Advertisement
Front, Kernel, Security, Distributions, Development. See your byline here on LWN.net.
| |||||||||||||
|
| |||||||||||||
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw)Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw)Posted Oct 12, 2007 18:28 UTC (Fri) by sepreece (subscriber, #19270)In reply to: Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) by ketilmalde Parent article: Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw)
"perhaps showing the same directory in two file system browsers"
At a guess, not this one, unless the browsers were showing the same display object (i.e., were driven off the same graphical elements), as opposed to the same storage structure. That is, opening two browsers on the same subject would be further from what the patent describes than opening one browser and showing it in two windows.
But this is all guessing...
(Log in to post comments)
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 12, 2007 22:29 UTC (Fri) by drag (subscriber, #31333) [Link] You can read it for yourself.
http://patft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&...
Keep in mind that the patent language is divided up into several sections.
The first section is the 'abstract'... The abstract is a general overview of what is held in the patent.
But remember, it's JUST A OVERVIEW. It's not realy what the patent is about and arguing or examining anything in the abstract is totally worthless. It's ment specificly to be generic and easy to understand, but have no real meaning.
The import part is the next section which is called the 'Claims'.
That is the 'meat' of the system. You only have to violate a single claim to violate the patent, if I understand it correctly... Although this patent seems rather insane.
IANAL
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 12, 2007 22:31 UTC (Fri) by drag (subscriber, #31333) [Link] In fact the patent is completely insane. It doesn't make sense... anybody that let this past them had to be a moron.
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 13, 2007 3:34 UTC (Sat) by sepreece (subscriber, #19270) [Link] I do know how to read a patent, I just haven't read these in any detail.
However, I don't know why you say they don't make sense. They describe a particular user interface system and claim the elements that make it up. Based on a quick look, I didn't think the descriptions were particularly hard to read by patent standards.
The work this is based on was interesting and innovative. I remember reading about and playing with Rooms back when it appeared. The work was done by very smart people in a very innovative environment and I don't see any reason to assume a priori that the work wasn't inventive in the sense required to grant a patent.
Now, this suit is trying to apply the patents to something quite different from Rooms, arguing that the claimed elements are infringed by some aspects of the window system in the respective Linux systems. After a quick look at the complaint, I wasn't particularly convinced by the specific items they cited (claim 1 of '412, claim 3 of '183, and claim 1 of '521). The claims in '521 and '183 actually sound like they're talking more about the hardware implementing the UI than the UI itself. But, I don't claim to be an expert in the area or to have read a lot of patents in the area.
I sat in the jury for a moot court patent litigation once (my daughter was in the class and on the team arguing for the plaintiff). Patent trials are typically around whether or not infringement occurred - whether the defendant product actually infringes the claims of the patent - with a lot of incredibly detailed testimony about the nature of the claims and the ways in which the allegedly infringing product matches them, with the goal of convincing a jury carefully selected to know nothing about the technology. Believe me, the arguments to the jury will be nothing like as hard-to-understand as the patent itself; they will be reduced to very clear comparisons of features. It's certainly within the realm of possibility that a jury could be convinced on this one.
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 13, 2007 7:09 UTC (Sat) by drag (subscriber, #31333) [Link] I am sure that it's makes some sort of sense. It's still insane though.
And if they do win a patent suit against Redhat it will be a tragedy of justice. It will be a victory of a group of asshats who don't produce anything or do anything constructive for anybody except themselves versus people that provide products, software, and support (as well as innovations) that are actually usefull for something by the majority of people. And not only do they provide this in order to make a living they also provide it in a manner that is absolutely able to be used by anybody freely for pretty much any purpose.
(ps: I had the same attitude with Eolas vs Microsoft. Eolas should of lost and lost hard. It was horrible that Microsoft lost that suite and those Eolas people should of been sued themselves for just plain being huge dicks.)
THATS why it's insane. It's insane because it's convoluted and it's insane because such a thing should not exist in the first place. It leaves a bad taste in my mouth.
Simple, easy solution Posted Oct 13, 2007 15:36 UTC (Sat) by tony (guest, #3654) [Link] The easiest fix to the patent mess is to throw it out. I don't think the "innovation without patents is impossible" argument holds any longer, if it ever did.
However, there is a simpler solution to the threat of patent trolls.
Make patents non-transferable, and only grant them to the individuals who apply for them. (Personally, I believe only individuals should be granted copyright and patents, anyway, and only corporations should be granted trademarks. But that's just me.)
Non-transferable, non-corporate grants for both copyrights and patents would go a long way to reforming the system, and bring it in line with the intent of limited government-granted monopolies.
Simple, easy solution Posted Oct 14, 2007 0:12 UTC (Sun) by sepreece (subscriber, #19270) [Link] Non-transferable patents don't work because an individual inventor is unlikely to have the resources to bring the invention to market and corporations that do have the resources want to own what they invest in.
[Again, I'm not happy with the patent system as-it-is, on a number of issues, but I do think that some patents are justified...]
Simple, easy solution Posted Oct 15, 2007 19:31 UTC (Mon) by tony (guest, #3654) [Link] Want in one hand....
If the system were set up like that, corporations and patent holders would come to some sort of agreement. This doesn't stop exclusive deals and whatnot. It just curtails the formation of patent trolls, who purchase up patents with the sole intent of litigating.
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 13, 2007 15:52 UTC (Sat) by sepreece (subscriber, #19270) [Link] Well, presumably the inventors got paid for their invention (I guess they probably got paid by Xerox and Xerox later sold the patents). The argument in favor of this model is that innovation needs to be stoked by investment; if you can't get a return on that investment, the rate of innovation would be lower. So, even though Xerox didn't manage to create successful products that leveraged this invention, the possibility of selling the invention itself was a justification of investing in creating it. The trolls don't do anything productive themselves, but the existence of people willing to buy inventions does encourage companies to pay productive people to innovate.
You still haven't said *why* you consider this patent to be insane. From a quick read, it seems to identify something reasonably specific and to have been filed at a time when that something might have been reasonably inventive. Or is it that you think that the whole principle of software patents is insane?
[Note that I DO think it's bad policy that a software idea invented in 1986 should still be covered by a patent in 2007. The barrier to creation of software inventions is low enough that the duration of such patents should be much more limited than in fields where the barriers are much higher. If, of course, software patents serve any public purpose at all, which is highly debatable. One could easily argue that the barrier to software invention has today dropped to a point where it's silly to grant patents.]
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 14, 2007 8:58 UTC (Sun) by oak (guest, #2786) [Link] The time a patent is valid is insane indeed. There are some things thatcould be done to limit frivolous patents and patent trolls.
I think the reason for the large number of patents is that for patent
Then to patent trolls... For a patent violation claim to be valid, the
Patent Infringement Lawsuit Filed Against Red Hat and Novell - Just Like Ballmer Predicted (Groklaw) Posted Oct 14, 2007 18:27 UTC (Sun) by sepreece (subscriber, #19270) [Link] There's an interesting analysis of patent policy in the current CACM (<http://portal.acm.org/citation.cfm?id=1290961&coll=AC...>)
The authors look at the interaction between public benefit and variations in the duration of software patents, the difficulty of getting one, and the barriers to use of the idea by others during the life of the patent. I'm not sure they caught all of the factors, but the analysis is interesting and the conclusions are worth reading. They think long durations are OK if the barriers to patents are high enough (patents have to be good patents) and their are reasonable options for competitors.
Their final conclusion is that the difficulty and data requirements for figuring out the optimal policy argue for just not having software patents at all...
I tend to think that unless they radically raise the bar for getting a patent, the term should be much shorter. The idea of statutory mandatory licensing is also interesting, but fraught with complexity in determining what reasonable terms would be.
I also think that the free-software movement is itself [without referring to its philosophy at all] a powerful argument against the value of software patents - the fact that products implementing software innovations can be brought to market by free-software projects with little or no capitalization argues strongly against the idea that inventors need protection to earn a reasonable return on the cost of bringing the invention to market.
Broken patent system Posted Oct 15, 2007 0:14 UTC (Mon) by bojan (subscriber, #14302) [Link] This lawsuit itself points to big problems in the U.S. patent system, because:
- licensor itself isn't using the patent for anything useful
Put differently, if there were no people implementing this idea, the licensor's contribution would be exactly zero. They could have sued (if this patent was so important to them) various people _way_ before, but they held off until Red Hat and Novell had the money to pay. And finally, this patent is old news and it is the people coding this idea that actually had to put effort and money behind it, so rewarding inventors for this now makes little sense.
In the end these are the possible outcomes:
- RH/Novell spend money on something inventors never wrote
I don't see how any of the above is contributing to progress and I especially don't see why the inventors should be rewarded for this 16 years after their invention has been patented, in a field that churns out new versions every few months.
PS. In Australia, patents on software are encouraged by the relevant government body (IP Australia), so the situation isn't much better here either.
Broken patent system Posted Oct 15, 2007 12:54 UTC (Mon) by sepreece (subscriber, #19270) [Link] While I agree that there are problems with software patents, I think your list of objects is off-the-mark:
(1) They ARE doing something useful with the patent - they're licensing it to people who want to implement it. Would you criticize ARM because their designs are licensed to other people rather than fabbed in-house?
(2) Not sure what you mean by "last possible moment" - they've been litigating these patents against various parties for years; few companies want to maintain large numbers of patent suits in parallel - they're a lot of work and cost.
(3) "they are only targeting deep pockets" - well, duh. Not a lot of point to suing people who can't pay.
(4) I agree patent terms are too long. However, there's a reasonable argument that a patent that is the basis for a lot of extension and refinement should be protected while that's happening. No idea is innovative after it's been thought of, by definition.
(5) No idea == no implementation or delayed implementation. It cost money to create the environment that led to this idea; Xerox ultimately recouped some of that money by selling the patents. I think it's fair to guess that the modern GUI would have evolved if PARC hadn't existed, but it would have been delayed.
As I noted previously, I think that open source has significantly changed some of the argument for software patents - the notion that it takes a lot of investment to bring a software idea to practice or to bring it to market is harder to sell, today, than it was 30 years ago, when the current patent rules were largely formed.
Broken patent system Posted Oct 15, 2007 22:06 UTC (Mon) by bojan (subscriber, #14302) [Link] About 1, if that were true, they would have attempted to license (or sue) people distributing XFree86, Gnome and KDE 10 years ago. So, they aren't doing anything useful here - they are just waiting until the last moment to extort money.
About 2, see 1.
About 3, if you own a trademark for instance, you must threaten/sue even people that don't have deep pockets. And the system is still alive and well (and a lot fairer). Oh, and you must use the trademarks yourself.
About 4, the term of software patents is about 20 years too long :-) I still recall the idiocy of XOR patent that Autodesk had to pay for.
About 5, in a world without software patents, Xerox could have created a reference implementation instead (which _is_ actually a lot of investment) and sold that to implementers. The folks licensing that would then still have the first mover advantage, while imitators would have to wait for a while to do their own.
The point is, in software, copyright combined with trade secrets offers enough protection to keep innovators ahead of imitators. There is really no need to impose a 20 year monopoly in a field where things change every month, as it doesn't contribute to anything.
|
|
||||||||||||