Slashdot Mirror
Is H.R.1907 Patent Reform that We Want?
"What is Trudel ranting against? I wasn't quite sure after reading his pages. Is his only gripe the publishing of patent applications 18 months after filing, whether or not the patent is granted? What does Trudel stand to gain or lose? Might Trudel have a hidden agenda against other provisions in H.R.1907?
What kind of "reform" are Trudel's antagonists proposing? I couldn't determine this from his rants either. (I haven't read H.R.1907 in its entirety, because frankly, I don't have time.) What do Trudel's antagonists stand to gain or lose?
What are the implications?
Could Trudel be right? Will the "reform" usher us into a world where US inventors and corporations are hurt by foreign concerns? I don't see how.
- If a patent is granted, the patent holder has his US monopoly, even if foreigners could read his application before the monopoly was bestowed.
- If a patent wasn't granted, it can't be argued that the patent application filer has lost anything because the idea was either: prior art, not innovative, or obvious. (Damn bloody obvious with how the PTO grants patents even for obvious methods!)
So what are the real issues?
What does it all mean?"
The only thing I can think of is that this would reduce the effective confidentiality period of an application from 30 months to 18. You see, when you file for a patent in one country, you have a one year grace period to file in the rest of the world, but still get the same effective filing date (known as the "priority date") as on that first application.
Now, the priority date applies for the purposes of patent protection, as well as first-to-file claims in other countries, but it is the actual filing date which determines the date of publication. So, a common tactic is to file in the U.S., wait a year, then file in the rest of the world. This way, you get all the benefits of the U.S. filing date around the world, but the application is not published until 30 months after the U.S. filing date. This would prevent this tactic and require all applications to be published 18 months after the true filing date.
So you can see why corporations might not like this--things they want to keep secret are now revealed a year earlier than they would be otherwise--but it is important to note this is not an issue only to U.S. inventors/corporations. Foreign inventors can and do use this tactic just as well as U.S. inventors, as there is no rule that your first application must be filed in your home country.
I made my comment as a constructive piece of criticism to be worked with, not a catch-all.
My belief is that it takes less resources to secure an equivalent revelation in software systems than in, say, tractor parts. Therefore, the reward (in government supported monopoly over a number of years) should be lower. Exactly how many years should be determined by the how many years the industry can tolerate being without key technologies.
But there is another problem lurking behind the current patent system. Besides the fact that these patent "clerks" get payed nothing and hit with sh*t from lawyers making their annual salary in a week. Besides the fact that Congress has made "business processes" patentable, in violation of the spirit of patent law. Besides the fact that Intelectual property is itself a dubious notion (remember when your ten year-old friends used to say when an idea worked "I thought of it first!").
The real problems stem from truth, logic, and lack thereof in the system of obtaining, enforcing, and repealing patents. Each of these are insanely expensive, abusable, and all too human processes. I mean, there is better logic in playgrounds and Nurseries than in the lawyer's technical arguments over why X should pay Y several billion dollars of hard-earned money.
Submarine patents are obvious abuses, and should not be valid. Ever played boggle? Two people look for words in any direction on a small letter board, and hide their sheets. The most, best words win, and the rule is if two people come up with the same word from the board neither player gets the points. Patents should work like that. If two people independantly come up with an idea, obviously it wasn't patentable to begin with.
-Ben
I'm not sure if I'm reading this right, but I guess in the past your patent term began when you filed the patent? So I guess here you get extra days for every day the patent office is unnecessarily slow in getting it processed. Maybe someone has a better analysis.
I think it was the other way around in the past, ie. the term began when your patent is granted. I remembered reading in Applied Cryptography about the patent stunts pulled out by the NSA and the PTO: the NSA always applies patents for each and every one of their cryptographic inventions, but the USPTO never grants them immediately (so that it need not be disclosed to the public). Then, when somebody outside came up with the same idea and applied for a patent (maybe a good 10 years later), the USPTO reveals the NSA patent, cite it as prior art, grants the NSA patent, and the 17 year clock starts ticking. Real smart, isn't it?
Some patents are granted for inventions that cannot be used without being disclosed (look-and-feel patents, anyone?). ... Since the progress in the useful arts is brought about by disclosure, schemes such as the above which cannot be used without being disclosed should be held to a very high level of scrutiny before any patent is granted.
I'm afraid I don't get you here. Every patent has to be disclosed (aside from some patents with "National Security" implications), when the patent is granted.
Most of the benefits of many inventions is not in the sales, but in the use. (See open-source software.) Techniques such as wavelet compression are far more useful if they are not patented, because the requirement to negotiate use rights is a barrier to their use, and their value is increased by ubiquity; the promise of "progress in the useful arts" is taken back by the barriers to entry.
And it always has been. The purpose of the patent system is not to "promote use of the useful arts", but, as you pointed out, to "promote progress of the useful arts". While the current system, especially the length of the patents and the inability of the PTO to recognize obviousness and intependently research prior art, impedes progress, so would eliminating software patents, as two things would happen. First, commercial funding for software research would be reduced, as the potential of making back the investment would be greatly diminished. Second, what research there was would largely result in work which would remain trade secrets, which impede progress as much as overbroad, overlong patents do.
I think the best solution is to greatly reduce the patent period for software patents to 3 years after award or 5 years after application, and to increase funding for the PTO to adequately research patents. Perhaps the funding could come from a sliding scale for patent application fees based on the applicant's annual revenues, with an additional fee for patents transferred within 1 year of issue (the additional fee being the difference between what the fee would be for the company gaining the patent, minus the amount paid to the original recipient of the patent.
This is really very old news. H.R. 1907 and its Senate counterparts were adopted at the end of the last session, folded into the appropriations bill and passed by a substantial majority of both houses.
The only problems with the provisions are that they didn't go far enough. Patent "reform" is far closer to the position of most Slashdot readers than the strong-IP, "patentee must always win, and anything that weaken's a patentee's rights is bad for America" views espoused by the bill's critics.
The key provisions provided for early publication (which gives companies and programmers a timely heads up of claims for software patents when international patent protection is sought), limited prior user rights for programmers and their employers who had used programs written and in use prior to the issue date of a method of doing business or process patent and a number of administrative changes.
Most significantly, it started to provide greater access to third parties who bring prior art to the patent office to seek reexamination of an issued patent. Reexamination is substantially less expensive, and sometimes far less risky, than litigating the validity of the patent in court. I recently posted in Slashdot concerning the virtues of liberalized reexamination for software inventions.
The subject of this news story represents a faction of so-called "independent inventors" who take a radical, pro-patent stance, arguing that virtually anything that weakens a patent owner's rights compromises the rights of all citizens. I have found advocates from this camp to adopt a "take-all-prisoners" stance, relying on ad hominem attacks and jingoism in lieu of debates on the merits.
While it is true that much of the movement for patent reform is associated with "harmonization" of U.S. patent law with most foreign nations to simpify the process of filing world-wide applications, I will note that even that does not equate to ceding rights of American inventors to foreign entities. Moreover, I will observe that Slashdot critics of software patents have frequently cited to patent policy in Japan and the E.C. as more enlightened than that in the U.S.
Having checked with the company shipping clerk, FedEx, Airborne, UPS and USPS should take a brick wrapped in plain brown paper or in a largish envelope. To send the largish 4 lb, 14 oz brick I rescued from its role as doorstop would cost me $5.81 with Airborne (overnight), $6.60 wih Fedex (2-day), or $6.50 with the USPS. I dare say a thousand bricks/day would grind the USPTO mail room to a halt. It would be even better if none of us included contact information until the last brick. Not only would it leave the USPTO worried, the press would grapple onto it faster because of the 'mystery' angle.
.sig: Now legally binding!
Wrong. The test is supposed to be one's learned peers, an expert in the field. The common individual is not a software engineer, and has no business judging software patents. Nor is he a chemist, an expert in bioengineering or a 'rocket scientist'.
Slashdot readers are (relitivly speaking) experts in software, networking, databases, hardware and e-business. Therefore, we are qualified to make the 'obvious' determination. It's not our fault the USPTO isn't.
.sig: Now legally binding!
But how many people could we get to participate? Say we get 10,000. That's 14,000 letters per day on average, and at a half ounce each, is 3.5 tons of mail.
Perhaps we should Airborne Express them a brick along with a letter that reads 'Judging from your decisions lately, everyone at the USPTO must be as dumb as a brick. Here's some company!/Now you're twice as smart' Our cost would soar to around $25, but they'd have to tolerate almost forty tons of insulting real-world spam. It's sure to make the news as one of the most inventive protests of all time.
.sig: Now legally binding!
You might also want to check out some of Bruce Schneier's work to see if he's patented his encryption algorithms. A quick check of counterpane.com turns up this page on Blowfish, indicating that it is now part of OpenBSD (and almost certainly not patented). Even if RSA did patent theirs, it doesn't mean that they set the standard.
--
Time is Nature's way of keeping everything from happening at once... the bitch.
I have a better idea, how about we DOS the patent office in meatspace, that is, use a modification of Milkman Dan's milk-a-pult to hurl burned-out software engineers at the US Patent Office building until they surrender?
--- Dirtside | "Spirituality" is the irrational belief in the supernatural
"Destroy science and religion. Science would re-emerge exactly the same; but not religion." - Penn Jillette, paraphrased
The flaw in your argument as stated: you are assuming that without a patent, inventors of things will get no compensation. But, many software companies make money from much unpatented software. Visicalc, Lotus 1-2-3, Microsoft BASIC and DOS, Oracle, etc., these things were not patented, and their inventors got plenty of compensation.
To sharpen your argument, I think you mean to say that you feel
Not that I agree with you, but if I'm to argue against the idea I'd at least like it to make sense first ;)
IIRC, Celera is using a different process than HGP, that Celera believes would result in a MUCH faster mapping of the genome. Further, Celera has tried in vain to get the government to fund their approach, and so have whipped out the "we'll patent everything we find first" threat.
Now, whether or not Celera's claims are true, this opens up a whole can of worms:
If discoveries yield patents (either directly or by patenting the means of discovery or means of use), is publicly-funded research competing with potential commercial enterprise? Or are commercial enterprises using public research to gain exclusive advantage? (Note that researchers/universities can obtain patents for the fruits of publicly-funded work.)
Is market incentive or public policy the best way to determine how research is funded? Both have their faults, and both can be biased.
Are patents that claim rights to future/unproven technologies (e.g. cancer cure based on a discovered gene) invalid on the basis of being obvious? (After all, the applicant came up with the idea without even having the expertise or providing the details to make it work). Is the PTO also failing to enforce the necessary level of detail in such patents?
If Celera could really use the discovered genes to cure cancer, doesn't that provide a likely monetary reward, regardless of exclusivity?
Do Celera's tactics demonstrate that patents are only useful as a weapon to keep others away from an innovation? Or are patents the best protection for a small player versus the government or a large player?
"You can't get something for nothing." - my grandfather, on the stock market and Reaganomics.
I apologise for not including this in my original submission but I was forced to disconnect for a brief while.
Celera Genomics have been running a massively expensive mapping of the human genome, and in doing so have been in competition with another state-run mapping.
A little over a month ago, they applied for patent rights on 6500 sections of DNA which they had discovered, which was a very controversial application for the reasons pointed out by other responses. It also should be mentioned that Celera plan to release gene sequences which they discover only to paying subscribers.
Tellingly, one Israeli newspaper has described Celera as 'The New Microsoft'.
"What is freedom of expression? Without the freedom to offend, it ceases to exist." Salman Rushdie
As far as I'm concerned, the only difference between genetic code and computer software (in this context) is that one program is run by biological mechanisms and the other by electronic ones.
Lots of people have been clammoring for the complete abolishment of all software patents, which I don't fully agree with. I think that this system needs a lot of work, and the number of software patents issued needs to be limited (at least) to truly unique, novel algorithms, not silly UI "innovations." Someone mentioned wavelets as an example of what I would consider patentable.
Genetic code should share these limitations, with the additional protection against genetic code found in nature. Of course then what happens when somebody develops a cool checksum/redundancy technique to resist mutation, they receive a patent, and then later on someone else discovers that technique being used elsewhere in nature? Is the patent revoked? Limited? Have there already been discussions or precedents relating to this?
My general impression (which hasn't changed (I'm on page 45 of 116) much since I started reading is that this is a bill by lawyers for lawyers.
The first section (17 pages) deals with "invention promoters", and on the surface, looks good because it creates a disclosure requirement and makes it easier for an inventor to end a contract with a promoter. But (rereading now) I didn't see a whole lot of legal teeth other than the right to sue (which we already had) for damages, and if the promoter was guilty of fraud, it was only a misdemeanor offense. Secondarily, if I'm reading correctly, the inventor only has five days to rescind the contract -- which is not long enough to find out whether or not the promoter can even provide value added service to the inventor.
Pg 19 -- has a definition sentence that is one paragraph long, and I'm still not sure I understand the legal ramifications of it. Pg. 20: requires that a process be reduced to "practice" at least a year prior to the filing of a patent in order to be a defense. Seems dangerous. What if I have developed a process in my business, but haven't documented it to a practice -- and some bozo tries to patent that same process and I don't get word of it. Did I just lose my "prior art" defense against infringement? Sure looks like it.
Pg. 20, bottom: correct me if I'm wrong, but current patent law allows infringement defense based on "substantial improvements" to an item, even if the new item was derived from an earlier patented item. My reading of the text is that this clause is ambiguous and could remove that protection. Pg: 21==>Burden of proof: So far my worst finding... It puts the burden of proof for infringment on the "alleged" infringer. 'xcuse me. I thought under US law a person was innocent until proven guilty. Pg:23==>A successful infringement defense does not necessarily invalidate the patent. Oh yeah? Isn't a successful defense somewhat a proof of "prior art".
Well, enough of that. I'll keep reading, but on first look this is bad.
...Open Source isn't the only answer -- but it's almost always a better value than the alternatives...
I read Mr. Trudel's article in Analog about a month ago; I must say that I thought at the very least it was filled with a lot of historical distortions and inaccuracies.
For example, he credited the authors of the Constitution with developing the first useful patent system, and that this system was responsible for the genesis of rapid technical progress and technology world leadership in the US.
Well, there are some very severe problems with this account. One is that the first effective patent system was devised in England, not the US. A second is that up until at least 1850 or so the US was hardly a world technological power.
Another problem I have with Mr. Trudell is his rant about the US patent system becoming like the Japanese system. In one respect I agree with him - the Japanese system is terrible. It is in fact a corrupt arm of the Keiretsu that use it to maintain their grip on the Japanese economy. HOWEVER, most of the changes that are working their way through the US patent system are actually inspired by standardization that occurred in Europe with the WPO and EPO. These changes, from my point of view are generally beneficial, and solve a number of problems with the US system.
Many people feel that the problems with the Japanese patent system are very detrimental to their development of new technologies. One thing is certain; basic research in Japan is a mess.
If you compare the three systems, you will find that the historical tendency has been in Japan to grant a patent for very small inventions - things I have seen patented there would not make it through the US Patent Office. There are other problems as well; for example the company filing a patent has control of the examining process, and can delay examination for many years.
On the other hand, European patents are generally tougher to get, and require demonstration of a larger step in innovation - something I would like to see in the US process.
I have not read HR 1907 in detail, but if it is directed towards making the US system more compatable with the WPO (and the WPO does publish 18 months after filing) I am all in favor of it.
Geeks should want not just software patents, but also "business process patents" (such as Priceline's reverse auction) to be banned. The spirit of the current patent law seems to already say that "algorithms and laws of nature should not be patentable." But, even though it is clear to each of us that all software programs are long copywritable collections of algorithms, there is such cluelessness in Washington D.C. that they were willing to listen to money rather than to law.
In fact, my humble opinion is that the whole judicial system in America is broken. Perhaps it is also broken elsewhere in the world and never worked at all. But when a judge can take the word a man who makes about $20,000 a year and has to deal with perhaps thousands of patent cases... and hold that over billion dollar industries...
My point is this. In our increasingly litigious society, people are flocking to building whatever fodder for lawsuits and protection from lawsuits they can find. The real heart of the problem is the breakdown of logic, facts, and truth in the judicial system in favor of patents, stories, allegations, and the threat of a lawsuit!
-Ben
I have always believed the simplest way is this:
Since software has just recently become patentable, it should have a short-lived patent system, say a 3 year patent on all software innovations. Later, say in 10 years, congress could then extend the length of patents for longer. This is because software is a new and quickly emerging field, without 17-year business cycles.
Also, any new field (such as bio-technology and quantum computing) should have these same short patent lifetimes, followed by extensions as the market is ready for it. The times could be universally agreed upon by the (patent and non-patent holding) people in the industry.
This is my most reasonable thought on the subject, and it would have, if instigated earlier, quite nicely have kept microsoft from being able to hold its government-enforced software monopoly for its 20 year stranglehold. The old dos technology would have been quickly giving to third parties who would have helped the entire market, like the IBM-compatable computers.
-Ben
(p.s. check out http://www.overlawyered.com)
While I agree that some patents are idiotic (IE the amazon.com one), others are not. There are real advances in computer sciance, and I think that the people thinking them up should get some compensation. Look at wavelet or fractal image compression. I would hardly call those "obvious."
:(
As for hardware patents, well that's the very purpose of patents. I mean, if we had no software patents, or hardware patents, then what the hell would we be able to patent?
I agree that the patent process needs to be reformed, but as far as I can tell those wanting its complete removal from the software feild is mostly from people to stupid to come up with there own cool stuff.
Sorry for my spelling errors, I'm away from a spellchecker
"Suble Mind control? why do html buttons say submit?",
ReadThe ReflectionEngine, a cyberpunk style n
Some of the idiotic software patents are giving patents overall a bad name. The point of patents is protect the little guy, not the mega corporation that everyone seems to think.
Let's say Joe invents the Great Widget in his garage. He begins to start trying to manufacture it (with micro money), but Big Corp sees it, ramps up production and marketing, and takes it over. However, if he gets a patent, he is protected.
Now, the cynical person might be tempted to say, "oh yeah, sure, like Joe can compete with the Big Corp's lawyers." Well, it happens every day, folks. If Joe has a clear patent, and it's a clear violation, Joe makes a lot of money licensing the patent. This means he can get attorneys on contingency.
Now, as for software patents, my thought is that since software engineering is such a new science, the "wheels" so to speak are still being invented. I say we should set an arbitrary date, say 2029, where we say "OK! All the obvious software algorithms have been invented. After that point, clearly it was not obvious and someone deserves to own it." Note that after that date any undiscovered algorithm is going to be extremely complicated (like the recipe for intelligence or something).
Some might argue that software is a mathematical formula, and thus shouldn't be patented. I reject that argument. A hardware machine is a pattern of matter that is organizied in a novel way to do useful work. A software machine is a pattern of electrons that is organized in a novel way to do useful work. What's the difference? Both are patterns; one happens to organize atoms, the other organizes electrons.
---
Although Dry it does have some interesting points>
1) The only sign of patent reform is sec. 705 Which pretty much says the the Patent and Trademark Commisioner has to come up with a study 6 months from passage about why American Biotech interests need to be able to file patents. (Bad thing)
2) Sec 311, "Any person, at any time" we request reexamination. Provide you pay 1) pay the fee, 2) disclose the real parties interested. (Generally a good thing) The director makes a ruling if a new quesiton has been raised. This decision is final. If it is reject they "may" refund the fee.
3) This is the part the Irks me. In sec. 709, National Security and sec. 304, technical clarification there are many claus to how a patent could be kept confidential. This can be at the request of the applicant or under the terms described in sec. 181. Sec. 181 is not published in the text of the bill. (Generally BAD)
4) No where is there any technical reform. Beyond a better appeals process this really is counter productive.
One way in which the patenting system ought to be reformed is to completely and indisputably outrule patenting of sections of the genetic code. Some companies, most notably Celera, have applied for patents on sections of the human genome and if these were passed such actions could be destructive to life-saving research because of the cost of licencing.
"What is freedom of expression? Without the freedom to offend, it ceases to exist." Salman Rushdie
It's 116 pages in the PDF form from the Government Printing Office, but the letters are big and it's a quick read.
Title I - Inventors' Rights
Basically protects inventors from those shady late-night TV commercials promoting "invention" services, where you sign your inventions over to them for the purposes of "evaluating" and "promoting" it, and they keep the patents, royalties, etc.
This is a good thing.
Title II - First Inventor Defense
Protects an inventor from patent infringement charges if the original inventor brought the subject matter to practical use at least a year before the filing of the other guy's patent and used the patent commercially at some point prior to the filing.
It seems to me like this kinda matches my own interpretation of what "prior art" was, but maybe not. Maybe this just makes it more explicit.
Title III - Patent Term Guarantee
Very basically, if the patent office is butt-slow in getting your patent approved, your patent term is adjusted accordingly.
I'm not sure if I'm reading this right, but I guess in the past your patent term began when you filed the patent? So I guess here you get extra days for every day the patent office is unnecessarily slow in getting it processed. Maybe someone has a better analysis.
Title IV - United States Publication of Patent Applications Published Abroad
Presumably this is where all of the fuss occurs. Unfortunately, it's very hard to read for me, and makes extensive references to the various patent treaties.
My understanding is this: Patents are generally published/made available to foreign countries after 18 months. You can request that it be made available earlier, or, if you're not filing patents in a different country, you can request that it not be published at all. If you do file patents in other countries, presumably treaties require these patent applications to be shared between member countries after 18 months.
I think this is what some people are angry about. I don't know the rationale behind the sharing of applications, so maybe somebody else can step up and elaborate for us? I guess by publishing these patent applications, anyone in another country where the patent holder isn't planning on applying for a patent could then in turn patent it himself. Anybody in a country where the US patent wouldn't be legal could use the information in the patent for his own good. I may be totally off-base here, but that's how I read it. Corrections are welcome.
Title V - Patent Litigation Reduction Act (I like it already)
Anybody can write the patent office with a request for re-examination, citing examples of prior art. Their letter will become an official part of the patent, and the patent office must make a determination (whether or not to re-examine) within 3 months. The requester gets copies of the progress.
Title VI - Patent and Trademark Office
This takes up about half of the actual bill. Since I'm not totally familiar with the current structure of the PTO, this could either be a lot of re-wording, or (more likely) a complete restructuring of the office itself. Note that this seems to only apply to the PTO office/organization. It doesn't have anything to do with the types of patents issued, but how they do their day-to-day business, how they evaluate their staff, and pick their work force.
But most importantly, the bill includes wording to bring the PTO out of the dark ages. They're given permission to make use of whatever equipment and technology they need (like broadband) and lets them hire their examiners more competitively.
Here's a link to the most recent version of the bill.
./ers.
I can't make heads or tails about what this guy is complaining about. Probably black helicopters are involved, somehow...
Section 303, requesting an investigation of business model patents, is probably of the most interest to
My Blog. Sela Ward can sell me long distanc
From the U.S. Legal Code here
-CITE-
35 USC Sec. 101 01/26/98
-EXPCITE-
TITLE 35 - PATENTS
PART II - PATENTABILITY OF INVENTIONS AND GRANT OF PATENTS
CHAPTER 10 - PATENTABILITY OF INVENTIONS
-HEAD-
Sec. 101. Inventions patentable
-STATUTE-
Whoever invents or discovers any new and useful process, machine,
manufacture, or composition of matter, or any new and useful
improvement thereof, may obtain a patent therefor, subject to the
conditions and requirements of this title.
-SOURCE-
(July 19, 1952, ch. 950, 66 Stat. 797.)
Go here to find more. I searched for 'discover' in title 35.
I do not believe that discoveries should be patentable, but the current law allows it. Also recall that the US governemnt's web pages are not always up to date, i.e. H.R. 1907 may have changed this in some way.
There are a number of problems with the patent system today. To list a few:
- Some patents are granted for inventions that cannot be used without being disclosed (look-and-feel patents, anyone?). Amazon's one-button patent falls into this category. Since the progress in the useful arts is brought about by disclosure, schemes such as the above which cannot be used without being disclosed should be held to a very high level of scrutiny before any patent is granted. This is certainly not true today.
- Most of the benefits of many inventions is not in the sales, but in the use. (See open-source software.) Techniques such as wavelet compression are far more useful if they are not patented, because the requirement to negotiate use rights is a barrier to their use, and their value is increased by ubiquity; the promise of "progress in the useful arts" is taken back by the barriers to entry.
- The USPTO considers "prior art" to be that which has been previously granted a patent, and precious little else. 'nuff said.
I've got my name on three, count 'em, three software patents. I'd like to see every last one of them invalidated, because I think that everyone should be able to build better stuff without having to jump through hoops to do it or worry about stepping on a legal land mine when they are trying to do engineering.--
Time is Nature's way of keeping everything from happening at once... the bitch.