trust. They review and vett code from anyone.
2:Meritocracy and mutual respect. Your stuff has to have merit from a very technical viewpoint; it's gonna get shredded, spindled, and mutilated until it makes the cut. Even IBM went through this for a year or so. If and when it gets in, merit and respect come into play but it *still* gets shreddded, spindled, and mutilated. Code quality and clear logic are above all.

3:Programmers who voluntarily go through all this are opinionated and emotional about their work. Do you blame them? This doesn't mean they're jerks; it means you have to *prove* it. They're quite gracious about being wrong, but you really have to show it first.

4:Flame-wars: PJ check this out OK? Most of the day-to-day list is quite dull unless you're a coder. You should see what happens when somebody brings up licensing, tho. I've seen lkml flame-wars drag on for *weeks*. See #3 above.

All-in-all, its good fun. You can subscribe here. I recommend the digest version with attatchments in-line. It's about 1Mb daily at 7AM EST.

Remember, it's totally open; anybody can post and read there,or just hang out.

---
"Truly, if Te is strong in one, all one needs to do is sit on one's ass, and the corpse of one's enemy shall be carried past shortly." (seen on USENET)

[ Reply to This | # ]

It's not a truck, it's a bus! :-)
Authored by: Anonymous on Tuesday, February 03 2004 @ 07:02 PM EST
See the Onion article for yourself!

[ Reply to This | # ]

  • thanks - Authored by: webster on Tuesday, February 03 2004 @ 07:22 PM EST
Ever Wonder How the Kernel Gets Built?
Authored by: Anonymous on Tuesday, February 03 2004 @ 07:07 PM EST
Why shouldn't they run their server with Linix? According to SCO, they own it.

[ Reply to This | # ]

OT: SCOX down 1.10
Authored by: red floyd on Tuesday, February 03 2004 @ 07:33 PM EST
SCOX closed down 1.10 today at 13.30

Of course, I wouldn't read too much into this, since the NASDAQ as a whole was
down.

---
The only reason we retain the rights we have is because people *JUST LIKE US*
died to preserve those rights.

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: Stumbles on Tuesday, February 03 2004 @ 07:53 PM EST
As others have noted here, the procedure, ie the order of the
commands to
recompile a kernel are fairly simple. The sometimes
tricky part is determining
what you want compiled as a module or into
the kernel or wether you want it
part of the kernel at all.

[ Reply to This | # ]

[OT] Wish I was able to question Darl.
Authored by: Anonymous on Tuesday, February 03 2004 @ 08:51 PM EST
While some of the questions put to Darl the other day were reasonably good, I
wish that they'd used more strategy.

You see, it is demonstrably futile to reason with Darl, or to convince him to
change his mind at this point. If nothing else, we have the "prior
investment" fallacy coming into play here--that is, they've bet the farm on
lawsuits. This won't end until SCO is legally "crippled" (and, most
likely, "gagged" ...) or insolvent.

Thus, most of the hostile questions missed the opportunity for something more
subtle. You see, Darl is clearly experienced in dodging questions. It is
obvious that he simply did not answer some of them--they were too blunt, rather
than subtle.

I'm sure you'd like to know just why the hell I know any better, considering
that several questions were still rather effective; at least, for the audience.
But I think we could do more harm via a different approach.

You see, Darl uses some old, familiar principles in answering pointed questions.
If you can't give an answer, don't. If you're accused, make a
counter-accusation. Action and reaction, a sort of verbal judo. Now, that's
not to say he's terribly skilled--some of his statements may well harm him in
court. We want him to make such statements, to dig his own grave as it were.
We should help him do that. Thus we must not protest his appearances, but
encourage them, that he might be given more opportunities to hurt himself.

Obviously, we ought to be prepared for this. First, let us examine the
techniques he uses. If a bit of thought is given into phrasing a hostile
question, we can to some degree, predict what strategy he will use to counter
it. That is, whether it gives him opportunity to make a counter-accusation, or
whether he must dodge it entirely. By proper application (and timing!) of
carefully constructed (perhaps even "flawed" in the sense that they
lend themselves to one of his particular rants) we can lead him into
self-contradiction, in the same presentation.

Yet this only helps the crowd. The most interested people here are generally
those who dislike Darl--we're not likely to have any need of convincing them
further. There is little point. "You're preaching to the choir."

What other tricks have we? How about non-hostile questions? How will those
hurt him, you ask? Hopefully, by revealing things he might not want us to know
but which we do not yet know, necessarily. One example I have thought up is to
ask him this: "There have been a number of online sources covering this
issue, from those rabidly against your cause like slashdot, to the articles
linked from your own website. Which news sources or journalists do you think
have given us the most accurate account of things so far?"

It's not "hostile" so it looks like a softball, no? But it reveals
who he feeds FUD to. This can be convenient for other reasons; for example, we
all know that Daniel Lyons should be canned--any "journalist" who
quotes slashdot trolls as sources deserves to be fired. Heaven help us if he
discovers IRC; his next article might well be entitled "hay d00dz im
l33t!!1!1!" Remind me to write the Forbes editors; I believe that their
editorial scrutiny is desperately needed...

But I digress. The topic at hand is how to help Darl get his foot further into
his mouth. Now I grant you, most of his leg is already in there, and I'm
wondering just how much further things can go, but we owe it to ourselves to
help him. In any event, if we use cleverly worded "softball"
questions like the above, especially between the "hardball" ones
(rather like playing "good cop, bad cop" here, no?), we may be able to
extract useful information. Done right, we might just get insight into things
like what part of the sealed BSD settlement he things might help him, what SCO's
exit strategy is (think of "where do you see yourself in five years"
type questions--and don't just let him talk about rolling in license money!),
and anything else we haven't already established from the public record.

Remember, he's making this up as he goes along doing whatever seems best at the
moment. He has long range goals more than concrete plans and a tendency to talk
too much. These are exploitable weaknesses; use them.

[ Reply to This | # ]

Hmmmmmm.
Authored by: Anonymous on Tuesday, February 03 2004 @ 09:28 PM EST
Towards the end of this account from an MIT student (from the /. post) there is this description of an exchange between Darl and Chis Stone:

He [D.Mc.] even looked back to his Software VP (Chris Sontag) guy and said "I guess what they're saying is that if we tone down the anti-GPL stuff a bit, we'll look better" (that's a mild paraphrase). The Software VP guy looked shocked, and didn't like that idea at all.

I wonder what to make of this?

[ Reply to This | # ]

OT: Found on SCO's website
Authored by: jachim on Tuesday, February 03 2004 @ 09:30 PM EST
Yes, this is totally offtopic but I just have to share! While searching for information about /etc/vfstab I discovered a very funny and apropos web page, the first result returned by Google for "vfstab" is the manpage for 'fud(1M)'. Unfortunately, it's not a Fear, Uncertainty and Doubt Howto, but simply the manpage for the file update daemon for the LKP. Anyway, I thought it was hilarious.....

[ Reply to This | # ]

OT: Fatal Attraction?
Authored by: Tim Ransom on Tuesday, February 03 2004 @ 10:00 PM EST
His Nibs has been disturbingly flattering of Google as of late. He apparently intimated that MS is working on a new search technology. The article includes speculation that they might integrate it into the OS.

Also, vexation about their anachronistic business practice continues:

'According to Eric Schmidt, Google’s ubiquity isn’t all that has Microsoft’s eyebrows rising. Mr. Schmidt reported back from Davos to a colleague in an email spotted by the New York Times that “Based upon [Microsoft’s] visceral reactions to any discussions about ‘open source,’ they are obsessed with open source as a business model.”

Google is, of course, the shining example of a business built on Linux and open source, so to applaud Google, is to cheer for a world that doesn’t need Microsoft. The Google service is run on more than 100,000 servers in 12 data centers around the world, and the company is rumored to be building the largest data center in the world.'


Thanks again,

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: Ted Powell on Tuesday, February 03 2004 @ 10:03 PM EST
If you have ever wondered how the kernel is built, take a look at Infoworld.
Or, for the executive summary, at a document featured here yesterday, the United Nations Conference on Trade and Development's E-COMMERCE AND DEVELOPMENT REPORT 2003 (UNCTAD/SIDTE/ECB/2003/1), Chapter 4: Free and open-source software: Implications for ICT policy and development, Section E3 (page 17 in the PDF, page 109 in the document).
GNU/Linux, as it expanded, developed a semi-formal organization for decision making about code. Clearly differentiated role structures exist within the GNU/Linux community. As the programme and the community of developers grew, Torvalds delegated responsibility for sub-systems and components to other developers, who became known as lieutenants. Some of the lieutenants onward-delegate responsibility to "area" owners whose work has a narrower focus. The organic result is what looks and functions very much like a hierarchical structure where decision making flows through a fairly well defined pyramid. The GNU/Linux pyramid works imperfectly but is evolving through trial and error towards greater scalability.

---
Truth is not determined by majority vote.

[ Reply to This | # ]

OT: Darl soon to enter FBI witness protection program
Authored by: belzecue on Wednesday, February 04 2004 @ 12:33 AM EST
Charles Murry, EE Times, writes:

"SCO Group is claiming that Linux builds on the company's Unix patent portfolio." [emph. added]


"When SCO was accused of fabricating the story of its attacks, it hired the University of San Diego's Cooperative Association for Internet Data Analysis, to verify the event. When the association did so, it too was hit with a denial-of-service attack.

We were attacked, we were attacked for saying we were attacked and then the agency who said we were attacked got attacked," noted a spokesman for SCO Group."


"The company spokesman said that SCO executives' lives have been threatened. When SCO Group chief executive officer Darl McBride appeared at the Mandalay Bay Convention Center in Las Vegas in November to deliver a keynote speech at CD Expo, the company brought a sharpshooter along for protection. "We were aware of specific threats from someone who had already served time behind bars," the spokesman said. "If they've been there [in prison] before and evidently aren't afraid to go back, we have to take them seriously."

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: Anonymous on Wednesday, February 04 2004 @ 12:36 AM EST
Well the official "Linux kernel" would be frozen. But distributers
like Suse and redhat would continue, since they maintain their own trees.

They would continue making on as they do, making improvements, releasing them
into open source and incorporating them into their products.

[ Reply to This | # ]

OT-Netcraft and SCO
Authored by: Anonymous on Wednesday, February 04 2004 @ 01:03 AM EST
I am looking at netcraft and they say thescogroup is offline. It is now 9:53
and for the last few minutes I have been looking at the sco site and netcraft.
I cannot resolve the address 216.250.128.21 using nslookup on windows XP pro. I
don't have time to look into this and see if there is some discrepency, but I
thought I would note that from 9:50 to 10:02 sco is working.

I wonder sco has been changing there web site from one os to another fairly
rapidly. Why the change? Could it be part of SCO's problems is that they don't
know how to set-up thier BSD system and before that during another outage they
were moving from one linux to another. If someone had the time, they might check
back and see how often these outages occur when they have changed thier OS.

Also it would be nice if some reporter would ask a McBride if proprietary is so
much better than opensource why doesn't sco's site run on sco unix? Or
Microsoft? Or Solaris or any other proprietary system. Could be cost? Is the
TCO of sco unix so high that even sco can't afford it? I assume they would give
themselves a discount. Couldn't they have negotiated a license from one of thier
partners. At least Sun and Microsoft run thier websites on thier own OS. Not
very encouraging to see a vendor that is afraid to use his own product for
running a website. I assume that apache does not spend a lot of time supporting
sco unix and with the shrinking tech staff at sco they probably don't have time
to make the needed changes to apache. It probably takes their best SW engineers
just to run grep a few times. Think how much time it took to find those abi
files that Linus wrote. If a normal engineer takes five minutes the sco guys
probably took a week or two. I suspect because they printed it out first on
paper like the pages they sent to IBM. Then they "greped" a pencil and
seached through the pages.

Sorry sarcasm got the best of me. But seriously how can someone take an OS
company seriously if they can't run a web site on thier own OS. Come on sco get
"YOUR" product and a commercially avalible and PROPRIETARY web server
and compete in the open market. Put your tech skills and money were your mouth
is! Have some guts for a change. Put up your stuff and see if it can play in the
real world. What do you have to lose. Your system is proprietary so it is
invincible according to your logic. So why not prove it where in counts on the
internet holding off the hordes. Or remain gutless.

Anyone is free to use any of the ideas in this post. Rewrite and amplify it.
Unlike Facist Darl who thinks every programmer in the last 20 years should give
SCO thier work so "ONLY" SCO can make money off it. IT is amazing to
me that anyone can listen to Darl and not think him a facist. In his world, I
am a criminal for writing a program and then sharing it. I wrote a ray-tracer
one time using some very smart persons ideas and code snippits and then I gave
it back to him as a working whole. Darl calls me a communist and wants the
federal govenment to control me. Darl that's Facist. For those too young to
remember good old hitler, then think of Sadam and people thrown in jail for
thinking and merely mentioning Sadam's name. How do you propose to control me
Darl? Thought police? I rarely use such rhetoric. I don't think it adds to a
discusion. But Darl has gone too far. He has called the open source movement
communists in so many words if not outrightly. (I haven't checked thoughly) He
has insulted the great programers in the open source movement at the same time
ripping off the IP of projects like SAMBA. Come on Darl, have to guts to live
your convictions. Remove SAMBA, the Linux code you ripped off for LKP. You know
you did not write it from scratch. You stole the code. Prove me wrong by
releasing it to the open source folk for thier review. Perens and others have
proven, PROVEN they can research the code. You have not shown any work that
substantiates that you have anyone with skills close to the guys who reviewed
you code from Vegas so I think the best, most capable people should review the
LKP. Proof Darl. Proof. If you do not know what it means, look it up in the
dictionary.

I am not opposed to a small company fighting for justice against a large one. I
would even be on your side if there was any shred of evidence that supported
your position. You have not shown a thing and if you boil down all of your
words it comes down to "TRUST ME." Sure Darl, Corporate CEO's have
earned our trust. Gee, Like Worldcom, Enron, the list goes on and on. How can we
trust a thing you say without proof?

Also, I was looking at all of the documents and arguments you are making. If you
paid 25 million for the advice you have been getting, what do you have to pay
for good advice? If you paid me 1 million I could show you where Stud boies, and
your brother and Hatch blew it. You had a line of reasoning that would have
forced IBM to give you what you want. It is clear that an alternative approach
might have worked. At least it would have been coherent and have had a chance.
You spent money on flash and glitz. If I were an investor, I would be asking for
your head on a platter. You and ken and all of the other CEO's that give
business a bad name. It is not the fact that you are trying to protect your IP,
all companies try to do that, no my biggest objection is that you treat everyone
as if they are stupid and will belive the line of bull you are putting out just
because you said it. The world used to look up to the CEO's of America's
corporations, but you have proven once again that we cannot trust a thing said
from most CEO's Thanks Darl for continuing the long slide downward. You Ken and
the others scam artists are one big reason America is sputtering economically.
If anyone should be outlawed, it should be you and your kind.

[ Reply to This | # ]

Ever Wonder How the Kernel Gets Built?
Authored by: grouch on Wednesday, February 04 2004 @ 01:05 AM EST
Alright, to give you artists some full body shivers and illustrate that gimp has
extremely low skill requirements, my attempt at illustrating the kernel
process:

http://edge-op.org/images/outline-tran.2-9-mal.png

Another professed non-artist going by the name of
"malifick_in_the_shell" helped, but wouldn't stamp his name on it.

Maybe somebody could fix this?

[ Reply to This | # ]

Stuff for IBM in the Harvard Video
Authored by: Anonymous on Wednesday, February 04 2004 @ 01:55 AM EST
c. 56 mins - lots of comments about infringement being identified by SCO, and
million lines attributed by IBM.

[ Reply to This | # ]

ATT vs. BSD is eerily like SCO vs. Linux
Authored by: Philip Stephens on Wednesday, February 04 2004 @ 03:40 AM EST
This is a bit off-topic, but I thought I'd post this here since we're talking
about the development process that lead to Linux, and SCO's scornful attitude
towards it.

I've been reading a book by Peter Wayner called "Free for all" which
documents the rise of the free software movement and Linux. The book is freely
available on the net, by the way, from http://www.wayner.org/books/ffa.

Anyway, this book has a very good chapter on how BSD UNIX came into being, and
the resulting lawsuit between AT&T and BSDI and the University of
California. I realise this has been covered before on these forums, but here is
a quick recap: Berkeley and AT&T did a lot of collaboration on UNIX during
the period when AT&T was not permitted to get into the computer business.
As a result, a fair amount of what was known as AT&T UNIX was in fact code
developed by Berkeley. In particular, the TCP/IP code for networking was from
Berkley.

Around the same time that AT&T was finally allowed to sell UNIX, Richard
Stallman started the GNU project, where he and a band of programmers started
writing their own versions of UNIX utilities and libraries, in the hope of one
day creating a UNIX clone that was totally free. In a delicious irony, Berkeley,
being a university, felt that they were being used by AT&T, which started
imposing substantial restrictions on the redistribution of UNIX, and charged a
fortune for the source code. Since Berkeley had contributed a lot of this code,
much of it paid for by government grants, they felt the need to get out from
under AT&T's thumb.

So what did they do? They got together a bunch of programmers and had them do
exactly what Stallman was doing, and rewrote all of the UNIX utilities, tools
and programs from scratch, as new "clean room" implementations. Then
they combed through the UNIX kernel, identified which bits were AT&T's and
which were Berkeley's, and replaced all of AT&T's bits--except for 6 files
that they had trouble untangling from AT&T's code. They then released this
as BSD UNIX (actually called Networking Release 2 at the time).

BSDI was a company that took BSD UNIX, rewrote those 6 remaining files, and sold
it for much less than what AT&T was charging for UNIX. Even though other
versions of BSD UNIX were released for free, the fact that BSDI was charging for
the OS is what got them on AT&T radar. So AT&T sued BSDI.

BSDI was able to show that apart from the 6 files they rewrote, everything else
came from Berkeley, and so the judge restricted AT&T's case to just those 6
files. So AT&T sued Berkeley as well, charging they had stolen their code
and released it as BSD UNIX.

See the resemblance with SCO vs. Linux yet? :-)

What is interesting is that the attitude that AT&T took towards a free
version of UNIX being released to the world is virtually identical to the
attitude that SCO is taking towards Linux being made freely available. The fact
that SCO "inherited" AT&T's version of UNIX just adds to the
irony.

Unfortunately for SCO, just as Berkeley was able to convince a judge that their
code was not encumbled by any of AT&T's code (even if it only lead to a
settlement and not an official judgement), SCO is going to have the same uphill
battle to convince a judge that Linux is encumbled by any of the same code. But
clearly Darl McBride has it in his head that since the case was settled and no
judgement rendered, he's going to do what his "predecessor in
interest" failed to do in the early 90's, and that's regain control of
anything that walks, talks and acts like UNIX. Well, good luck to him I
say...he's going to need all the luck he can get!

As far as the AT&T copyrights that McBride claims were stripped out in
violation of the settlement, included with the copyright notices was a
"license" that allowed the code to be used freely for any purpose
whatsoever...so even if McBride was to argue those copyrights should be put
back, it won't change the fact those files are still "unencumbered".
A fat lot of help that will be for SCO's dream of charging the same kind of
exorbitant license fees that AT&T once enjoyed with UNIX.

The settlement also had a couple of files stripped out of BSD UNIX. The book
says there are rumours these files were only removed in order to allow
AT&T's lawyers to save face, rather than because they had any AT&T code
in them. But of course, unless the settlement details ever become unsealed,
we'll never know the truth. McBride seems to think there are some terms in
there that will help SCO--I'm willing to bet there are none.

So, BSD UNIX was declared completely free of AT&T source code. Linus
Torvalds started developing Linux during the AT&T lawsuit, but never
actually knew anything about free versions of BSD UNIX being available...and so
he developed Linux totally free of any knowledge of BSD UNIX source code, let
alone AT&T source code! So how SCO believes they will be able to convince a
judge or jury that Linux is an unauthorised derivative of UNIX is beyond me.

[ Reply to This | # ]

OT: SCO 10-Q
Authored by: gbl on Wednesday, February 04 2004 @ 06:15 AM EST
Not much has been said about SCOs recently filed 10-Q (or perhaps I've missed it.) Of special interest is their research and development costs. Here is a quote

Research and development expenses consist of the salaries and benefits of software engineers, consulting expenses and corporate allocations. The decrease in research and development expense in fiscal year 2003 of $6,546,000 compared to fiscal year 2002 was primarily attributable to work force reductions. The increase in research and development expense in fiscal year 2002 of $797,000 over fiscal year 2001 was attributable to development work on our UNIX operating systems acquired during fiscal year 2001. Our research and development personnel decreased from 136 as of October 31, 2001, to 75 for fiscal year-ends October 31, 2002 and 2003.

For fiscal year 2004, we anticipate that research and development expenses will increase from fiscal year 2003 due to our recently initiated efforts to maintain and enhance our UNIX products.

For a software company to lose 61 R&D staff (45%) over two years is quite an achievement.

Then we can look at the next section and learn that their General and administrative expenses have also fallen mostly due to the fact that they got rid of 48 staff (47%) over two years.

At that rate SCO has 2 years to live.

---
If you love some code, set it free.

[ Reply to This | # ]

SCO value on ft.com
Authored by: jmc on Wednesday, February 04 2004 @ 07:17 AM EST
I do apologise for this not being on-topic on the kernel but there isn't (as far
as I can see) a current story for this to be on-topic for. As a humble
suggestion - could Groklaw be subdivided according to broad topic? Or perhaps
the current stories be kept alive longer? I'm sure there will be a need for
Groklaw when SCO is a smoking ruin and Linux's next enemy won't be quite so
inept, so we need to advance on all fronts.

If you look up SCOX on www.ft.com (the link is so horrendous that it's easier to
just give that) and go to "Consensus forecast" the recommendation is
"buy" at the top of the scale despite yesterdays large drop. Do they
know something we don't - or is it vice versa???

I don't play the stock market myself but if I did I'd be worried about seeing
advice like that on a subject I knew about in case I was following equally bad
advice on situations I didn't.

[ Reply to This | # ]

Odd entry in Changelog!
Authored by: Anonymous on Wednesday, February 04 2004 @ 08:43 AM EST
I saw the changelog, and noticed an odd entry...

Could someone research thi?

<DMcBride@thescogroup.com>
[Patch] Added SCO I.P. to Kernel so we would have a case.

[ Reply to This | # ]

Can we please work on a SCO faq?
Authored by: ra on Wednesday, February 04 2004 @ 10:32 AM EST
It seems that Darl McBride was able to convince a Linux advocate that SCO's
position may be reasonable about SCO owning AIX and having a need to press its
case.

I take this as a failure of Groklaw that should be corrected as soon as
possible.

The fact is that people who read just about every post and comment here and
follow nearly every news story have not made it easy for those who aren't as
obsessed with this issue to catch up to speed quickly.

A person who is committed enough to the Linux cause to create Knoppix cds for
all of the attendees of the speech should completely embarrass Darl when he
claims "control rights" over AIX.

But that did not happen. Tons of facts and links are buried in the archives but
only the already knowledgeable know where to find it.

We need to make an FAQ so that things like this do not happen again. And so
everyone interested knows exactly how ridiculous SCO's claims to control AIX
are. Groklaw in a small way has failed to inform. We should make sure that
does not happen again.

Please lets start.

Over at Yahoo someone started an FAQ. Maybe something in there can be used by
us here along with appropriate disclaimers and plenty of links and well edited
language.

http://quick_sco_faq.tripod.com/index.html

Once we agree on an FAQ, it should be in the top left pane so people new to this
issue can catch up quickly.

I think this is really important.

[ Reply to This | # ]

[OT, joke] Fast Forward to 2014
Authored by: Anonymous on Wednesday, February 04 2004 @ 10:34 AM EST

Look at this article: Eyewitness account of the Linux monopoly trial, it is so funny...

Washington DC, January 31, 2014 -- Riot police have finally managed to beat back the milling throng of displaced Visual Basic programmers who attacked the courthouse after Judge Cotter Kathelly announced that Linux was not an illegal monopoly and that neither Linus Torvalds nor his company, Linux Development, Inc, owed damages to former employees and shareholders of now-bankrupt Microsoft or to any of its business partners.

...

At first David Boise, attorney for Microsoft's former employees and shareholders, seemed to enjoy his turn in the limelight after his years in seclusion following his unsuccessful attempt to stop Linux development on behalf of a small, long-forgotten Utah company called "SCOrn" (or perhaps SCOre; no one seems to recall the company's exact name now, and its former CEO, Daryl McBrood, has not been heard from since he started serving his 20 year federal prison term for securities fraud back in 2006) and the disbarment that followed this ill-fated legal battle.

[ Reply to This | # ]

Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details )