backtop


Print 47 comment(s) - last by TomZ.. on May 17 at 9:52 PM

Linus Torvalds takes issue with Microsoft's claims of patent infringement

In its most recent bout with Linux and the open-source community, Microsoft claimed that Linux distributions, along with other open-source software, actually infringe on its patents. The Redmond giant claimed that 235 of its patents have been violated by the Linux kernel, Linux graphical user interface and Open Office/other open-source applications. The company went on to say that it believes it has the right to demand royalties from Linux distributions.

Despite Microsoft taking a strong stance on its argument, Linus Torvalds, lead developer of the Linux kernel, failed to blink. Talking to InformationWeek, Torvalds simply said, "It's certainly a lot more likely that Microsoft violates patents than Linux does. If the source code for Windows could be subjected to the same critical review that Linux has been, Microsoft would find itself in violation of patents held by other companies."

Torvalds went on to reason that although "fundamental patents" for basic operating series were already filed in the 1960s, the protection for those patents have been expired for quite some time. He also suggested that Microsoft name the patents that it believes are being violated in order for its claims to actually be tested in court.

"Naming them would make it either clear that Linux isn't infringing at all (which is quite possible, especially if the patents are bad), or would make it possible to avoid infringing by coding around whatever silly thing they claim."



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Closed source code
By jtesoro on 5/16/2007 9:38:50 AM , Rating: 5
That comment is very interesting and has a chance that it actually holds water. Can one make a claim and sue a software company (Microsoft or otherwise) if their source code isn't publicly visible or available?




RE: Closed source code
By jtesoro on 5/16/2007 9:40:52 AM , Rating: 2
Oops, I meant that in the context of the quote "If the source code for Windows could be subjected to the same critical review that Linux has been, Microsoft would find itself in violation of patents held by other companies."


RE: Closed source code
By lukasbradley on 5/16/2007 9:59:44 AM , Rating: 3
Absolutely. If you own the patent, you don't even have to be a producer of code.

One of the reasons intellectual property and software patents have become so cantankerous is the advent of third-party firms (law firms, mostly) purchasing older patents from expiring companies. These third-party firms then pursue damages from other firms who have produced something that is close in functionality to what the patent describes.

Owning a patent, in many ways, allows you to legally claim you own the *ability* to exercise an idea or concept, even if you choose not to do so. If other entities so desire, they can license that ability from you. Otherwise, they are subject to legal ire.


RE: Closed source code
By jtesoro on 5/16/2007 10:19:03 AM , Rating: 2
Interesting. In this case, the patent is in the context of source code, so I assume that it involves a certain way of coding. For example: "using a 'for' loop to search through an array" (let's assume that's patentable). How can one make an assertion that someone violates the patent if you can't see their source code?


RE: Closed source code
By TomZ on 5/16/2007 11:31:37 AM , Rating: 2
I agree, and I think that, as a practical matter, it is practically impossible to write code that doesn't infringe on software patents. You would have to do patent searches for all new code you write - and who does that?


RE: Closed source code
By jtesoro on 5/16/2007 11:54:18 AM , Rating: 2
I think there has to be a way to protect truly novel or unique coding innovations (though how to define "truly novel or unique" will be difficult at best). On the other hand, as time goes on, programmers will likely stumble on the same innovations eventually. This will make it practically impossible to write non-infringing code, as you say.

Given all that, I think Microsoft should back off on these royalty claims they're making. You know, kinda using the elementary school concept of "fair's fair".


RE: Closed source code
By TomZ on 5/16/2007 1:01:14 PM , Rating: 1
I don't know if Microsoft should back off or not. To the extent that Linux contains infringing code and it is responsible for decreased sales for Microsoft and other companies/individuals are profiting from it, there is a chance that such claims would make sense.

But that is a long chain of conditionals that all have to be true, and I think it is more likely that Microsoft is reminding people that Linux is maybe "free" like free speech but not "free" like free of IP ownership problems for its users, especially corporate users.

I'm not advocating strategies that border on FUD, however, we have to remember that companies are making money off of Linux, and they need to be held to the same standards as everyone else. After all, there should be no free ride for companies that make money off of "free" software.


RE: Closed source code
By leexgx on 5/16/2007 9:42:22 PM , Rating: 2
heh i guess that did not go down well with DT users (thay should make the Voting system in here like Digg.com as some users can get some realy low - numbers)

companies that sell linux are norm selling an service for and there is norm None free software that is sold with it that is not Open source


RE: Closed source code
By TomZ on 5/16/2007 10:49:14 PM , Rating: 2
Some people can't handle the truth. Oh well.


RE: Closed source code
By Scrogneugneu on 5/16/2007 8:11:30 PM , Rating: 2
quote:
You would have to do patent searches for all new code you write - and who does that?


Lawyers. Why?


RE: Closed source code
By mars777 on 5/17/2007 12:10:48 AM , Rating: 2
In Europe they just dont.
Software patents on "schemes, rules and methods for performing mental acts" are not allowed.

So you cannot patent a method for "doing something" in code.
You can patent a product only.

And that is fair.

Everything in software if "obvious" to programmers. it's not obvious to morons at the departments for patents.

America has gone wrong with software patents, and even if it would be fair to patent menthal acts in programming it would not be cleaver.

Why?

Because programming has a rate of progress and speed of developing x times faster than other developments.
So if a patent on electronic engeneering lasts 20 years, a patent for software cannot.
You would hinder progress for 20 years!!!

Patent a recursive function call to solve a treeview search and you have just killed 30% of all programming on earth ... (it's just an example, its obvious that this has allready been done in the past and is not patentable)


RE: Closed source code
By theapparition on 5/17/2007 11:31:57 AM , Rating: 2
I think everyone is getting a little off base with the patenting software "code" example. Patenting "FOR" loops and "IF" functions are part of the software compiler that was purchased and is licensed legally for you to create what you want, with any sort of code.

However, from reading a little about Microsoft's claim, it has nothing to do with the code, rather, it is that they don't like how the Linux shells emulate the "look and feel" of Windows. If you remember back in the day when Apple sued Microsoft over Finder. It had nothing to do how it was coded, but how it was similar in "user" operation.

That's the point of Microsoft's current sabre rattling.


RE: Closed source code
By theapparition on 5/17/2007 11:31:57 AM , Rating: 2
I think everyone is getting a little off base with the patenting software "code" example. Patenting "FOR" loops and "IF" functions are part of the software compiler that was purchased and is licensed legally for you to create what you want, with any sort of code.

However, from reading a little about Microsoft's claim, it has nothing to do with the code, rather, it is that they don't like how the Linux shells emulate the "look and feel" of Windows. If you remember back in the day when Apple sued Microsoft over Finder. It had nothing to do how it was coded, but how it was similar in "user" operation.

That's the point of Microsoft's current sabre rattling.


RE: Closed source code
By Dactyl on 5/16/2007 2:16:55 PM , Rating: 2
How can one make an assertion that someone violates the patent if you can't see their source code?

If you suspect they are violating your code, first you notify them of your patent and demand that they say they are not violating it. (If they agree they are infringing, then you can settle with them.) If they won't settle, you can sue.

If Sun sues MSFT, Sun doesn't get to see MSFT's source code, but Sun's lawyers do. And when I say "Sun's lawyers," I'm talking about a team of lawyers employed by a law firm that Sun has hired to represent it. Sun's in-house lawyers (lawyers directly employed by Sun) can't see MSFT's source code either.

So the lawyers Sun has hired for this case, and their experts, get to see MSFT's source code. Everybody has to swear up and down that they will keep it secret, with serious consequences if they leak it.


RE: Closed source code
By carage on 5/17/2007 5:14:43 AM , Rating: 2
And the lawyers probably have to hire another team of programmers, I doubt there are many lawyers capable of analyzing source codes by themselves.


RE: Closed source code
By KasjaajaK on 5/16/2007 1:50:12 PM , Rating: 2
Could Microsoft potentially trying to muddy the waters in anticipation of future patent infringement case?

It seems highly unlikely that the immense MS canon doesn't infringe on the GNU license in at least a hundred ways. It also seems to Microsoft's advantage to get in the first shot, while establishing in court the pragmatic fact that it's long been impossible to code without doing something that's been done before.

Such a move could help to insulate against a future case of deliberate plagiarism by providing plausible deniability.


RE: Closed source code
By borowki on 5/16/2007 4:28:05 PM , Rating: 2
quote:
It seems highly unlikely that the immense MS canon doesn't infringe on the GNU license in at least a hundred ways.


GPL is based on copyright. Unless a programmer at Microsoft actually copied and pasted code from a GPL'ed software, there would be no violation.


RE: Closed source code
By TomZ on 5/16/2007 5:15:33 PM , Rating: 2
Just curious - has any court ever dealt with whether downloading and making use of GPL-licensed open source actually legally binds the user to that license?

What I mean is, I remember a while back (years ago) it being stated that shrink-wrap EULAs were basically untested in courts as to whether they were legally binding. In other words, just because I tore open and installed the software may or may not mean that I am legally bound to the agreement.

Just to be clear, I'm not talking about right/wrong; I'm just questioning the legal effectiveness of the license.


RE: Closed source code
By borowki on 5/17/2007 5:17:57 AM , Rating: 2
As far as I know the GPL hasn't been tested in court. There is really no practical reasons to challenge it, since showing the GPL to be unenforceable would not strip a software licensed of its copyright protection. This point is explained in the text of the GPL itself:

quote:
You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works.


RE: Closed source code
By NoSoftwarePatents on 5/17/2007 1:59:43 PM , Rating: 2
Actually it *has* been tested in court in some ways. Look up an anti-GPL character named Daniel Wallace. In short, the lawsuit against the GPL did not go well for him.

Here is a link to get you started...

http://www.fsf.org/news/wallace-vs-fsf

This doesn't mean every aspect of the GPL is highly enforceable, but it's not worthless either.


RE: Closed source code
By TomZ on 5/17/2007 2:58:16 PM , Rating: 1
Interesting information - thanks guys.


RE: Closed source code
By borowki on 5/17/2007 6:26:18 PM , Rating: 2
A curiosity, though it says little about the enforceability of the GPL. The suit was probably thrown out on procedural ground: Mr. Wallace simply has no grounds to sue.

As it was express yet again in the linked page, the terms offered by the GPL is better than nothing. If you refuse to accept it, you end up with the latter.

The only legal question of interest, I think, is whether those who license their software under the GPL are obliged to follow its terms. Can the author of a GPL'ed software "un-GPL" it--i.e. stop people from using it? Given that the GPL is not a valid contract, the answer is probably yes.


bullseye
By dome1234 on 5/16/2007 9:37:57 AM , Rating: 5
quote:
Torvalds simply said, "It's certainly a lot more likely that Microsoft violates patents than Linux does. If the source code for Windows could be subjected to the same critical review that Linux has been, Microsoft would find itself in violation of patents held by other companies."


MS can examine linux code in detail while the opposite can't be said.




RE: bullseye
By codeThug on 5/16/2007 10:36:22 AM , Rating: 1
Who keeps mod'ing these posts down?

Are there that many MS "tools" out there acting like Scientologists? sheesh...


RE: bullseye
By smitty3268 on 5/16/2007 11:01:12 AM , Rating: 3
"If Microsoft were to actually tell people what patents they claim we violate, we could either laugh in their face and show prior art, or just show them to be obvious, or we could do things differently,"

Torvalds


RE: bullseye
By Oregonian2 on 5/16/2007 1:51:23 PM , Rating: 2
Although being obvious is theoretically something to make something non-patentable, I've found that it doesn't slow patent awards down any.

But it's doubtful if Microsoft would want to list the patents because they'd be worked around and code made clean as a whistle which wouldn't be useful to Microsoft. Would effectively make Linux a Microsoft certified clean-of-violations product. Last thing Microsoft wants. They're taking a risk as it is because that could be a result if things aren't managed adequately.


RE: bullseye
By borowki on 5/16/2007 4:43:13 PM , Rating: 3
quote:
But it's doubtful if Microsoft would want to list the patents because they'd be worked around and code made clean as a whistle which wouldn't be useful to Microsoft. Would effectively make Linux a Microsoft certified clean-of-violations product. Last thing Microsoft wants. They're taking a risk as it is because that could be a result if things aren't managed adequately.


If the Open Source people are so sure of non-fringement, they can seek a declaratory judgement from the court stating that there's no violations. From what Microsoft has said so far, I believe they're entitled to do that.

Whether a patent can be worked around is pretty much a moot point, since the damage would have already occurred, allowing Microsoft to seek compensation from Linux users. Even a single infringement, if proven, would effectively discredit Linux as a business solution.


RE: bullseye
By TomZ on 5/16/2007 5:06:38 PM , Rating: 2
quote:
If the Open Source people are so sure of non-fringement, they can seek a declaratory judgement from the court stating that there's no violations

I think that would be the proverbial "picking a fight you can't possibly win." As you said, even a single infringement tarnishes Linux, and considering the number of patents involved plus the inability of anyone outside of Microsoft to evaluate the details, most likely Linux would be found to have some problems. Once proved, then Linux users could be held liable. That doesn't sound like a good path for the Open Source folks to go down.


RE: bullseye
By Oregonian2 on 5/17/2007 4:46:05 PM , Rating: 2
There is a better path if Microsoft insists on making Linux go away through FUD instead?


RE: bullseye
By TomZ on 5/17/2007 9:52:03 PM , Rating: 1
Sorry, I'm not sure I understand what you mean.


RE: bullseye
By TomZ on 5/16/2007 11:29:30 AM , Rating: 3
Linus is right about that - if Microsoft were to make its source code openly available, there would almost certainly be a large number of new lawsuits challenging them. The same is probably true for any company, since it is practically impossible to write software without unknowingly infringing on existing software patents.

Your statement is right, but it is how it should be, given the above. And I don't think there is an expectation of symmetry in the two cases since they operate under completely different IP protection models.


RE: bullseye
By MatthewAC on 5/16/2007 1:02:03 PM , Rating: 2
Only one word can be said in this situation, pwn3d :).


Confused
By ninjit on 5/16/2007 1:59:34 PM , Rating: 2
My first impression after reading the article was that MS was complaining about code within linux that was taken directly from them (e.g. reverse engineered or something).
It reminded of the whole SCO debacle.

But from all the comments it seems that I'm mistaken and that they are complaining about ways of coding.

I'm not exactly sure how something like that is patentable - pretty much all "features" of coding are governed by the language you choose.
So C++ and java promote object-oriented programing (as opposed to functional program like in C or Fortran), and so treating a display device, or printer, or mouse as an object that interact with one another is a natural extension of the language.
Pretty much anything you write up can be viewed that way.

I assumed that was one reason companies patented processes and algorithms in general terms, otherwise If MS owns a patent on say some unique way to read and display images in C++, someone else could write the same function in java or fortran, and claim its different because the underlying code is different.

So is MS complaining about actual code, or general algorithms/processes to which they hold patents?




RE: Confused
By wakingxeden on 5/16/2007 2:08:37 PM , Rating: 2
quote:
In fact, according to Smith, the Linux kernel itself violates 42 Microsoft patents. Worst yet, Linux graphical interfaces such as KDE violate another 65 patents.


This statement leads me to believe it is the actual code and not the way it was processed.


RE: Confused
By TomZ on 5/16/2007 2:42:27 PM , Rating: 2
"The code" would be covered by copyright. "What the code does" would be covered by patent. Both could also be trade secrets.


RE: Confused
By smitty3268 on 5/16/2007 2:45:33 PM , Rating: 6
Wrong. A patent can not be a trade secret, because you have to file lots of paperwork explaining exactly what it does in order to be granted a patent. The details of exactly how it is achieved (ie source code) might be kept secret, but the technology being patented can't be.


RE: Confused
By TomZ on 5/16/2007 2:55:54 PM , Rating: 2
You're right - I didn't mean something could patented AND a trade secret; I meant OR. Trade secrets rely on something being kept "secret."

Copyright is another story...


RE: Confused
By smitty3268 on 5/16/2007 2:43:52 PM , Rating: 4
You can't patent actual code, just processes or ideas. Code itself is subject to copyright law, which just means you can't copy it - which isn't a problem since that code isn't released to the public anyway.

With patents, the details of how you do something doesn't matter (C++ vs Java), it just matters what the result is. (implementation of the FAT filesystem)

Some of it could be reverse-engineered stuff, but I get the feeling that they are mostly talking about more general stuff, that OSS advocates would claim is obvious and therefore not properly patentable. Stuff like virtual memory, sudo, etc. Torvalds says all the really fundamental stuff in the Linux kernel is from the 60's so the patents on it would have run out by now anyway.


Patent Litigation
By PolPot on 5/16/2007 1:01:44 PM , Rating: 5
All someone has to do is come up with a good faith basis for alleging patent infringement by Microsoft and then they can probably issue a subpoena duces tecum for the source code to discover whether a patent has been infringed or not. Based on the technical expertise and time needed to review that much code, an in camera inspection would be impractical. Microsoft should STFU before it invites a counterclaim that can stick and opens its code up to legal discovery proceedings.




RE: Patent Litigation
By MrJim on 5/16/2007 1:32:08 PM , Rating: 2
PolPot>Excellent post, i agree with you, maybe MS legal divison has some secret weapon, or is it that MS is starting to get seriously afraid of Linux? Im still waiting for MS Office 2007 for Ubuntu myself.


RE: Patent Litigation
By wakingxeden on 5/16/2007 1:35:10 PM , Rating: 2
quote:
All someone has to do is come up with a good faith basis for alleging patent infringement by Microsoft and then they can probably issue a subpoena duces tecum for the source code to discover whether a patent has been infringed or not. Based on the technical expertise and time needed to review that much code, an in camera inspection would be impractical. Microsoft should STFU before it invites a counterclaim that can stick and opens its code up to legal discovery proceedings.


+1

This might bring a chance to see what MS is really up to. If said happened it may put big dent in MS and their reputation.


RE: Patent Litigation
By TomZ on 5/16/07, Rating: -1
RE: Patent Litigation
By leexgx on 5/16/2007 9:46:24 PM , Rating: 2
torll ?


An enlightening exchange!
By Maxmars on 5/16/2007 10:38:58 PM , Rating: 2
I very pleased to have discovered this site. Impressive content.

Not being an expert by any measure, would someone be so kind as to comment on the following: Mathematically speaking, there are a finite number of possible valid constructs within the code (any code). Then doesn't logically follow that it will eventually be possible to identify a patent for everything possibly useful 'snippet' of code within the code-base? I mean to ask, is this a precedent-setting tactic? Would it be in MS' best interest to 'throw the first rock' so-to-speak, perhaps foreseeing a vulnerability in their own position?




RE: An enlightening exchange!
By TomZ on 5/17/2007 8:57:20 AM , Rating: 1
If patents were free, that might be a viable strategy. But it takes time/money to research existing patents, write up the patent copy, prepare the applications, etc. So there is a cost/benefit tradeoff that needs to be considered in each case.


its ok to steal, just dont copy
By rika13 on 5/17/2007 8:44:17 AM , Rating: 2
actually, code can be both patented and copyrighted since it both an invention and a work of "art"; trade secret means that it is kept within the company (since patents are public, look up 10NES on wikipedia, patent 4,799,635; the library of congress is supposed to have copies of all copyrighted work [since copyright is for the public interest, not just money making]) and is probably the strongest protection

and its perfectly legal to steal ideas, not the invention; the most famous is when m$ stole the GUI from apple, who stole it from xerox




RE: its ok to steal, just dont copy
By TomZ on 5/17/2007 9:03:14 AM , Rating: 1
quote:
and its perfectly legal to steal ideas, not the invention; the most famous is when m$ stole the GUI from apple, who stole it from xerox

Your post was pretty good right up until that ridiculous statement. If you were involved in computers at the time, the trend was for all computers to go from text to graphics. The only thing that slowed it down was the availability of suitable hardware. There were lots of people and companies working on this.


"We shipped it on Saturday. Then on Sunday, we rested." -- Steve Jobs on the iPad launch














botimage
Copyright 2012 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki