Schneier on Security
A blog covering security and security technology.
« How Antivirus Companies Handle State-Sponsored Malware |
| Keeping Track of All the Snowden Documents »
December 2, 2013
The TQP Patent
One of the things I do is expert witness work in patent litigations. Often, it's defending companies against patent trolls. One of the patents I have worked on for several defendants is owned by a company called TQP Development. The patent owner claims that it covers SSL and RC4, which it does not. The patent owner claims that the patent is novel, which it is not. Despite this, TQP has managed to make $45 million off the patent, almost entirely as a result of private settlements. One company, Newegg, fought and lost -- although they're planning to appeal. The story is here.
There is legislation pending in the U.S. to help stop patent trolls. Help support it.
Posted on December 2, 2013 at 12:48 PM
• 49 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
This is definitely a start, and hopefully it passes; I think more is needed to be done, however. Patents are destroying innovation; 20 years is an eternity in the computer world, and I'd argue we'd be better off without patents entirely these days, although 5-7 years would be significantly more reasonable.
Anura - that doesn't make any sense. If a patent isn't worth much after a few years, why do you care?
The fact is, USPTO allowance errors remain a tiny minority of issued patents. Go ahead and subject yourself to the process. You'll find out real quick how nearly impossible it is to find an examiner who will even read or can understand applications anymore. Do you have any idea how long it takes to get a patent anymore, or how much money it costs. Stop listening to these people.
I know, you could have been a millionaire by now if only those trolls hadn't gotten in your way.
If a patent isn't worth much after a few years, why do you care?
The truth of this case is that TQP's patent is invalid.
However, a jury is incapable of discerning the truth of a case, which is supposedly the jury's job. Why? Because before the trial starts, the lawyer for one side or the other makes sure to eliminate any potential juror who knows anything about the subject matter. Such a juror would see that the lawyer's case is b.s. and would convince the other jurors likewise.
TQP won the case, not because their patent is valid, but because the jury wasn't capable of discerning the truth.
Newegg says they will appeal the verdict. The appeal will be heard by a panel of judges, not by a jury of ignorant citizens. The judges are much more likely to discern the truth of the case and throw out the patent.
I would much rather patents simply be eliminated and let first mover advantage rule, but this bill is certainly better than the status quo so it seems a no-brainer to support it. Fee-shifting, however, really needs to come along with an insurance system available to small-time patent owners taking legitimate enforcement action, or it simply shifts the balance too far in the other direction.
It was not just any jury. Apparently that Texas county has specialized in patent cases, and is making a very good living off it. Don't bite the hand that feeds you.
The proposed legislation is a good first step, but by far not enough.
As long as software patents are not entirely abandoned, there will be no fair game and the problems will stay.
it should not be allowed to settle a patent lawsuit out of court. the entity suing for infringement should be required to pay triple the defendant's legal costs plus the amount being sued for if such entity (troll) loses in court.
of course another way to do this would be to instantly invalidate *all* software patents. stallman has discussed this in great and excruciating detail much better than I can: http://www.gnu.org/philosophy/... the gist of it is software and hardware are different. copyright protects the text of software just fine, but copyright and patent are totally different concepts.
@supersaurus - not quite as cut and dried.
I implement an invention in brass gear wheels=patentable, I update it to electronic circuits=patentable, I implement those circuits in an ASIC=patentable, I do it in an FPGA=software and only copyright. So anyone is free to copy the code, rewrite it and sell it, as Compaq did with IBM's PC bios.
The problem with most of these "software patents" isn't that they are software - but that they are really either mathematical algorithms, business methods or artistic designs - and are patented simply because the law is more advantageous.
This isn't something I know much about, so I've got some (non-leading, non-snarky, honest) questions...
If these "patent trolls" are buying patents to begin with, doesn't that mean that they *can* serve a useful function?
Joe inventor may be a clever bastard with a valid patent...what if he can't afford to protect it when it's infringed? Does it not make sense that he would want to get at least some compensation for having done the brain work in the first place? If the only way he can capitalize on his innovation is to sell the patent to an aggregator who sees its value, isn't it good that he at least has the option to do so?
I can grasp the reality, that some aggregators are shifty-eyed jerks bringing frivolous lawsuits, but ideally they *could* be serving a legitimate purpose if they went about it differently, no?
@aboniks - that's the problem with all the attempts at "fixing" the law.
1, Don't allow NPE (non-producing entities) to hold patents. So ARM can't patent its chips because it doesn't fab them?
2, Make the "troll" pay costs. So the lone inventor is liable for all of IBM's legal costs if they lose?
3, Only allow the original inventor to hold a patent. So do companies allow their employees to own the patent? How much is the patent worth to the inventor or their heirs if they can't sell it?
It's very difficult to write a law that only applies to the bad guys.
The problem is that patents are getting more abstract and more trivial, to the point where no one can really develop any technology without worrying about patents. You patent a specicifc type of brakes with the intent to sell to auto manufacturers, great. If instead you patent the concept of using friction to stop movement, this is a problem. The latter is what's happening today; the system is getting flooded with trivial patents, but the fields they apply to are so specialized that a system that could actually work to determine whether they are novel does not, and probably cannot, exist.
As technology becomes more advanced, as patents become more abstract, as they get to the point where you can't actually make something in any new market, e.g. cell phones, without having to license thousands of patents (and still face lawsuits for all the ones you didn't know about at the time), then patents are no longer protecting innovation, just hurting it.
The trolls are one part of the problem, but the entire system is obsolete.
The problem is not so much that people abuse the patent process but the legislators are encoraged by those with deep pockets not to fix the issues.
Most people agree that people who come up with original ideas should have some way of recuperating costs and losses of turning an idea into a reality and some kind of ongoing income from their originality and effort to make it work. This is not an unexpected viewpoint and even those who might disagree for ideological reasons realise that society would be poorer if originality that benifits societyy was not protected in some way.
The question then arises as to how you do this...
Without going into the history of it since Queen Elizabeth the first, it's clear that the process went wrong somewhere.
The underlying issue as to why it went wrong is that legislation and regulation creates artificial markets and this was not realy realised untill around a hundred years or so ago.
The problem with any market is although it's not a zero sum game most people involved treat it as such which gives rise to undesirable consiquences, which actualy hurt all players in the market be they inventors, manufactures, consumers or the tax taking government.
Invention is a "societal good" that all benifit from, and it is thus better to not create legislation that makes invention an "individual good" which is what currently happens. Nor does a lack of regulation or a "free market" benifit society due to the inevitable "race to the bottom" of all free markets (the cost of the distance vector being the only real limitation on this, removal of which is why Internet Markets don't work the way assumed).
The trick is to find legislation that encorages society benifiting activity not legislation that constrains benifit to a few individuals.
Some years ago the idea of an open royalalty model was moted, but as has been seen with the entertainment industry the issue of individual benift becomes an institutional not societal benifit.
Various standards institutes tried to implement this by saying that all algorithms put forward as part of a standard would be treated either "institutionaly" or later when that was found not to work as "unencumbered". Which has likewise been found not to work when large organisations not individuals are involved. Either way society did not benifit.
Society only appears to benifit when the inventor does not take direct financial benifit. We have seen this with the NIST competitions and with the various Open Software initiatives. But man cannot live by fresh air and plaudits, inventors need to get benifit for their work or else they would chose to do some other endevor.
Which leaves the question of how do you protect inventors and give them reward whilst also protecting society from those who wish to pervert such a process for their own benifit at the expense of both the inventor and society as a whole?
And it's not an easy question to answer as at every turn there is somebody looking to become a "troll" by fair means or foul, including buying off legislators.
There's another issue as well, in that advances in technology happen exponentially, so while patents at one point may have helped things grow when technology was moving more slowly, major technological advances are happening within a matter of years rather than decades. I would conjecture that as technological advancements are more rapid, it gets to the point where patents only slow down innovation, and that we have already passed that point.
The original intention of patents was to allow new ideas to be published and added to the common knowledge for use after a decent interval had passed to allow the original inventor to extract some value. It's clear that this is now a rarity.
In the ideal case today, patents act merely as an expensive government witnessed NDA so a lone inventor can show their idea to $MEGACROP without it being immediately stolen.
In reality they are either used by trolls to extort money from $MEGACORP or for a cartel to lock up a market with patent encumbered standards.
I followed this case through the ARS reporting of it.
Although their bias is evident given their general coverage of patent trolls, I was surprised with the verdict, especially after Diffie's testimony.
"To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries."
-- The United States Constitution, Article 1, Section 8, Clause 8
The current problem is that patents are being issued for things that aren't new or novel. The USPTO is issuing patents for, essentially, the wheel. (They did allow a patent to Apple for the rectangle, and to IBM for abbreviations and the index.) Patent holders are grossly stretching what their patents could possibly cover, and suing world+dog for about half of what it would cost to fight them.
The problem of the lone inventor fighting the corporate giant is typified in Robert Kearns' plight. He invented the intermittent wiper control, and then had to sue the Big Three auto makers to make them pay. I do imagine that we still have that problem, on occasion. For that, I do support patents, just not the racketeering that's going on now using patents.
Support the RICO act!
In a parallel universe, some merciless crypto god would have instantly struck down by lightning the idiot lawyer questioning my former CSO Whit Diffie. For anyone even remotely familiar with the case and the subject at hand, there simply is no case. It's pretty much unbelievable that a jury of laymen can be deemed qualified by anyone to judge over this kind of highly specialised matters. I hope Newegg wins on appeal, the Innovation Act passes and all patent trolls are finally banned to the 8th circle of hell where they belong.
Invention is a "societal good" that all benifit from, and it is thus better to not create legislation that makes invention an "individual good" which is what currently happens.
Metaphysically speaking, society is an abstraction, and only individuals actually exist. It follows that a term such as "societal good" ascribes a property to a non-existent being.
Individuals have the capacity to value things as good because they ultimately face the alternative of dying if they don't choose the means required to live. A society has no such natural basis or capacity for decision making.
There are many ways to draw boundaries around individuals in order to regard them as a society of one kind or another.
To avoid contradictions, a general solution to the patent problem needs to follow policies that respect and affirm individual property rights.
The problem in this case seems to have less to do with patent validity (which is what several comments have focused on) and more on the issue of whether the patent is applicable to the use of RC4 in SSL.
This particular case seems more about a failure in the court system than a failure at the patent office. The patent pretty clearly talks about a specific way to use a stream cipher to encrypt modem-to-modem communications. Whether or not this is a particularly valid patent to have granted in 1989 (my guess is not), it's even more ridiculous to claim that it applies to using RC4 and SSL to encrypt web traffic. In other words, less a case of an overly broad patent being granted and more a case of the court siding with an overly broad interpretation of the patent. And when you think about it, that's potentially an even more serious patent issue.
That said, I think Newegg's defense wasn't as good as it could have been. Their case seemed to be mainly about how SSL used public key cryptography (which the patent didn't mention) and how RC4 was invented slightly before the patent. Fair enough, but that ignores the fact that the claim is basically that the patent covers the invention of stream cipher cryptography in 1989. This seems like a pretty easy claim to discredit without having to bring in Diffie or discussing the history of RC4.
Coming from India, where there's a tad more corruption than most other countries (or at least it's more visible), I naturally wonder if that is the explanation for the inexplicable decisions that the East Texas District Court seems to come up with.
Since no one else seems to have come up with this, perhaps I am completely wrong.
@ Tim L,
Metaphysically speaking, society is an abstraction, and only individuals actually exist. It follows that a term such as "societal good" ascribes a property to a non-existent being
Biologicaly speaking taken at one level/viewpoint multi-cellular organisms are an abstraction, and only the individual cells exist, likewise bees, ants and other colonies only the individual insects exist. However taken at another level the cells interact to produce more complex behaviour than individual cells are capable of and likewise the individuals in colonies collectivly behave in much more complex ways than the individuals can. Thus it can be clearly seen that individuals can collectivly be rather more than the sum of individual parts. You only have to look at your environment around you to realise this. So the "metaphysical argument" is at best moot.
Individuals have the capacity to value things as good because they ultimately face the alternative of dying if they don't choose the means required to live. A society has no such natural basis or capacity for decision making.
Sorry that argument is false. Societies live and die by the collective actions desires of the individuals with regards to other individuals and the environment. Such individual actions/desires on mass become the societal rules both informal (morals) or formal (laws). To try to argue that a society does not live or die defies the evidence of history.
There are many ways to draw boundaries around individuals in order to regard them as a society of one kind or another
It's a fundemental statment, but as in set theory it has underlying assumptions that make it context sensitive in use and outcome, and you have not realy provided a context for it's use. For instance it can be used to argue that in general terms only three numbers or viewpoints make sense, Zero, One and Infinity. That is there can be "none of something", "something can be unique" or "there can be many somethings the number of which is unknown (and in effect unbounded)".
Which brings us onto,
To avoid contradictions, a general solution to the patent problem needs to follow policies that respect and affirm individual property rights
As an argument it fails on your assumption that there will be contradictions, and thus argue eroniously to your view point that "individual property rights" are the only possible course.
It so happens that Patent law recognises higher social rights by way of "National Security" which you should be aware has been used to nullify "individual property rights".
But further consider the difference between primary or original patents and secondary or derivative patents. This can give rise to a deadlock senario, you hold a primary patent but I hold a derivative patent that prevents you exploiting your "individual property rights" because it conflicts with my "individual property rights". In effect your individual proprty rights have been nullified by mine and it can be clearly seen that the solution you propose is actually just as subject to contradictions as the current system of patents.
no one can really develop any technology without worrying about patents
--Take a chance for one in your life, who cares. It really depends on the context of the invention though.
An observation I have of someone near and dear to me, is he grew up in a rural place and had freedom to do almost anything he wanted. I want to conduct some experiments but I can't b/c I will have someone call the coppers on me or I will interfere w/ some people's signals (I still manage some b/c they're holding me back). Instead children today are continually having cops in schools, cameras in the ceilings, walk in a single file line, do not get out of line, do not walk on the grass...and it's killing original thought and creativity, I know it is...
Until the end of the 20th century, all formal methods, that is mathematics, were excluded from patents. Including software and business methods made a complete mess of the system. Because a business methods/software patent is a patent on an idea. Such ideas do not have good boundaries.
The other part is USA only. Defendants cannot recuperate legal costs even when they win. So there is little downside to suing. In the rest of the world, a patent troll will have to pay legal costs whenever they lose. That changes the equation enough that patent trolls are rare and far between outside of the USA.
To avoid contradictions, a general solution to the patent problem needs to follow policies that respect and affirm individual property rights.
Metaphysically speaking, "individual property rights" is an abstraction, and only an individual's power to retain possession actually exists.
Your argument is kind of bizarre. You want "society", the abstraction which you choose to deny, to grant and enforce monopolies granted to individuals, while claiming that such enforcement is a "property right", the abstraction which you choose to affirm.
I agree with the statement that RC4 and SSL/TLS do not infringe upon the patent, but I disagree with the statement that the patent is invalid (unless one considers it obvious and that's too easy to do with hindsight). The patent is simply about using different keys for encrypting blocks in a block cipher where the changing keys are generated via synchronized pseudo-random number generators on each side of the communication triggered by counting blocks (amount of data transmitted/received). Block ciphers have problems when each block is encrypted using the same key (identical plaintext results in identical ciphertext) and this is why the Electronic Codebook (ECB) mode is not the only mode available. Cipher Block Chaining (CBC) forces each block to produce different ciphertext even when the plaintext is identical into each block. The Cipher Feedback (CFB), Output Feedback (OFB), and Counter (CTR) modes all turn the block cipher algorithm into a stream cipher where the block cipher just becomes a key stream generator.
Anyway, the idea of changing the key used in a block cipher is not specified in prior art including that presented at trial. You need to actually read the Denning book "Cryptography and Data Security" which is found here:
The book says that "A block cipher breaks M [the message] into successive blocks M1, M2, ..., and enciphers each M1 with the same K" and does not talk about changing keys for different blocks, nor generating such changed keys by using synchronized pseudo-random number generators, nor triggering the change after a predetermined number of blocks (amount of transmitted/received data). There is some discussion about using different keys for different records in a database and of course there is discussion about how stream ciphers generate a continuous stream of keys (bits if the key stream and plaintext are XOR'd).
The 5,412,730 patent went through a USPTO re-examination requested by TD Ameritrade after reviewing "150 additional prior art documents not considered during the original patent examination" and confirmed patentability of all original claims, and added several new claims:
The Newegg defense attorneys were so focused at the trial on proving patent invalidity that they hopelessly confused the jury regarding infringement. They not only referred to Denning's book as prior art when as shown above it really isn't, but they also claimed that RC4 was prior art when RC4 is a stream cipher that has nothing to do with the patent. You can't claim RC4 to be prior art for the patent at the same time as claiming that RC4 does not infringe because the patent doesn't relate to it. No wonder the jury was so confused.
The Newegg defense attorneys also told the jury “In today's systems, the keys change every time a block is encrypted. There's nothing to count!”, but that statement was incorrect because as I noted earlier block ciphers don't change their keys (for a given session) and stream ciphers by definition change their keys for every bit (when using XOR). The patent would read upon a stream cipher if the seed key were changed after so many "blocks" (i.e. amount transmitted/received) and such new seed keys generated via synchronized pseudo-random number generators, but as far as I know no one does this today.
As for RC4, this is a stream cipher and stream ciphers were used in mechanical machines by Hagelin in the 1930's, by the German Enigma machine in commercial use in the 1920's and used during WW II in the 1940's. They were then used in computers using vacuum tubes and then in the 1960's in transistor-based computers that used Linear Feedback Shift Registers (LFSR) as pseudo-random number generators that produced one bit of key stream at a time. RC4 was the first stream cipher to do byte manipulations in memory to produce one byte (8 bits) of key stream at a time, but the key stream is still applied to encrypt bits individually (that is, there are no "blocks").
As for SSL/TLS, these do not infringe because the session keys (including encryption keys) do not change during the session and new sessions are not initiated at any predetermined amount of transmitted/received data.
So again, the patent is likely valid and neither RC4 nor SSL/TLS nor their combination infringe.
I must confess to feeling bizarre about this whole topic. You see, I wrote a short story, "flash fiction" of less than 500 words, titled "Mephistopheles in Silicon which I submitted to AntipodeanSF about August this year, following the death of someone highly respected and liked on the Yahoo Viola email list. My story was about something that suddenly seemed obvious to me - that with a suitably advanced set of AI functions and a suitably large collection of data of human responses, emulating human responses would become trivial.
And then last week Slashdot points me towards an article titled Google Wants to Write Your Social Media Messages For You. It's eerie:
Mephistopheles in Silicon
"We've got an advanced AI that will follow all your lines of thought as expressed in your online postings, and produce arguments that you yourself would find indistinguishable from your own. So your passing will not leave your friends bereaved. Even better, it will take your patterns of composition and produce compositions that are like your current ones, but which will develop with time, and which will be published under your name, thus ensuring a constant income for your loved ones. All we need is your signature on the relevant documents, and your first payment, and the appropriate passwords, and we will set it up for you."
Google Wants to Write Your Social Media Messages For You
Essentially, the program analyzes the messages a user makes through social networks, email, text messaging, microblogging, and other systems. Then, the program offers suggestions for responses, where the original messages are displayed, with information about others reactions to the same messages, and then the user can send the suggested messages in response to those users. The more the user utilizes the program and uses the responses, the more the bot can narrow down the types of responses you make.
So, the only difference is that the salesdroid in my story doesn't depend on interaction with the end-user to certify his responses in order to respond, once he has the signatures, unlike Google's patented software.
And I consider my story trivial. I consider the speculation that led to the story, trivial - heck, Bruce's been discussing public and private surveillance for ages on this blog. And with my neuroscience background I place a lot more trust in emergent properties than in designed "AI".
It's just that now I've lapped Google Corp without intending to "forecast" anything, and it feels eerie.
I think you're missing the point here. 20 years are not too long because the invention would be worthless by then, 20 years are too long because it is a lot of time for key technologies to completely block out competition, and technology is always based on previous advances. If you can't get a critical part for your product (say, a computer), and have to use something from almost 20 years ago instead (say, a 25 MHz CPU), the whole thing will be useless - even if the patent only covers a small part of it. And you have no incentive to develop anything based on patented technology, because you can't use it anyway.
Luckily, most large companies are reasonable enough to trade parts (so we at least get usable products), but this still makes it practically impossible for new players to establish on the market, and limits the possibilities for new developments.
I need to add more info regarding my statement that neither RC4 nor SSL/TLS infringe upon the patent. As I noted, if one were to change the seed key for a stream cipher or the key for a block cipher after a predetermined number of transmitted/received bytes and that the new keys were generated independently without transmission by synchronized pseudo-random number generators, then that would be infringing upon the patent. Note the following:
“Renegotiate Size” in
So one can apparently define a size of transmitted/received plaintext for triggering a renegotiation that will generate new keys.
This appears to be an implementation-dependent issue, not something necessarily in every TSL implementation as I could only find such info in the above references for the IBM and BIG-IP from f5 SSL/TSL implementations. Renegotiation is defined in SSL and TSL, but how often to renegotiate and specifically basing it on counting the amount of transmission appears to depend on the specific SSL/TSL implementation.
This still should not be infringing upon the patent because the renegotiation requires creation of a new premaster secret key that must be transmitted and that is done using asymmetric public key cryptography. The patent explicitly requires that there is no transmission of keys and that they instead are generated via synchronized pseudo-random number generators on each side. Now it is true that the premaster secret that is transmitted then results in both sides generating a master secret and then session keys (including encryption keys), but this is for the session as a whole and such independent pseudo-random number generation is not done for subsequent keys (as there are no subsequent keys).
This is what makes SSL/TLS vs. the patent confusing because of these extra elements of renegotiation based on amount transmitted/received and because of the identical pseudo-random number generation of session keys (though based on a premaster secret that must be transmitted).
Form letters from authorities should probably begin with the words:
Dear Establishment Unit,
And some hazy recollection I have was some pack of ****s trying to patent the XOR operation. That someone even considered something so foolish was a warning bell that the system was broken or at least badly injured.
Correction: renegotiation does not create a new premaster secret key because there is already a secure connection available. Instead, new encryption keys are generated and exchanged via the existing secure connection. The generation of the keys may be with pseudo-random number generators but having them exchanged seems like a gray area with regard to the patent. For the patent, there should be no need for exchange since synchronized random number generators on the two sides should be able to generate the same keys without the need for transmitting them.
TQP is violating my Business Methods patent on Patent Trolling!
Like the attempts to rein in the NSA, this is a good cause but one with a longer history than many people realize. I recently read a book from 1934 which had much the same complaints about the patent system. Navigating a thicket of patents just to get one piece of technology to market? It had an example of a mechanical loom which was subject to 80 different patents. Patent trolling? It mentions "the special vehicle for exploiting patent monopolies" among types of predatory capitalism, and then says nothing further about it, because apparently the contemporary reader was familiar with that sort of thing.
Considering how a patent also discloses information, sometimes it can be a better idea to not to file for a patent and just go the "trade secret through obscurity" route instead.
trying to patent the XOR operation
--Unbelievable...Someone should patent the patent process.
"Someone should patent the patent process."
Actually, I was going to do that myself but I was certain someone tried/failed already. Turns out it was Haliburton:
Patenting the surveillance process - or various techniques used in it - would be great if it provided a means to severely limit all the surveillance
@ Figureitout, Nick P,
The one process you should try and patent is that of "lobbying"... there's so much money sloshing around in that game that a licence fee of 0.1 cents on the dollar would give you more money than you could honestly find a use for.
--That is truly disgusting...
--I want to feel a sense of pride earning my money, witnessing that garbage during an internship at a statehouse was all I needed to see that I was done w/ public affairs. That and having to have police protection everywhere during the legislative session b/c we don't want to face the angry constituents.
But the purpose of patents is not to protect the patent holder or the invention, it's actually intended to protect society. The rest of us.
The standard of disclosure in a patent is that an expert at the subject matter can duplicate the invention.
This disclosure and the limited protection period of the patent is to encourage inventors to disclose their invention, instead of keeping it secret.
This benefits society, after the protection period anyone is welcome to use the invention. Much better than inventions kept secret. For the rest of us.
That said, I think most software patents are ridiculous, and the patent office does not nearly have the expertise to examine those applications. They should just stop.
An example of a valid reason for a patent is in medical technology, drugs for example. It can cost hundreds of millions of dollars to invent and test a drug to treat disease. What company could afford to develop and bring a drug to market if it were appropriated immediately by their competition who spent nothing to develop or test it and could then undercut the price?
Those patents encourage incredible inventions, which would not be able to be invented without an appropriate patent protection period.
--True, but patents on math aren't in my interest; in fact that makes you an enemy in my view (I take that personally, I'm sure other mathematicians probably more so).
Also drug patents, there's some incredible drugs no doubt; but there's some evil of course when the drugs can be produced cheaper yet someone's life doesn't take precedence over a drug company's profits.
Also mass surveillance that would scoop up all that R&D anyway makes me personally, likely some others, to not even waste time if your work is going to be stolen instantaneously and someone claim credit.
This popped up on Reddit, http://www.usatoday.com/story/news/nation-now/...
This is embarrassing...
Shows just how dumbed-down we're becoming, that this clown of a journalist is calling the guy a "visionary" for a taco bell taco shell. And of course taco bell gave only $1k as he died from brain cancer. People then went and ate some more taco bell in his honor...
Figureitout, without some period of exclusivity drugs would not be developed.
No company would spend money on research, testing, clinical trials if their intellectual property were immediately stolen by their competitors.
If you consider drug company profits to be evil think of what the world would look like if there were no drug companies, no drug patents and no profits.
It would be much worse, wouldn't it?
Certainly we would all like for everyone to live a happy life and all diseases to be cured, but there has to be a profit motive to spur the research.
Without that it would be like you and I working at a soviet union bicycle factory. You can make 20 bicycles a day, and I can only make 4, but we are both paid the same.
Why should you make that many again? Just because you can? I don't think so.
You make decent points but your argument makes two assumptions I think are unwarranted:
1. All drug patents must be treated equally.
2. Allowing drug patents and profits = current patent system.
Neither is true. Here's my proposal:
1. Reduce cost of FDA evaluations as much as is practical.
2. Reduce number of years a drug patent is valid.
3. Nationalize them with profit.
The first option the drug companies are all for as it increases their profits. The second option is one compromise I came up with. The idea is that we need to give them enough monopoly and ROI to develop/market the drugs, but not more than that. I was trying to figure out how long when I decided it might be better to measure how long it's in the marketplace rather than how long it exists. Being necessities rather than luxuries, people will buy good ones the second they're available. So, 20 years of guaranteed sales seems excessive. I'm thinking more 5-15 depending on what a thorough analysis of the data says about ROI. (to be fair to the companies)
Number 3 is my own original solution. People in this debate forget that laws were originally created to protect people and in the US the people can push for new laws. People might push for the government to seize intellectual property that is priced too high and considered essential. I'm going to pick on cancer drugs here because there's just a few, it's one of the leading causes of death, and the drugs are *ridiculously* expensive.
My proposal isn't necessarily to seize the I.P. Instead, the government would force the company to sell it. They'd get their full investment into it (including FDA evaluation) plus a percentage for their profit, maybe even 100%. Only certain essential drugs would be treated this way. The people would get what they want (and pay in taxes). The businesses would still have plenty incentive as all they'd have to do is invent it rather than market and distribute it. Once the government bought it, all manufacturers in the US would instantly be granted a license to it and free market (with some safety regs) takes over from there. The end result is certain widespread problems have cheap solutions without making drug companies want to stop investing.
What do you think about my socialist + capitalist option? I mean, not the likelihood of adoption (almost zero). More the effectiveness if adopted and business argument.
I gave an example of a patent that does serve it's purpose. Unlike software patents which often serve no purpose.
It can easily cost hundreds of millions of dollars to conduct research, clinical trials, and get a drug approved. Some medical patents expire after 17 years, before the drug can be tested and approved. That can be very costly. Business risk.
You indeed make some socialist proposals. We live in a capitalist society, and it's no secret that many people from all over the world would rather move to the US if they could. Perhaps billions of people. So something must be better in their view. I don't much want to muck with that.
As to streamlining the FDA, no argument there. But though our FDA might indeed be too conservative and their attitude and caution makes things too expensive, there needs to be a balance. Where that balance is, well, remember Thalidomide? That was a great drug for pregnant women. For 8 or 9 months. So caution is in order.
I do think the drug patent system is working pretty well, and i do think most research does take place because of the profit motive.
I don't feel nearly as good about software patents, I imagine we have some common ground there.
Perhaps non practicing entities should not have standing to sue. That would be some interesting case law. I would like to see that argued.
Schneier.com is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc.