Nearly every inventor I have met believes that everything they make is novel. I would never leave it up to inventors to decide what is unique and how much protection under the law they should be afforded.
Oh, I absolutely agree. But I don't think anyone is proposing a "inventor-is-also-the-examiner" reform to the patent system, are they?
the office only allowed about a third of the cases applied for
...and half of those shouldn't have been granted either. I'm not the only crackpot of this opinion; anyone remember the study that showed upwards of 80% of all software patents could be invalidated with prior art software?
Large companies hire firms or establish large in house shops to manage their portfolios. They buy "quality" patents from smaller firms or inventors. If they want to use something from another large company they usually cross license. All of this is predicated on the fact that they will not buy or cross license bad patents.
All of what you are saying is predicated on the fact that the players are all large corporations. Guess what? They aren't. Producing a piece of software, even a very trivial free one that you give away as a hobby, puts a big ol' 'send me a cease and desist letter' or 'sue me for infringement' target on your back. You can't tell me this isn't true -- I've lived through it.
Also it's rare that a firm sues another on a single patent.
I agree again, but for a different reason: it's rare that a firm sue anyone when they can simply send out threatening cease and desist letters. If the target is small enough, all they can do is cease and desist, even if they don't believe they have infringed any claims or if they believe the claims to be invalid. Once again, I beg your humble forgiveness for not being richer so that I could afford to defend myself in court.
The system encourages invention. Take White LEDs for instance.
A wonderful example! I'm glad you brough this up. The inventor of the blue and white LED worked 12 hours a day, 7 days a week except holidays for his employer Nichia. When he finally achieved the bright blue LED, they gave him a $200 dollar bonus. That's it. He later got more in a court settlement, but it's hard to say that he or his company were motivated by the patent. They discovered a process that their competitors couldn't have duplicated without duplicating Nakamura's research and knowledge gained over those years, and there was plenty of lucrative money to be made by being able to produce these. In fact, that's what Nichia did as right away: announce the production of blue LEDs, not announce licensing deals for a patent that had been granted. Admittedly, the licensing deals now give Nichia a nice revenue stream, but that is beside the point -- the natural, market-based motivation to create this innovation was exactly that: natural and market-based. The patent rights DID NOT cause the invention of the blue LED.
Slashdotters, while being surprisingly well informed in some areas, keep extolling the virtues of socialism as every oppertunity (remind me again when that has worked in the history of man?).
Hmm. That is a curious observation. From what I've seen, the crowd here tends much more toward libertarianism and less government, not more. And that, my friend, is the common thread that ties this to the patent system: patents are not natural -- they are government enforced monopolies. Patents don't exist without a government to back them. And, as a matter of setting you straight: monopolies have a lot more to do with communism than they do with capitalism. Under a communist system, the idea is that the state can more efficiently centrally plan industrial production, and can do away with the 'inefficiencies' of capitalist competition. I don't know about you, but that sounds an awful lot like the patent system, w
Societies don't have rights. Individuals have rights.
You are absolutely correct about human rights, but wrong about patents. Patents are not a right, but rather a restriction on the rights of others. A patent is a government enforced monopoly over the implementation of an idea (and sometimes, unfortunately, over the idea itself). And forced monopolies are not rights, but rather the taking away of rights of others.
What you do have, as a natural right, is the right to create. That right is pre-society, pre-government, pre-law. It is only when government comes into play that patents can exist, otherwise who will prevent all but the patent holder from excercising their right to create?
Some of the first "patents were granted on manufacturing salt, soap, glass, knives, sailcloth -- things that people had first created many centuries (or even millenia) before, and that until the time of grant, could be made by anyone with the resources and knowledge to make them" (from this post).
I think you need to be educated on the concept of hindsight (everything looks obvious in hindsight)
This is a common refrain from defenders of a broken system. The truth is, many patents are obvious not only in hindsight, but in foresight. Many patents are granted on methods that are already well-known and widely used in practice, but because they don't have corresponding write-ups in literature, the patent database, or other easily accessible resources, they are given monopolies by the patent office.
Pretending that every granted patent that is non-obvious because patents are only supposed to be granted on non-obvious inventions is circular reasoning, and I don't need a law degree to see through that.
examiners are under no pressure to allow, we get counts towards our quotas for abandonments and when the applicant elects to start the clock over again via a request for continued examination.
It is called 'regulatory capture', and the USPTO is absolutely a good example of it. You bring in money for granting applications. Token rejections are good because it costs the applicant money to 'start the clock over again.' Eventual granting of the patent is good because it results in continuance fees over the lifetime of the patent. Congress is inclined to make laws that make it easier to grant more patents, because Congress likes to stick their hands in your very well endowed USPTO coffers. The USPTO is a profit center for the government, not a cost center.
Look, I'm not saying that patent examiners are corrupt. I'm not saying your bosses are corrupt. All I'm saying is that the system encourages favoring the applicant, and the potential rewards for receiving a patent are so great that people will do all sorts of immoral things to obtain them. A system which encourages corruption is corrupt by definition, even if the minor players in that system are trying their darnedest to be honest and forthright.
It's quite common for these specialized examiners to be as adept in the field as any inventor working in that same field.
Do you really believe this? I'm sorry, but it just doesn't ring true to anyone who has browsed through many of these patents and read their contents. The truth is, patent examiners are overworked, underpaid, and under incredible pressure (by those giving the USPTO money for applications) to grant patents. Sure, there is usually the few token rejections and rewrites, but anyone who has gone through the process of obtaining a patent can tell you that persistence usually wins out.
There is a reason why nearly all cases brought before the office are by lawyers representing inventors and not by the inventors themselves.
This is damning evidence against the patent system. If you'll recall, the great compromise of the patent system was that the government would grant a monopoly if the inventor would publicly disclose how the invention works -- the main impetus was to keep secrets from getting tied up in guilds or going with the inventor to the grave. If it is as you say it is, the patent system is broken by definition, because only lawyers can understand the applications. Us lowly slashdot readers don't have a chance, you imply, because we just are not smart enough.
Bad patents cannot be enforced in a court of law and are therefor not valuable to the inventor...people who pursue bad patents only harm themselves
Once again, this rings false. The average cost of defending oneself in court against a patent claim averages around 2 million dollars. That gives the holder of a 'bad' patent incredible leverage -- as long as they ask for something reasonably less than 2 million dollars in licensing fees, the prudent "infringer" will pay up rather than fight. Don't tell us this isn't how it works -- one need only look at how many billion dollars the Lemelson "computer vision" patents brought in before eventually being challenged and invalidated.
A patent on some random element of a flower pot is only useful in that third parties find it useful and without an alternative
...unless the patent is overly-broad, in which case the patent on a random element of the flower pot gets applied to all sorts of new technologies that weren't envisioned when the patent was filed. Again, I'll reference the Lemelson patent portfolio, but that is by far not the only example. Just do a 'patent' query on news.google and you'll see a host of others, for example the NTP patents on sending email over wireless medium.
Lawyers understand the quality of the USPTO far better than the average public.... feel free to ingore this or mod it down so you can continue to sound ignorant to those that have bothered to understand the details of Patent Law.
Once again, I beg your merciful forgiveness for not being a super smart lawyer or lawyer-admirer like yourself! If only all of us could become IP lawyers the world would be a better place!
In the meantime, feel free to continue your naive perspective in which the wonderful patent system not only doesn't ever impede progress, never hurts inventors, doesn't retard products from making it to market, and never sets scientific research back by decades, all the while enriching all of us equally according to the merit of our cleverness. Oh, what a wonderful system it is!
I hate to make such an analogy, but this sounds exactly what Boromir said in The Lord of the Rings. Using evil for good is a very dangerous thing.
You are right. I probably should have said "could be used reciprocally," as in, an author gets cease-and-desisted or sued for patent infringement by X, so they reciprocally cease-and-desist or sue X using patents from the pool. This is how the big corps get away with software development that is otherwise off-limits to the smaller guy. But I absolutely agree -- it is still evil, but then again, there is no other solution available without getting new legislation passed.
If we can get new legislation passed, then my absolute favorite reform is the independent invention defense, which invalidates a patent if another party independently discovers the same principle/idea. The onus of proving that the discovery was independent lies with the second discoverer. Nonetheless, this shouldn't be any harder to prove than is the current "first to invent" principle from easily producible evidences.
Byfield writes, "no patent claim has ever been upheld against FOSS." This isn't entirely true. I know of at least one open source project that shut down after receiving a cease-and-desist letter from a patent holder, and I'm sure there are many more. Technically, none of these claims have been 'upheld' by a court of law, but I think that stems more from the fact that us poor open-source developers don't have the resources to fight cease-and-desists or other methods of shakedown. Our only option is to fold.
Nice Rebuttal to Wyne at Right to Create
on
The Demise of IP?
·
· Score: 1
Right to Create has a nice rebuttal to Wyne's insanity, pointing out the errors in her analogies and logic. To quote just a tiny bit:
If Ms. Wyne were truly concerned about confiscatory government policy, she would be outraged at the growing number of commonly held ideas that are confiscated by our government and turned into private monopolies through the abuses of our patent, copyright, and trademark policies. To suggest that the opposite is true, that freeing ideas is the same as confiscating them, is to pretend that water is dry and fire is cold.
#1 is good, but #2 is problematic -- it shifts the 'power to patent' further in favor of the large corporation and further away from the small startup or independent inventor.
Right. But if the patent does issue, and I refused to give him royalties or cease from my "infringing" activities, where do I stand? Wouldn't he be entitled to damages from me in that case, since the claims in the provisional have the date of publication, and those extend to the real patent upon grant?
Okay, "provisional application." I was wrong to call it a "provisional patent."
But I don't see how this changes anything. The patent, when granted, gets "the benefit of the earlier filing date," as you said. Exactly. So if I get a cease-and-desist from Mr. Knight tomorrow (Nov 4 2005) because I wrote a screenplay about a guy who falls asleep for X years but lives a separate life during that time as a zombie, and I ignore it and publish/produce the movie anyway, and then Mr. Knight gets his full patent granted within 12 months (Nov 3 2006), but with the filing date as Nov 3 2005, where do I stand? I stand as infringing from the moment he sent me the cease-and-desist, do I not?
4. This is also being covered at Peter Zura's Two-Seventy-One Patent Blog (Peter Zura "is a registered patent attorney practicing in the Chicago area. He is a former patent examiner and has considerable experience in patent litigation and patent portfolio development and management.")
In short, the provisional patent application becomes enforceable as of the publication date, if the patent is eventually awarded. This is why "According to the official Patent Office website, provisional rights 'provide a patentee with the opportunity to obtain a reasonable royalty from a third party that infringes a published application claim provided actual notice is given to the third party by [the] applicant, and a patent issues from the application with a substantially identical claim.'"
In other words, you could infringe this provisional patent if you wanted, even after being notified by Mr. Knight, but if he is successful, you will owe him royalties.
I hope this doesn't make sense to you -- because it doesn't. We are laboring under an extremely broken system which, rather than rewarding innovation, rewards monopoly stagnation and locking up ideas from the public.
You really should get involved in the fight against this nonsense. Right to Create and freeculture are good places to start.
"Of course, any 'infringers' of this patent could ignore the cease-and-desist letters sent out by Mr. Knight, but they'd be doing so at great risk, as those actions would be actionable once the patent is awarded."
I should clarify that. The reason ignoring the cease-and-desist could be actionable is that if the patent is granted, the provisional patent grant date becomes the patent issue date. Meaning that all knowing infringers from today onward would be liable. The only way you could ignore a cease-and-desist and be okay is if the patent gets rejected.
Go ahead and read the linked story again. Note the term "provisional patent." Then read my response here.
I agree that this is insanity, but it is not as innocuous as you think. Mr. Knight can begin sending out the cease-and-desists right now, with the full force of the law behind him.
I hope this makes you mad. I hope it makes you mad enough to take some action. There are a number of things you can do. See some of the other posts at Right to Create and freeculture for things you can do to put an end to some of this non-sense. Letting your representatives and congressmen know how you feel is a good place to start.
You don't have it quite right. This is a provisional patent. A full patent has not yet been granted on merit, but its author can begin the shakedown today.
Please see the "Features" section of the USPTO explanation of provisional patents: "enables immediate commercial promotion of the invention with greater security against having the invention stolen."
Of course, any 'infringers' of this patent could ignore the cease-and-desist letters sent out by Mr. Knight, but they'd be doing so at great risk, as those actions would be actionable once the patent is awarded.
Does this make any sense? Of course not. But that's our patent system for you.
As one who has recently been on the wrong end of a cease-and-desist order for a very simple idea embodied in a bogus patent that was entirely unenforceable[*], as one who could not defend myself against said claims because litigation was too expensive, and as one who therefore removed said technology (which was available freely as open source) from the Internet, I can tell you that our system does not protect innovation but rather rewards IP-trolling and massive IP-land-grabs.
[*] The first patent on the technology was taken out in 1980. Since that time, roughly 40 academic publications have been authored showing ways to vastly improve the technology. Five or so additional patents had been granted in the 1990s. The implementation I used infringed none of these patents, except perhaps for a claim or two of the patent that was issued in 1980 and had since expired.
"Obviousness" is such a tricky, subjective criteria that the USPTO seems to have given up on it. Or, at the least, they've combined it with the "prior art" criteria, so that anything that isn't already in the patent database is both without prior art and non-obvious.
The other test for obviousness (the one that is somewhat sane) is apparently left for the courts to decide after the patent is granted.
You might also be interested to know that our patent system originated in 15th century England, and had nothing to do with novelty, non-obviousness, or prior art and everything to do with exclusive, state-sponsored monopoly (see A Brief History of Idea Monopoly for details on how such commonly manufactured items as soap, salt, glass, and sailcloth were granted patents).
The striking thing here is that our current patent system is starting to look a lot like the old 15th century English one, where "low quality" patents are granted willy-nilly, punishing the general public by levying a sort of tax on everyone except the owner of the patent. In other words, it seems more and more to have everything to do with exclusive, state-sponsored monopoly and nothing to do with protecting innovation and inventors.
There is an interesting post at Right to Create that discusses Roberts and his penchant for being a strong IP-maximalist (in other words, a weak supporter of the freedom to create and invent), and points out that the Senate Judiciary Committee didn't ask him one question on this topic.
The problem is largely one of patent quality. One important test is, 'were the inventions that they patented non-obvious?" The problem with that test is that it is entirely subjective, and so the USPTO seems to have combined it with another important test, that of prior art. Prior art for the USPTO means, "is there already a patent that covers this idea." If there isn't, then the chances are (from a patent examination standpoint) the patent is both non-obvious and there is no significant prior art. Anything outside of that criteria is decided after the patent is awarded by a judge or jury during litigation.
Now, of course that all flies in the face of common sense. But that's our patent system. Until we change it, that's what we are stuck with. If you are interested in doing your part to fix it, you should be reading sites like:
> If one were to do a somewhat more sophisticated analysis, a better model for hard drive failures is the Bathtub curve.
If one were to do a somewhat more sophisticated build-the-danged-thing-and-watch-the-failures, one would be hard pressed to find a bathtub curve anywhere at all in the data. Your analysis appears sound, but analysis is often misleading.
> I think what you are referring to is how multiple observations of a uniformly distributed stochastic variable generally look. It doesn't have anything to do with fractals, though.
You statistical guys always want to avoid self-similarity!:) I really do think, though, that you'd find a self-similar model to be more accurate representation of the combined failure rates of such a system than a multiple-observations-of-a-uniformly-distributed-s tochastic-variable model.
Not to nitpick back at you or anything, but have you ever sat in front of a system with 100s of cheap-off-the-shelf drives and recorded the failure times? I'll be a monkey's uncle if they aren't self-similar.
Slashdot Mirror
Oh, I absolutely agree. But I don't think anyone is proposing a "inventor-is-also-the-examiner" reform to the patent system, are they?
All of what you are saying is predicated on the fact that the players are all large corporations. Guess what? They aren't. Producing a piece of software, even a very trivial free one that you give away as a hobby, puts a big ol' 'send me a cease and desist letter' or 'sue me for infringement' target on your back. You can't tell me this isn't true -- I've lived through it.
I agree again, but for a different reason: it's rare that a firm sue anyone when they can simply send out threatening cease and desist letters. If the target is small enough, all they can do is cease and desist, even if they don't believe they have infringed any claims or if they believe the claims to be invalid. Once again, I beg your humble forgiveness for not being richer so that I could afford to defend myself in court.
A wonderful example! I'm glad you brough this up. The inventor of the blue and white LED worked 12 hours a day, 7 days a week except holidays for his employer Nichia. When he finally achieved the bright blue LED, they gave him a $200 dollar bonus. That's it. He later got more in a court settlement, but it's hard to say that he or his company were motivated by the patent. They discovered a process that their competitors couldn't have duplicated without duplicating Nakamura's research and knowledge gained over those years, and there was plenty of lucrative money to be made by being able to produce these. In fact, that's what Nichia did as right away: announce the production of blue LEDs, not announce licensing deals for a patent that had been granted. Admittedly, the licensing deals now give Nichia a nice revenue stream, but that is beside the point -- the natural, market-based motivation to create this innovation was exactly that: natural and market-based. The patent rights DID NOT cause the invention of the blue LED.
Hmm. That is a curious observation. From what I've seen, the crowd here tends much more toward libertarianism and less government, not more. And that, my friend, is the common thread that ties this to the patent system: patents are not natural -- they are government enforced monopolies. Patents don't exist without a government to back them. And, as a matter of setting you straight: monopolies have a lot more to do with communism than they do with capitalism. Under a communist system, the idea is that the state can more efficiently centrally plan industrial production, and can do away with the 'inefficiencies' of capitalist competition. I don't know about you, but that sounds an awful lot like the patent system, w
What you do have, as a natural right, is the right to create. That right is pre-society, pre-government, pre-law. It is only when government comes into play that patents can exist, otherwise who will prevent all but the patent holder from excercising their right to create?
Some of the first "patents were granted on manufacturing salt, soap, glass, knives, sailcloth -- things that people had first created many centuries (or even millenia) before, and that until the time of grant, could be made by anyone with the resources and knowledge to make them" (from this post).
Pretending that every granted patent that is non-obvious because patents are only supposed to be granted on non-obvious inventions is circular reasoning, and I don't need a law degree to see through that.
Look, I'm not saying that patent examiners are corrupt. I'm not saying your bosses are corrupt. All I'm saying is that the system encourages favoring the applicant, and the potential rewards for receiving a patent are so great that people will do all sorts of immoral things to obtain them. A system which encourages corruption is corrupt by definition, even if the minor players in that system are trying their darnedest to be honest and forthright.
In the meantime, feel free to continue your naive perspective in which the wonderful patent system not only doesn't ever impede progress, never hurts inventors, doesn't retard products from making it to market, and never sets scientific research back by decades, all the while enriching all of us equally according to the merit of our cleverness. Oh, what a wonderful system it is!
If we can get new legislation passed, then my absolute favorite reform is the independent invention defense, which invalidates a patent if another party independently discovers the same principle/idea. The onus of proving that the discovery was independent lies with the second discoverer. Nonetheless, this shouldn't be any harder to prove than is the current "first to invent" principle from easily producible evidences.
Now, if the open source patent pools could be used offensively, or the Independent Invention Defense were allowed, we'd probably see some action.
If Ms. Wyne were truly concerned about confiscatory government policy, she would be outraged at the growing number of commonly held ideas that are confiscated by our government and turned into private monopolies through the abuses of our patent, copyright, and trademark policies. To suggest that the opposite is true, that freeing ideas is the same as confiscating them, is to pretend that water is dry and fire is cold.
There are some really good reforms that could take place, short of giving up the goal of abolishing the patent system altogether. A list of a few is given at this post on China's entry into tighter patent law. My favorite? The Independent Invention Defense.
And, sadly, it looks like this guy is serious. Looks like he's even set up a practice to promote helping others get storyline patents.
Right. But if the patent does issue, and I refused to give him royalties or cease from my "infringing" activities, where do I stand? Wouldn't he be entitled to damages from me in that case, since the claims in the provisional have the date of publication, and those extend to the real patent upon grant?
But I don't see how this changes anything. The patent, when granted, gets "the benefit of the earlier filing date," as you said. Exactly. So if I get a cease-and-desist from Mr. Knight tomorrow (Nov 4 2005) because I wrote a screenplay about a guy who falls asleep for X years but lives a separate life during that time as a zombie, and I ignore it and publish/produce the movie anyway, and then Mr. Knight gets his full patent granted within 12 months (Nov 3 2006), but with the filing date as Nov 3 2005, where do I stand? I stand as infringing from the moment he sent me the cease-and-desist, do I not?
1. US Patent Application 20050244804
2. USPTO explanation of provisional patents
3. FAQ about provisional patents
4. This is also being covered at Peter Zura's Two-Seventy-One Patent Blog (Peter Zura "is a registered patent attorney practicing in the Chicago area. He is a former patent examiner and has considerable experience in patent litigation and patent portfolio development and management.")
In short, the provisional patent application becomes enforceable as of the publication date, if the patent is eventually awarded. This is why "According to the official Patent Office website, provisional rights 'provide a patentee with the opportunity to obtain a reasonable royalty from a third party that infringes a published application claim provided actual notice is given to the third party by [the] applicant, and a patent issues from the application with a substantially identical claim.'"
In other words, you could infringe this provisional patent if you wanted, even after being notified by Mr. Knight, but if he is successful, you will owe him royalties.
I hope this doesn't make sense to you -- because it doesn't. We are laboring under an extremely broken system which, rather than rewarding innovation, rewards monopoly stagnation and locking up ideas from the public.
You really should get involved in the fight against this nonsense. Right to Create and freeculture are good places to start.
I should clarify that. The reason ignoring the cease-and-desist could be actionable is that if the patent is granted, the provisional patent grant date becomes the patent issue date. Meaning that all knowing infringers from today onward would be liable. The only way you could ignore a cease-and-desist and be okay is if the patent gets rejected.
I agree that this is insanity, but it is not as innocuous as you think. Mr. Knight can begin sending out the cease-and-desists right now, with the full force of the law behind him.
I hope this makes you mad. I hope it makes you mad enough to take some action. There are a number of things you can do. See some of the other posts at Right to Create and freeculture for things you can do to put an end to some of this non-sense. Letting your representatives and congressmen know how you feel is a good place to start.
Please see the "Features" section of the USPTO explanation of provisional patents: "enables immediate commercial promotion of the invention with greater security against having the invention stolen."
Of course, any 'infringers' of this patent could ignore the cease-and-desist letters sent out by Mr. Knight, but they'd be doing so at great risk, as those actions would be actionable once the patent is awarded.
Does this make any sense? Of course not. But that's our patent system for you.
[*] The first patent on the technology was taken out in 1980. Since that time, roughly 40 academic publications have been authored showing ways to vastly improve the technology. Five or so additional patents had been granted in the 1990s. The implementation I used infringed none of these patents, except perhaps for a claim or two of the patent that was issued in 1980 and had since expired.
The Right to Copy? Hardly.
The other test for obviousness (the one that is somewhat sane) is apparently left for the courts to decide after the patent is granted.
You might also be interested to know that our patent system originated in 15th century England, and had nothing to do with novelty, non-obviousness, or prior art and everything to do with exclusive, state-sponsored monopoly (see A Brief History of Idea Monopoly for details on how such commonly manufactured items as soap, salt, glass, and sailcloth were granted patents).
The striking thing here is that our current patent system is starting to look a lot like the old 15th century English one, where "low quality" patents are granted willy-nilly, punishing the general public by levying a sort of tax on everyone except the owner of the patent. In other words, it seems more and more to have everything to do with exclusive, state-sponsored monopoly and nothing to do with protecting innovation and inventors.
There is an interesting post at Right to Create that discusses Roberts and his penchant for being a strong IP-maximalist (in other words, a weak supporter of the freedom to create and invent), and points out that the Senate Judiciary Committee didn't ask him one question on this topic.
Both of which routinely suggest actions you can take to try to turn this thing around.
Now, of course that all flies in the face of common sense. But that's our patent system. Until we change it, that's what we are stuck with. If you are interested in doing your part to fix it, you should be reading sites like:
Both of which routinely suggest actions you can take to try to turn this thing around.
For a nice collection of material against patents and IP-maximalism, Right to Create is good.
Doesn't change the argument -- just adjust the MTBF I quoted to whatever the real MTBF of the drive is and go on.
Or, do as you say and toss good drives during scheduled maintanence (or at least eBay them :) )
If one were to do a somewhat more sophisticated build-the-danged-thing-and-watch-the-failures, one would be hard pressed to find a bathtub curve anywhere at all in the data. Your analysis appears sound, but analysis is often misleading.
> I think what you are referring to is how multiple observations of a uniformly distributed stochastic variable generally look. It doesn't have anything to do with fractals, though.
You statistical guys always want to avoid self-similarity! :) I really do think, though, that you'd find a self-similar model to be more accurate representation of the combined failure rates of such a system than a multiple-observations-of-a-uniformly-distributed-s tochastic-variable model.
Not to nitpick back at you or anything, but have you ever sat in front of a system with 100s of cheap-off-the-shelf drives and recorded the failure times? I'll be a monkey's uncle if they aren't self-similar.