Software Patents Could Stop EU Linux Development

sebFlyte writes "An expert in computer and Internet law has advised that if the CIID is passed in europe (which looks likely but is not certain) then the threat of patent litigation could bring European Linux development to a grinding halt." From the article: "There is no question that some of the open source software that is out there -- such as the Linux kernel itself -- has got patent violations in there. That is acknowledged. There is more danger that those potential violations will be litigated..."

Comments

that is trademarks

[badriram]

Trademarks work that way. Patents can be used to sue people no matter how long they may have violated it with or without the knowledge of patent holder. and IANAL

Re:that is trademarks

[Macadamizer]

"Patents can be used to sue people no matter how long they may have violated it with or without the knowledge of patent holder. and IANAL"

Not really true -- except maybe for the IANAL part, I don't have any info either way on that...

There is a concept in the law (U.S. law, at least) called laches -- basically, the law acknowledges that it is simply "unfair" for a patent holder (or any other rights holder, for that matter) to recognize that someone else is doing something wrong, and simply wait until the other guy has racked up damages, or whatever, before suing them. If a rights holder (such as a patent owner) waits too long to sue, he could lose the right to assert that patent against that particular infringer. The infringer could raise a laches defense, and if the court agreed that the patent owner "sat on his rights" for too long, the court could throw the case out.

Now, how long it too long? There isn't a set amount of time, it will depend on the overall "fairness" of the situation. It's not like a statute of limitations, where there is a hard line after which you can't take action -- rather, this is an "equitable" defense, where the court basically says, "yeah, Mr. Patent Holder, you do have a case here, but you waited way to long to bring it, and now it just wouldn't be fair to sue the defendant after all of this time."

An example would be holding a patent, seeing your competitor build an infringing product, and then waiting for 6 or 7 years for the product to take off in the marketplace, then suing once the company has made some money and racked up damages. That would be an example of where laches would come in -- if you wanted to protect your invention, you should have sued when you realized the infringing activity was taking place -- waiting for 6 or 7 years to rack up damages just wouldn't be "fair" to the other guy.

But again, that doesn't mean there is a statute of limitations are anything -- as long as the patent is still valid, you can sue someone who has been infringing for 10 years, as long as you only recently found out about the infringement. And even if you are barred by laches from asserting your patent against one guy, that doesn't mean your patent is invalid -- you can still go after other infringers.

This goes for patents, copyrights, trademarks, etc. It can also apply to torts, breaches of contract, really ANY type of court case -- but in most cases, the statute of limitations will kick in before enough time passes for a laches defense to be successfully raised. However, in certain types of cases, even a very short delay -- a couple of months even -- can be enough to raise a laches defense.

Re:Really?

[jd]

I guess it could infringe on the IsNot() patent, but then the counter would be that Linux is prior art and therefore IsNot() as a patent is invalid.

Actually, there's not a shred of evidence that Linux has any patented code in it. What has been agreed is that the current development model wouldn't stop such code entering the kernel. This is not the same as saying that this has already happened.

If it had, SCO would likely have presented this as evidence of potential violations as part of its case. That it has yet to present a single example, either of a violation of their own patents/IP or even a violation of someone else's, is strong evidence no such violation has actually occured.

The closest I can recall was an uncredited piece of network code that fell under the BSD license. That violated the author's copyright - not patent, just copyright - but was resolved. Oh, and some jerk tried to trademark the word "Linux" for an OS, suing Linus and a whole bunch of distros for trademark infringement. The thing was settled out of court, IIRC, with the a*hole actually getting his money back for the trademark filing. (That's like a convicted blackmailer demanding the victim reimburse them for the cost of postage.) The other terms of the settlement were sealed.

Linux uses stuff donated by SGI and IBM, but that code has been clensed of any IP. Both companies went through a lengthy - and impressively transparent - process to really illustrate just how clean the code was. The only stuff that's even remotely suspect is the NUMA code, but since that was one of the prime targets for SCO, you can be certain it has been gone over with a fine tooth-comb by patent lawyers and coders on both sides of the fence. Nothing has been found, because there's nothing there to find.

Re:Maybe Linux has violations in it..

[Anonymous Coward]

Since there are no software patents in Europe right now

There are thousands upon thousands of software patents in Europe right now. Various countries' patent offices have been granting them for years, despite the fact that they cannot be enforced.

One of the more disturbing things about this patent legislation moving forward in the EU is that these will be grandfathered in and will come into full force once patents are allowed. I can't even imagine the feeding frenzy which will happen at that point, and frankly I wonder if these people have really thought it through at all. It almost certainly will cripple the software industry in Europe for years, if not permanently.

Laws passed doesn't mean it is valid.

[AKosygin]

In California, at least, if a city or some local government post a speed limit sign that says 10 MPH on an interstate freeway (for example), you don't have to follow it because the speed sign is not "reasonable" from an engineer/public safety perspective. All speed limit signs MUST be backed by engineering studies that says the proper speed limits should be set at x MPH for this stretch of road. If you get a ticket for speeding, and in court you demand the prosecutor providing engineering studies proof to back-up the vailidity of the speed limit, and they cannot, your case is then dismissed and you are free.

Therefore, even if the law is passed, it doesn't mean you should follow it blindly.

On another tangent, if the traffic is going 80 MPH, and you go 70 MPH on a 60 MPH highway it was possible to get a ticket for going too fast and too slow at the same time. Though usually the court will throw out one of the two.

Re:Really?

[darkonc]

Actually, there's not a shred of evidence that Linux has any patented code in it.

Excuse me??? Microsoft took out a patent on sudo. Yes, I realize that this patent should have been shot at the gate, but right now sudo violates a patent that was probably written by somebody who was in diapers when sudo was first written.

The fact that a patent is absolutely assinine and should never have been granted may lessen, but does not remove it's chilling effect on Linux development. That's part of why it's so important to leash (if not put down) the software patent industry.

Re:Hasn't slowed US Linux Development

[nchip]

Fedora/RedHat lacks mp3 and divx/mpeg/etc playback because of patent issues. Mandrake (France) and SuSe (Germany) ship with a lot better multimedia experience. WPA (802.11x) is another patented like hell tech, so don't hold your breath for a smooth out-of-box wireless support on RedHat either.

As more patents are filed all the time and old ones get enforced, the Linux experience will get dampened even more.

While the western tech media has blamed Chinas relucancy to adopt 802.11x on their jingoism, It is more likely that China doesn't think that RAND terms of those patents are that Reasonable and Non-discriminatory after all..