Author of Linux Patent Study Contradicts Ballmer

An anonymous reader sends us this EWeek story, following-up on the recent Linux patent scare. The author of the patent study is contacted, and says, "Open source faces no more, if not less, legal risk than proprietary software. The market needs to understand that the study Microsoft is citing actually proves the opposite of what they claim it does."

Comments

Math Error in Article

[Noksagt]

In fact, the study said Linux potentially violates 283 software patents, not "over 228" as Ballmer said in his speech.

Last time I checked, 283>228. Please also note this quote, which provides the best summary:

"The point of the study was actually to eliminate the FUD about Linux's alleged legal problems by attaching a quantifiable measure versus the speculation," he said. "And the number we found, to anyone familiar with this issue, is so average as to be boring; almost any piece of software potentially infringes at least that many patents.

Realized that: www.fudfactory.com

[Anonymous Coward]

Some people have realized that part about the fud factory.

I for one.................. overlords.

:) Geena

Re:Liability of implementors of patented ideas

[happyemoticon]

If I implement a piece of software that is already patented and put it in the public domain, can I be sued for this?

Strictly speaking, yes. Patents protect the abstract concept of something. If the code was just copyrighted, you could release a competing open source product though. That's why companies like patents.

Can someone write a free implementation of a patented idea just for the sake of other people doing research on the idea (e.g timing and comparing certain patented algorithms, etc.)? This doesn't seem different from publication of said idea/algorithm.

IANAL, but they probably have permission. Publishing firms have armies of lawyers to prevent infringement. However, whenever you are about to use something patented, ask yourself, "Am I making a profit based on this person's idea?" In the case of a book which touts X's patented sort routine or Y's mipmapping algorithm, the patent holder would probably concent because it's like free advertising. If you release a GPL'd implementation of something, you are still violating their patent. You'd have to talk to the company first and get their expressed, written concent.

Re:except...

[jkabbe]

...or maybe the patent owner will decide to:

1. Collect royalties directly from the end user

Patent's don't work that way. Patents prohibit you from manufacturing or selling the patented invention. If you're just using one you got from someone else they can't come after you. They would have to go after the person who gave/sold it to you.

2. Issue an injunction to the end user to cease using their IP - immediately.

Courts issue injunctions, not patent-holders. What the patent-holder would do is send a cease-and-desist letter. If it is ignored then the patent-holder would have to go to court.

And how many patent lawsuits has Microsoft had...

[mikael]

And how many patent lawsuits has Microsoft been involved in:

Microsoft, Tiscali sued over European download patent

Sun, Microsoft settle suit in billion dollar pact

Microsoft settles Intertrust patent lawsuit

Microsoft settles suit with Immersion

Microsoft settles 1999 Patent Infringement Case

Re:Liability of implementors of patented ideas

[Halo1]

Whether you try to sell it or use it for comercial purposes has little to do with the matter. What matters is whether you are distributing a description of the invention, or the invention itself.

Yes, obviously the problem is here that when it comes to software the task of drawing the line between the two is a real bitch. Maybe impossible.

Indeed. Depending on the language (e.g. C vs English) the description is the invention or not. It's one of the things that illustrates very well why patents are ill-suited to software. And possibly also one of the reasons they don't want source code in the patents, otherwise one patent description could infringe on another :)

Re:Really! Really!

[Halo1]

And while we're at it, we'll do it under a license that has never been tested or sanctioned by a court of law. And, while we're at that, let us also continue to use technologies developed by independent research labs funded by the United States government!

Right, that doesn't increase risk of patent violation at all.

No, it indeed doesn't. You're missing the point that large companies do not check either whether what they do violates any software patents. The reasons are that such an activity would be prohibitively expensive, everything but watertight and open them up to treble damages in case of lawsuits.

They simply deal with patent problems as they are contacted by people who claim to own patents they infringe upon. They'll look at the patents, the demanded licence fee and at the company, and depending on the outcome they'll cross-license, pay a license fee or try to get the patent invalidated in a lawsuit (and possibly countersue the other company for patent infringement)

Those "solutions" are usually only viable for large companies though (regardless of whether it's about open or closed source), and not for small ones or individuals (again regardless of whether it's about open or closed source). In that sense the patent problems are indeed entirely independent of open vs closed source (who's going to attack IBM about patent infringements by Eclipse?), and only a matter of big vs small.

Re:Open source is more vulnerable to patents

[Halo1]

Access to source code

Access to software

Most software patents do not require access to the source to determine whether or not they are infringed, and often not even to the application to determine whether or not they infringe. Most software patents monopolise very high level features. You'll almost never find "low level" software patents. The ones that you can find and which are enforced, are generally those that cover standards (e.g. on mp3, gif, jpeg).

Licensing capacity

I don't think IBM would have any problems getting a proper license for some patents infringed by e.g. Eclipse. So this is generally more a big vs small than open vs closed issue (though definitely not always, e.g. the shareware GraphicConverter was able to keep offering GIF encoding, because the author paid patent license fees to Compuserv).

Re:Sue-ability

[rseuhs]

Another example is FireFox, many claimed it's flawless

Actually, no Firefox developer ever claimed such thing, no other serious OSS-person ever claimed that and even altough I find loads of crazy opinions on Slashdot and other forums, I haven't seen anybody claiming that on Slashdot or anywhere else.

Re:Liability of implementors of patented ideas

[Halo1]

That's correct, although in Europe there are some exceptions for research. For example, you are allowed to use a patented process for research purposes. I don't know whether it extends to publishing programs that describe (= implement) patented algorithms though.

Re:Not really

[Anonymous Coward]

you may also have noticed that the linux equivilent was a mainframe...while the windows box was a dual xeon. apples to oranges

Re:The author needs to learn how to do math ...

[Keeper]

Please cite the patent number that you are using as an example. It sounds like you're making the (common) mistake of reading the abstract/summary and making assumptions about what exactly the patent covers.

Most likely, the patent you are referring to covers a small, specific enhancement to the mouse. Kind of like how auto manufacturers patent small specific enhancements they make to engines.

Re:if anything MS patens violate open source!

[servoled]

No, Microsoft filed a patent application for the isnot operator which probably hasn't even been looked at by anyone at the USPTO yet. Go back and read the discussion from that story or better yet look at the published patent application. As a former patent examiner you should know the difference between an issued patent and a PGPUB.

Re:Someone should call MS out on this patent issue

[servoled]

"Your assertion is that Linux violates a set of patents, and therefore as an end user I might be found liable and forced to pay dammages at some future date.* Does using Windows remove this problem for me? Are you willing to either guarantee that Windows does not have any IP property issues or to indemnify me if someone decides that they want $699 for every copy of XP that I use because they think one of their patents is being violated?"

It looks like the answer to those questions is yes. See here and here.

So where is this bullshit of which you speak?

Culture of negativity

[HangingChad]

MSFT is taking a page from Karl Rove's playbook. Between the paid-for TCO studies and the patent scare tactics they've turned relentlessly negative.

I think it makes an interesting statement about MSFT products to hear them bashing the competition instead of selling the competitive strengths of their products.

To me it's almost an admission that their product line is not cost competitve with OSS. That's not strictly a price comparison, it's a value comparison. MSFT products do not give you the same value for the $$$ that LInux and OSS. Something most of us here have known for a long time.

MSFT has been sticking it to their customers for years with higher and higher license fees, back-stabbing EULA's and Naziesque business practices. I think they're really underestimating how bad people dislike being dicked and how long their memories can be.

I know it sounds a little Pollyanna but it's unfortunate that going negative can be so effective. It's kind of like spam. Everyone complains about it but as long as it's effective we're going to keep getting buried buy it.

Re:And how many patent lawsuits has Microsoft had.

[mikael]

It's also interesting to see the lawsuits that Microsoft has filed:

Microsoft sues controversial system assembler

Microsoft Sues Lindows.com Over Name

Microsoft takes on teen's site MikeRoweSoft.com

Microsoft sues Lucent in old dispute

Microsoft sues Brazilian magazine, IT official for defamation

Microsoft files lawsuit against five Md. firms

Of course, since they usually either buy out the company, develop and market a competing product, they don't need to resort to lawsuits for those type of situations.