Linus Responds To Microsoft Patent Claims

An anonymous reader writes "Linus Torvalds has a sharp retort to Microsoft executives' statements in a Fortune article that Linux violates 235 Microsoft patents. In an emailed response to InformationWeek's Charlie Babcock, Torvalds writes: 'It's certainly a lot more likely that Microsoft violates patents than Linux does.' He added: 'Basic operating system theory was pretty much done by the end of the 1960s. IBM probably owned thousand of really "fundamental" patents... The fundamental stuff... has long, long since lost any patent protection.'" Torvalds also commented on Microsoft's stated intention not to sue Linux users: "They'd have to name the patents then, and they're probably happier with the FUD than with any lawsuit."

Comments

Sad or Telling?

[u-bend]

Is it kind of sad that such retorts are necessary? Or is it telling, that M$ is continues to resort to such SCO-like tactics?

Re:Sad or Telling?

[cyphercell]

I hope that it's telling, after all Microsoft is essentially following tactics from IBM a company that is not, primarily in the software industry. They have essentially hired IBM's patent lawyer, Marshall Phelps

Re:Sad or Telling?

[MightyMartian]

I think we all know that the kernel is probably safe; except for some of the drivers. I'll wager that Microsoft is eying FAT and NTFS. Still, it's awfully easy to fix that. Just distribute a kernel without those drivers in the source, and just let someone outside the US distribute the patches, compiled modules, as well as compiled kernels. Is Microsoft seriously going to demand that everyone turn over their kernels to check whether the FAT file system support is compiled into it?

The problem here is not that MS would ever dream of going to court, it's that the FUD could be very effective at slowing adoption. I suspect that if anything, it's OpenOffice that would feel the wrath of being dragged into court. Going after the kernel is ludicrous, and would likely turn up absolutely nothing.

Re:Sad or Telling?

[Anonymous Coward]

Is it kind of sad that such retorts are necessary?

Unfortunately, the damage is done. I work for a large financial organization that was *just* venturing outside of Microsoft operating systems and the lawyers sent out a notice today that we are to remove all traces of "open source" software, effective immediately.

I suspect that lots of organizations were in such a boat and Microsoft played their cards accordingly.

Re:Sad or Telling?

[rainman_bc]

I'll wager that Microsoft is eying FAT and NTFS I wouldn't be surprised if they've somehow tried to patent SMB/CIFS also... (Which incidentally is shipped with SCO Unix also.)

Re:Sad or Telling?

[cayenne8]

"Is Microsoft seriously going to demand that everyone turn over their kernels to check whether the FAT file system support is compiled into it?"

Is compatibility even an issue here? I thought you could reverse engineer or the equivalent thereof, to make systems compatible. Isn't this even one of the tennents of the DMCA? I know that isn't really patent related, but, is it against the patent law to interoperate with another system?

Re:Sad or Telling?

[Dan Ost]

Why not take some time to write a well reasoned response that you can send up the management chain and to the legal department. Point out how your company is currently benefiting from open source and how painful it would be to replace the open source currently in your infrastructure.

This is an opportunity to educate. Don't waste it.

Re:Sad or Telling?

[kimvette]

Not practical, considering that every USB "drive," flash card, and external drive ships Fat32 formatted. However, FAT is >25 years old now, so any patents they MIGHT have on it are either long expired or invalid.

Re:Sad or Telling?

[Hal_Porter]

Microsoft was writing Windows 3.1, an important upgrade to the hugely popular Windows 3.0. In September 1991, a plan was hatched to use this upgrade to kill DR DOS. In an email discovered by the Dept. of Justice, the head of Windows development and Microsoft VP David Cole wrote, "aaronr had some pretty wild ideas after three or so beers--earleh has some too." The plan was to plant code into Windows which would "put competitors on a treadmill" and cause the system to "surely crash at some point shortly later." In order words, Windows would intentionally bomb if it detected DR DOS.

At this time, many computer vendors were considering switching from MS-DOS to the superior, cheaper DR DOS. Microsoft was especially concerned about IBM. Wooing these PC vendors was crucial to the future success of DR DOS, as was the good will of "early-adopters" (i.e., technically savvy users who drive new trends in the computer industry).

These vendors and early-adopters were also the same people who received a Christmas "beta" pre-release of Windows 3.1. They discovered--to their horror--that using DR DOS would cause vague system errors to pop up in Windows 3.1; they dumped DR DOS in droves. By the fourth quarter of 1992, sales of DR DOS had dropped from $15 million to only $1.4 million. The once mighty competitor became a has-been and was sold to Novell and later Caldera.

This is the famous AARD code, written by aaronr

http://members.ozemail.com.au/~geoffch@ozemail.com .au/editorial/aard/firstpublic.htm

It's actually not as dastardly as you think. First it was disabled in the release code, it only warned people in the Beta. Secondly, Windows was really mixed up with Dos. For example, it called an undocumented function to get a pointer to the Dos data segment and modified that data segment. It also patched parts of the Dos code segment - it literally rewrite the code in memory. So most likely it would have issues running on any Dos that wasn't an exact binary copy of MSDos, and only a relatively recent Dos at that. I.e. Windows 3.x has code like this

code_ptr = NULL;
data_ptr = NULL;
 
// get the code and data segments. Both functions are undocumented
data_seg = GetDosDataSeg();
code_seg = GetDosCodeSeg();
 
if (DosVersion == 6.0 )
  {
  code_ptr = FARPTR(code_seg:aaaa);
  data_ptr = FARPTR(data_seg:bbbb);
  }
 
if (DosVersion == 5.0 )
  {
  code_ptr = FARPTR(code_seg:cccc);
  data_ptr = FARPTR(data_seg:bbbb);
  }
 
if ( code_ptr && data_ptr )
// Patch away.
else
// shit unrecognized Dos version. Disable some features or die with an error?

Re:Sad or Telling?

[Creepy]

I posted something about this on the last thread - SMB/CIFS has a couple of obsolete Microsoft patents that Samba implements, but not as documented in the patents themselves (they would not work in a UNIX-like OS). The main part of Samba is based on specs designed by the Storage Industry Network Association, which MS used more as a guideline than how they implemented it. Here's an article about it, and also see Samba's web site.

The only other patent I know of on CIFS is not owned by Microsoft, it's a Cisco patent

MS probably still argues that those two patents are being violated and MS even spread some FUD by issuing a license for using CIFS on other OS's but excludes the GPL (Samba's license).

Things that are obvious, but Microsoft has patents on, that I'm aware of (thank you, bookmark file):
patent on RSS feeds
FAT patent 5579517 (which I believe has now been rejected as obvious after appeal and my ref was link-dead)
Spam filtering
IsNot in BASIC
or how about this one, which is basically sudo
or this one which would be violated as far as I can tell by a Linux OS module updated over an https connection, though I think it would also need to include verification like an md5 checksum to fall under that patent.

and a couple that I don't think would affect Linux:
a patent that is basically the same as XUL, but for Windows only.
a patent on this one on learning, which is broad and vague - see this guy's response I found in a search which explains the stupidity better than I could (my original link is again dead - I need some housecleaning).

Linus, Linux, IBM, and patents

[symbolic]

I'm not clear how IBM could own thousands of patents back in the 1960s- at that time it was clearly understood that software is a non-patentable "invention". Oh how I wish this common sense would be restored.

Heavens, the breaking news!

[FatherBusa]

Linus's comments strike me as indistinguishable from the hundreds of comments we've had on Slashdot on this issue in the last 48 hours.

How about we wait until there's some actual news on this story?

Re:Heavens, the breaking news!

[MightyMartian]

The problem is that Linus's comments hold no more weight with XYZ Inc.'s legal department, who are by now recommending that the IT department's move to transfer file sharing from an expensive Server 2003 farm to license-free Samba network be suspended indefinitely. Sure, the guys in the IT department know as well as Linus, you or I that Microsoft is playing a dirty game, but Microsoft doesn't give a shit about Linus, you or I, but about the lawyers and officers of the companies that are actually considering Linux boxes with Samba and OpenOffice running on them.

Do you think it's an accident that OpenOffice was the only OSS project specifically named? Put this in the perspective of MS fighting various governments to stop OpenOffice file formats from becoming the defacto document standards. Hey hey Mr. Massachussetts, that document standard your talking about, well the baseline software that produces it violates a bunch of our patents. Now the talking heads that get into public meetings will have a new and very potent tool in the arsenal, the threat of legal repurcussions if a switch to open software is made.

Re:Heavens, the breaking news!

[Daniel Phillips]

Insightful perspective, except this is a little too glib:

That doesn't make him smarter than the rest of us (any more than Linus Torvalds or Tim Berners-Lee smarter than the rest of us. I know first-hand that Linus is a brilliant engineer, a legendary bug hunter and more skilled with the English language than I am. The last being a pretty good trick since English is not his first language, or second even. The thing I hope to see before the end of this Mordor-against-the-world story: a head to head debate between Billg and Linus, if Billg has the nerve. I know Linus does.

Why not start debunking FUD now?

[digitalderbs]

This is an innocent question. If the OSS community is ready to debunk these patents, do we really need Microsoft to reveal which of the 235 patents/infringements they're talking about? Couldn't we start a site/database that organizes all of Microsoft's patents and start documenting prior art and such for each. The patents themselves aren't hidden :

Microsoft's patents (6723 patents)
Microsoft's UI patent (155 patents)
(for example)

Why not start debunking the FUD to prove how spurious their claims are? Is it because this would be too much work? (Admittedly, 6723 >> 235)

If it could it would

[Recovering Hater]

But Microsoft can't actually bring the heat on any of the claims. For example: Gutierrez also said Microsoft is not likely to publicly list which specific patents it believes are infringed upon by open source software. "We're not going to have a discussion publicly with that level of detail," he said.

It seems to me that it is just more noise from a blowhard company that is losing steam in the arena of operating systems. Sound and fury signifying nothing. Too bad the general public won't recognize it for what it is.

Re:If it could it would

[init100]

"We're not going to have a discussion publicly with that level of detail," he said.

This and other similar quotes were what turned on my FUD warning light. The fact that they don't want to be specific is quite telling. They just want a dark cloud to hang over Linux, so that people wouldn't look that way when they are disappointed by Microsoft's latest offering.

I know MS trolls and fanboys won't like this, but the fact remains: If Vista and Office 2007 were booming and replacing Linux and other F/OSS in the marketplace, why care about it at all? Why risk their in many circles very positive corporate image to FUD Linux and other F/OSS with patent threats? No, this probably means that the uptake of Microsoft's latest products is far below their expectations, almost to be considered a flop. Flops they spent many billions of dollars developing.

Past infringement?

[KarmaMB84]

Assuming MS really does have valid patents, how could just rewriting the code prevent Microsoft from seeking royalties for past infringement? Why does Linus think that Microsoft can't have patented anything that might be in the Linux kernel now just because basic operating system theory was done in the 1960s? Surely Linux 2.6.x is more modern than 1960s technology, right?

Re:Past infringement?

[MightyMartian]

Software developers do not read patents.

And that's the saddest thing about software patents. I'll wager you won't find any non-trivial code that doesn't trample on some software patent. I doubt you could write anything over a few thousand lines long that wouldn't violate some software patent by somebody. If this continues too much longer, law schools will have to start turning out lawyers who can read C, C++, Java, C# and x86 assembly.

Re:Schwartz (Sun) responds

[skoaldipper]

Great insight, but a tad bit Utopian I think. If linux were not commercially usable, which it most certainly is, there would be no money trail to follow, nor a suit for a suit (so to speak).

First, they don't, and secondly they can't sue

[postbigbang]

Microsoft will shortly have to expose the patents they believe are violated. That will be their undoing.

If they sue, it will be against major distro makers that aren't otherwise indemnified (like Novell, it's believed). If they sue end users, they will rue the day, and it will become like the RIAA except that there'll be alternatives to software, where there is a monopoly on music distribution.

Who will suffer? Microsoft. They're already in trouble with sliding OS sales because they can't make a quality product because of decisions made more than a decade ago that are architectural in nature. Would a Linux user be sued? Hardly. A distro maker? Sure. And how much money are they going to get? Not very damn much-- that's the interesting part. It'll be like SCOx vs IBM all over again. Watch the smoke, watch the mirrors.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:constitutional lawyers?

[whoever57]

- Science must be making progress in order to be promoted. DRM is regression of rights, thus it cannot be protected. - Arts must be useful to be protected. I doubt entertainment can be considered "useful" in the way that was meant when the Constitution was written.

You are applying logic to laws. Never works. In this case, I think you will find that the sole judge of how the laws meet the goal is Congress. I think this came out in the challenge to the Sonny Bono/Mickey Mouse copyright extention. The SCOTUS decided that it was up to Congress to decide how best to promote the "useful arts", and if Congress felt that it was best achieved by locking up our cultural heritage, then so be it. One would also expect that SCOTUS would hold that it was up to Congress to decide what constitutes a "Useful Art", since that it a preamble to the actual phrase that authorizes Congress to create copyright laws.

Essentially what I am saying is that the part "To promote the progress of science and useful arts" is meaningless and the only important part of that section is: "by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries;"

Sue for FUD?

[eille-la]

Is it possible for an organization like IBM (who pay to keep the Linux kernel in development) to sue Microsoft because they spread misinformation on purpose about GNU/Linux which is bad for it's reputation?

I think it is possible to sue someone (at least in Canada) who act to make your reputation go dirty.

Is there a law for that?

Duty to Mitigate

[EricTheGreen]

I would be interested in comments from any lawyer-types on whether the duty to mitigate applies now that Microsoft has publicly announced knowledge of alleged patent violations. If they don't pursue specific action now, but defer it to some later time, could a defendant mount a defense based on Microsoft's failure to address the violation in a timely fashion? Or does this not apply in IP law?

Accusation - substantiation = slander

[KFury]

If Microsoft publicly declares that Linux violates patents but won't disclose those patents it seems Microsoft would be guilty of slander. Clearly they're trying to make businesses think twice about buying into OSS solutions by giving the perception that those OSS solutions could be illegal.

Just raising the threat is enough to swing business into MS's camp. Without a single company to take point and sue MS for slander they'll get away with it.

I got questions...

[Anonymous Coward]

I'm not familiar with the US legal system, but here goes...

1. Can't FOSS or some other Open Source-heavyweight counter-sue M$ for slander in the likely event their allegations can't hold water? We can't sue for lost profits (there isn't any), but instead propose some kind of gag restricting M@ from making likewise allegations again? I mean, lawyers like money, and this seems like stealing candy from a baby (who throws chairs).

2. Why doesn't FOSS-supporters construct websites touting all the patents M$ breaks, or patents that M$ claim as their own although prior art exists? It wouldn't be FUD, it would be stepping up to the table and unveil the shitty patent-practice M$ has. I mean, that's what patent-databases are there for, right?

3. In essence, wouldn't FOSS/The Open Source community be "the bus to get onto" for lawyers? It seems like an infinite generator of easy-to-win lawsuits waiting to happen because no-one steps up to the plate due to the de-centralization of open source. The lawyers would make money, the community would get M$ off their backs, everybody except M$ goes home glad.

IANAL, but gee, If Naomi can do community-service for flinging shit at her maid...

Re GoodWay:Sad or Telling?

[OldHawk777]

Also, M$ is possibly seeking mutual indemnification from theft of property prosecution. If M$ can press the Linux Foundation and FSF-GPL+ to provide such a settlement, then they are safe from being financially raped, forced to release M$ products to F/LOSS-GPL, or destroyed by forcing removal of all OSS-GPL copyrights protected code from M$ products.

So, who will be the first to flinch in this obvious game of chicken/standoff. I suspect, that M$ is looking to be another dickless empty SCOrotum.

I ain't good enough with code, but I will continue to put some loss money on the Linux Foundation, FSF-GPL, EFF .... It is a very good idea to start loading the evidence weapon that will put a through&through hole in M$. I mean we may as well help them commit economic suicide. Work with the Linux Foundation folks ... I will be sending them a couple hundred more $ this year for the TEK-War. I hope I can get a hat that says I supported the F/LOSS troops in battle against the M$techno-fascist.

!HAVEFUN!