Weekly Edition
Return to the Announcements page
| |||||||||||||
|
| |||||||||||||
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community
Florian Mueller writes
about the dispute between IBM and TurboHercules SAS; TurboHercules is
trying to commercialize an open-source emulator for IBM's mainframe
systems. "To add insult to injury, the list of patents with which
IBM tries to intimidate the Hercules project even includes two of the 500
patents IBM originally 'pledged' to the open source community. Patent
numbers U.S. 5613086 and U.S. 5220669 appear on page 4 of IBM's 2005
'patent pledge', and also appear as patents #83 and #106 in the letter IBM
sent to TurboHercules. This betrayal of the promise is unbelievable, but I
never believed that IBM was sincere about that pledge in the first
place."
A Scribd link to IBM's patent-threat letter is provided in the article, but it can be read without Flash on this page. It's worth noting that TurboHercules has been involving the lawyers on its side as well. (Log in to post comments)
The worth of patent pledges Posted Apr 6, 2010 14:07 UTC (Tue) by bjacob (subscriber, #58566) [Link]
Am I the only one to never have believed in the worh of patent pledges? Software patents undermine the very concept of a written Law that is the same for everybody. Indeed, since it's impossible to write software beyond "Hello, world!" without infringing dozens of patents, everybody is guily of patent infringement, and if everybody's guilty, the law is moot and the richer companies win. In that context, to make "pledges" is almost insolent. The only useful thing is to wholeheartedly campaign against the patentability of software (which the US supreme court never approved, and has never been recognized in Europe).
Involuntary infringement Posted Apr 6, 2010 14:51 UTC (Tue) by michaeljt (subscriber, #39183) [Link] It seems to me that in a lot of cases (I don't know if this is one, and IBM careful state that it isn't), an individual "infringer" could make a good case that they had no reasonable way of knowing about patents they infringed, as patents are plentiful as sand on the beach, and take a lot of time to read properly, which you can't really expect a time and cash-strapped developer to do. That developer is still liable for lots of money for involuntarily infringing on the patent(s). At least here in Germany though, I would expect (I don't know) that liability to be challengeable in court as it places unreasonable requirements on the "infringer" - I wouldn't expect them to be allowed to continue to use the patent indefinitely, but at least to be given a grace period so that they wouldn't have to cease all operations until they did.Anyone know how likely that might be in the US? It would certainly weaken the threat quite a bit if it were realistic.
Involuntary infringement Posted Apr 6, 2010 16:55 UTC (Tue) by drag (subscriber, #31333) [Link]
> which you can't really expect a time and cash-strapped developer to do.
You cannot expect _anybody_ to do. That's the problem. Nobody has the time or money to protect themselves.
If your a large company all you can do is make as many patents as possible and use them to cross-license with other companies and protect yourself that way. It's much more effective, much cheaper, and having numerous patents add to the perceived value of your corporation by the bean counters in the stock market.
If your a small company all you can do is ignore them, keep your head down, try to get a few patents of your own, and do your best to deal with them when they come up and people tell you your infringing.
> It seems to me that in a lot of cases (I don't know if this is one, and IBM careful state that it isn't), an individual "infringer" could make a good case that they had no reasonable way of knowing about patents they infringed,
Absolutely NOT!
Patents are absolute and they are designed specifically so that even if your completely ignorant of the patent you can still easily violate it.
Patents are a trade-off... If you get a patent then you are forced to publish your invention into the public domain... the compensation for this is that you are awarded a temporary monopoly on the invention.
Software patents themselves are probably invalid, legally. It's just that it's very difficult, very expensive, and very dangerous to argue that in front of the U.S.A. Supreme Court.
But anything short of that all the courts care about is if your violating the patent or not. It's not a judge's place to decide on the morality or applicability of the laws... they are simply judges that are there to decide whether or not you violate the law.
Either you violate the patent or not. Even if you have the judge sympathetic to your view point it's not going to mean jack shit.. your still going to lose hard.
The place to argue the practicalities of patent law is in front of Congress. They are the ones that can decide whether or not patents related to software can be invalid or not. They decide the rules, they make the laws, and they can change them.
Involuntary infringement Posted Apr 6, 2010 18:27 UTC (Tue) by butlerm (guest, #13312) [Link] If you get a patent then you are forced to publish your invention into the public domainThis is a joke. When was the last time anyone ever learned anything of significance by reading the text of a published patent? All the more in the field of software patents, where "invention" is something close to an oxymoron.
Involuntary infringement Posted Apr 7, 2010 8:27 UTC (Wed) by cladisch (✭ supporter ✭, #50193) [Link]
> When was the last time anyone ever learned anything of significance by reading the text of a published patent?
Two examples from the last two weeks:
US6212171 describes how to do certain optimizations on a FireWire bus; deriving the numbers for this algorithm would be very hard without buying the $400 standard.
US6405275 has a better explanation of TI's FireWire controller enhancements than their own datasheets. (This is a pure hardware patent.)
Involuntary infringement Posted Apr 8, 2010 6:40 UTC (Thu) by butlerm (guest, #13312) [Link]
"deriving the numbers for this algorithm would be very hard without
buying the $400 standard"
As compared to tens of thousands (if not millions) in patent licensing
Even where a patent documents something useful, the information will be
The only way to make the patent system moderately sensible would be to have
Involuntary infringement Posted Apr 8, 2010 8:01 UTC (Thu) by cladisch (✭ supporter ✭, #50193) [Link]
> "deriving the numbers for this algorithm would be very hard without buying the $400 standard"
> > As compared to tens of thousands (if not millions) in patent licensing fees?
That algorithm is an improvement over an earlier (free) one, so this patent implicitly documents how to avoid infringing it. (This happens quite often.)
> Even where a patent documents something useful, the information will be worthless by the time the patent expires in seventeen to twenty years.
US5052029 expired last Sunday, but it's still used (and needed) in every FireWire device. (But yes, by now that information is known anyway.)
> In fact, everyone is a lot better off by not reading patents. Treble damages and all that.
Andrew Trigdell famously disagrees for software patents.
For hardware standards like FireWire that are well known to be heavily covered by patents, using patents is unavoidable, so this defense probably wouldn't fly anyway. (The MPEG LA offers a 1394 patent portfolio license that every FireWire hardware manufacturer has to buy, so the situation here is similar to H264.)
Involuntary infringement Posted Apr 8, 2010 8:16 UTC (Thu) by michaeljt (subscriber, #39183) [Link] >> In fact, everyone is a lot better off by not reading patents. Treble damages and all that.>Andrew Trigdell famously disagrees for software patents. >For hardware standards like FireWire that are well known to be heavily covered by patents, using patents is unavoidable, so this defense probably wouldn't fly anyway. (The MPEG LA offers a 1394 patent portfolio license that every FireWire hardware manufacturer has to buy, so the situation here is similar to H264.) Which brings me back to my original question. I sort of assume that the reasoning behind the high damages for involuntary infringement of patents is that the "infringer" was just too lazy to research into what patents exist in the field. Which can be argued in a way in a case like this in which it is well known that there are a lot of relevant patents. In the more general software case though, it is clearly not feasible to research all patents which might possibly affect some code you are writing, particularly for "smaller" players. Does the US legal system not have any way of challenging laws that put that sort of unreasonable requirement on you?
Involuntary infringement Posted Apr 8, 2010 8:38 UTC (Thu) by cladisch (✭ supporter ✭, #50193) [Link]
> I sort of assume that the reasoning behind the high damages for involuntary infringement of patents is that the "infringer" was just too lazy to research into what patents exist in the field.
The intention of the patent laws is to encourage inventors, by giving them a temporary monopoly on their inventions. That involuntary infringers get hurt is a consequence that is accepted in order to reduce the risk to the inventor. In practice, it would be impossible to prove if an infringer was actually ignorant, lazy, or just managed to hide his knowledge of the patent.
As for treble damages for willful infringement: in practice, they are awarded only when the patent holder and the infringer talked previously about the patent, and the infringer didn't buy a license anyway. It's very hard to prove the necessary knowledge otherwise. (Well, flinging patent numbers about in public, like I'm doing, would be proof.)
> Does the US legal system not have any way of challenging laws that put that sort of unreasonable requirement on you?
"There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." (Ed Howdershelt)
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community Posted Apr 6, 2010 14:08 UTC (Tue) by chithanh (guest, #52801) [Link]
Some more information about TurboHercules
http://boycottnovell.com/2010/03/23/turbohercules-and-mic...
If I am not mistaken, the IBM patent threat is only against TurboHercules' commercial version, not against the Hercules project.
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community Posted Apr 6, 2010 18:12 UTC (Tue) by clugstj (subscriber, #4020) [Link]
So, Florian Mueller is actually full of sh*t when he says that IBM is breaking their promise. It doesn't help your credibility when you lie in the title of your article.
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community Posted Apr 6, 2010 19:23 UTC (Tue) by Cyberax (subscriber, #52523) [Link]
Hercules' commercial version is the same as its OpenSource version. They just sell support.
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community Posted Apr 6, 2010 21:44 UTC (Tue) by clugstj (subscriber, #4020) [Link]
Don't know if that's true or not, but I don't get that from the article. He says "commercialize" which implies something different than "sell support". Since we don't have the original letter, we can't say for sure what they told IBM that pissed them off. Until that is made public, I'd say it is too soon to pass judgment on IBM.
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community Posted Apr 6, 2010 18:34 UTC (Tue) by muwlgr (guest, #35359) [Link]
Yes, it is very usual for IBM. When their crown jewels and cash cows are touched, they fight toohs and nails in response.
Unfortunately, PearPC project is also abandoned now. They did not manage to emulate PREP/CHRP iron in a good way and boot AIX on it ...
Mueller: IBM breaks the taboo and betrays its promise to the FOSS community Posted Apr 6, 2010 21:59 UTC (Tue) by sepreece (subscriber, #19270) [Link]
Maybe I'm missing something - is there an actual patent threat anywhere in the published exchange? It looks to me like Hercules asked permission to do something and IBM said no. There's nothing in the published letter that looks like a cease-and-desist letter or a claim of infringement.
Since the request was for IBM to do something (sell licenses for running in a particular environment), saying "No" seems to be wholly within their rights and to have nothing to do with any "patent threat". Doesn't mean that Hercules can't build it's emulator; just that IBM won't license people to run their OS on it.
Is there another letter somewhere that threatens Hercules with anything?
|
|
||||||||||||