| |||||||||||||
|
| |||||||||||||
Patent browsingPatent browsingPosted Mar 27, 2009 23:51 UTC (Fri) by giraffedata (subscriber, #1954)In reply to: Patent browsing by man_ls Parent article: An afternoon among the patent lawyers
I have got the idea that infringement is unavoidable for every substantial program. Two things wrong with this statement:
according to Microsoft Linux infringes 235 patents. Now I would bet good money that no prominent Linux developer has spent any time reading Microsoft patents, but do you really believe that reading them would have reduced that number significantly? But what about the people actually risking infringement liability? IBM, for example, ships its own version of the Linux kernel in its embedded devices. I don't know how it differs from kernel.org, but rumor has it IBM expended some effort to determine if the kernel.org version infringes patents and omit parts that might. You are the expert, No, I'm not. I'm a contract lawyer with a side interest in computers. Virtually everything I know about patent law I learned from LWN.
(Log in to post comments)
Patent browsing Posted Mar 28, 2009 0:06 UTC (Sat) by man_ls (subscriber, #15091) [Link] I am not trying to argue for argument's sake, but this is actually an interesting discussion, at least to me.It's not infringement in the patent is invalid, which most of these are.If it was easy to determine if a patent is valid or not, then validity would matter. As shown in the FAT patent it isn't so it doesn't: PUBPAT tried for several years to get this stupid patent invalidated and in the end failed. True, it was against Microsoft, but in an ideal world (and we were talking about one) the size of the adversary should not matter but who is ultimately right. IBM, for example [...]Whoa whoa whoa. It's not "IBM for example...", but "IBM (which is probably the only company in the world with enough resources to do this)..." IBM probably employed a number of patent lawyers for some time to do this job, and it is part of their work to read patent applications. But this job still does not fulfill the stated objective of patents, which is to publicize inventions -- I doubt any of these patent lawyers approached an IBM engineer and suggested "hey, I read about an interesting algorithm, let's get a license and use it for our next project". And besides IBM has the largest stockpile of patents around, so they probably don't care too much unless the patent is really prominent.
Patent browsing Posted Mar 28, 2009 2:38 UTC (Sat) by giraffedata (subscriber, #1954) [Link] I am not trying to argue for argument's sake, but this is actually an interesting discussion, at least to me. Me too, but the argument might be wandering. I thought we were talking about whether the strategy of deliberately not reading patents to reduce one's patent infringement liability works, and consequently whether people actually do it. You're also talking about ideal worlds and the stated objective of patent law.
It's not "IBM for example...", but "IBM (which is probably the only company in the world with enough resources to do this)..." I don't think IBM is that big for the purposes of this discussion. The size that matters is the revenue the company can expect to generate by using a Linux kernel. IBM's entire hardware revenue is only something like 20% of the company's, and only a few products contain Linux. I think Cisco might have more money than IBM to spend on Linux legal work. But if the situation is as bad as you say -- that there's just no way to know until you've lost a lawsuit whether your software product infringes a patent, then it probably doesn't matter how much you spend on a patent search, and then you might as well cut your risk a little by maintaining ignorance of all patents.
I doubt any of these patent lawyers approached an IBM engineer and suggested "hey, I read about an interesting algorithm, let's get a license and use it for our next project". Actually, I don't think anyone ever intended the patent files to be used that way -- the patents themselves aren't written to be enlightening. I think the model was that an engineer would see that Windows boots twice as fast as AIX and call up Microsoft and ask how it does it. Microsoft happily tells him, gives him the patent number, and quotes a price to use it. An IBM lawyer checks the patent to make sure it really does cover the method described, then recommends payment. Or maybe the IBM engineer reads an article by a Microsoft engineer in a technical journal.
Patent browsing Posted Mar 28, 2009 10:49 UTC (Sat) by man_ls (subscriber, #15091) [Link] You're also talking about ideal worlds and the stated objective of patent law.I just don't want to lose track of the original discussion: whether it is worth reading patents or not. There are pros and cons; the cons are triple damages and extreme boredom, while the pros include being able to avoid filed patents -- but also learning new tricks or finding technnical solutions to existing problems. Actually, I don't think anyone ever intended the patent files to be used that way -- the patents themselves aren't written to be enlightening.You will not mind an engineer to quote the law... 35 USC 112 says: The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.It should be enough for making a working prototype. It is true that there may be things missing, like failure modes. According to the wikipedia article, which is actually very informative: In some countries there are requirements for providing specific information such as the usefulness of the invention, the best mode of performing the invention known to the inventor, or the technical problem or problems solved by the inventionThese requirements look more complete than the US ones. However I would imagine that an inventor would want prospective licensors to get as much as possible from the patent application alone, without any direct intervention.
Patent browsing Posted Mar 29, 2009 20:55 UTC (Sun) by oak (guest, #2786) [Link]
> I just don't want to lose track of the original discussion: whether it
> is worth reading patents or not. There are pros and cons; the cons are > triple damages and extreme boredom, while the pros include being able > to avoid filed patents
So you need to go through also patent applications, not just granted
Patent browsing Posted Mar 29, 2009 21:58 UTC (Sun) by man_ls (subscriber, #15091) [Link] Well, you could make the argument that patents not granted are an untapped source of great ideas... but I am not even sure they can be browsed. No, I meant patent applications that have been granted, sorry.
Patent browsing Posted Mar 29, 2009 22:12 UTC (Sun) by giraffedata (subscriber, #1954) [Link] I think oak's point is that if you want to avoid inadvertently using a patented invention, you also have to know about the things that have already been invented, but not yet had patents issued.That would entail knowing not only about the filed but not yet granted patents, but also the inventions that haven't even been filed yet (you have a year after invention to do so). Just like the unfiled inventions, you can't know about the filed but not granted ones either, because an essential part of the system is that patent applications are confidential. That way, until the patent issues -- and forever if it never does issue -- the inventor can still have his monopoly via trade secret. This is the source of the "submarine patent" controversy. Some inventors deliberately slow down the patent application process to give competitors lots of time to independently invent, and grow a dependency on the invention, and then when the patent finally issues, it's worth a lot more.
Patent browsing Posted Mar 29, 2009 22:36 UTC (Sun) by man_ls (subscriber, #15091) [Link] As Colonel Kurtz in Apocalypse Now, the more I learn about software patents the more I want to write in big bold letters: Drop the bomb, exterminate them all. So glad we don't have them in the EU.
Patent browsing Posted Mar 30, 2009 19:22 UTC (Mon) by oak (guest, #2786) [Link]
> Just like the unfiled inventions, you can't know about the filed but not
granted ones either, because an essential part of the system is that patent applications are confidential.
Even if they would be available, as they haven't been prior-art-filtered
Basically I consider patenting as rigged game of PTO & court roulette
Patent browsing Posted Mar 30, 2009 21:03 UTC (Mon) by man_ls (subscriber, #15091) [Link] You are assuming that prior art screening somehow works, when everything points to its pervasive failure. I think that granted patents do cover all kinds of stuff in our own software, as it is. You have probably seen what happens here on LWN every time a software patent is discussed: commentators routinely find prior art quite relevant to the patent. It follows that it is useless to browse not only applications, as you correctly state, but any patents issued by the USPTO.
|
|
||||||||||||