Slashdot Mirror
Patent Office Proposes Reform
leabre writes "The NY Times (subscription required) is running a story about how the Patent and trademark office is trying to reform itself. Among some of the reforms sought, is higher fees for the initial processing fee, higher fees for more than 20 claims, higher fees for the more work the examiners have to due (lower fees for less work and fewer claims), 2000 more examiners, and required continued relevance of the examiner in their field (certification and re-certification). My favorite quote "...Mr. Rogan says excessive claims not only slow patent processing but contribute to poor-quality patents." They are trying to crack down on abundant claims and too-technical jargon which they claim overworks the examiners, reduces the quality of the patent, and other things. Worth a read."
Higher fees means more income for them and less work. The major beneficiaries of the patent system is the patent offices themselves.
... here.
Higher fees are not a solution. They just raise the stakes, so companies will try even harder to win patents. Likewise, or contrastingly, the "little guy" who comes up with a legitimate invention is even less likely to be able to win a patent for it. These "reforms" will serve only to line the bureaucrats' pockets with the blood of the independent inventor.
Karma: Good (despite my invention of the Karma: sig)
A Method For Reforming A Patent Office
doh!
So, it'll be run like a dot com? When's the IPO?
If you were blocking sigs, you wouldn't have to read this.
Ok, it's a big "IF", but if they spent the money on better prior art searches, perhaps it might improve the system. For example, in my experience with patent submission, the US patent office only appears to search for prior art in its own published patent database, while say, in Europe, the EPO seems to look a bit harder.
Of course, extending it to looking through well-known journals relating to the particular art would be even better, but just looking at foreign patent databases (relative to the USA) would be a start.
Notice how there is no mention of changing the process for "business process" patents, like the Bezos "One-Click" and now infamous "eBay" patents.
Raising the fees only help big corporations, which of course want to patent everything under the Sun, probablly including the Sun, just like BT's frivilous patent on "links".
There needs to be some sort of improvement in prior art review. How come a couple thousand of us /.ers can find prior art, but the USPTO can't even use Google?????
Patents we devised to be accesible to the small guy and were designed to help increase innovation. Now they are used as ways for big corporation to squash people from even thinking, and the DCMA only adds to that.
D.O.U.O.S.V.A.V.V.M.
Bigger fees means nothing to corporations but is very hurtful to individuals. This is not a step forward for the rights of individuals, just another leg up for corporations that will do little for the quality of patents. It might stop the 1% most absurd, that's all.
The PO's reforms seem to consist of just charging higher fees. Have they been reading from the Microsoft Guide to Innovation?
But he's not going far enough. Simply charging more for more patent filings isn't going to stop the companies that exist to do nothing other than file patents. They'll just factor it into their business models and pass the costs on to licencees.
What the PTO needs to do is to charge punitive fees when they reject patents. Yes, you heard me. Currently, they get their income from granting patents, so there's absolutely no, zero, zilch, nada incentive to reject, and so there's no disincentive to file.
Let's turn that on its head. Patents should be granted grudgingly. Examiners should be looking for excuses to reject them.
I'd quite seriously propose a deposit of $10,000 for each patent filing, most of which would be refundable on granting. I want filers to be sure that they're actually filing genuine inventions, and I want the PTO examiner (and/or subcontractor) to be eyeing that $10,000 as her reward for finding prior art that you've missed or "forgotten" to mention.
If $10,000 looks like a lot, then consider how many genuinely novel inventions you're likely to have during your working life, and compare that to the number of cars you might buy over that same period. If you still think that's too much for basement inventors, then consider that they can always sell their idea to one of the patent swallowing companies, and we can go back to business as usual.
If you were blocking sigs, you wouldn't have to read this.
And exactly how would all this have prevented the sideways swinging patent #6368227? How much expertise and certification to you need to spot the prior art in THAT one?
"How to Do Nothing," kids activities, back in print!
The Patent & Trademark Office seems to have spent most of its time over the past decade trying to reform itself. With "customer" satisfaction surveys running in the 50 - 60% range, they know they have a problem.
For example, about a year back, they came out with software for electronic submission. Codes things in XML. Nice concept, but the software was virtually unusable. God knows how much money they spent on that. Their flawed electronic search system is another example of ineffective, grossly expensive automation projects.
Another very basic issue is that they seem to lose half the papers people send to them, and then commit significant resources to reviewing and ruling on the proof that the submitter actually sent the papers. This is routine. The most important part of any submission to their office, regretably, is the proof of mailing.
Then there's the touchy issue of quality. Some of the people who work there are highly competent and dedicated. But a lot of them are really inexperienced. Adding 2000 more will just make this worse.
The commissioners (who have been rotating with considerable frequency of late) always say they want to run the office "like a business." Well guess what? It ain't a "business" and it never will be.
They keep talking about their mission to serve "customers," i.e. the people who file patent applications. This is infuriating. They seem totally to forget that the key part of their mission to to represent the PUBLIC. At one point, a past commissioner actually wanted to privatize the office (and make himself the CEO). They should start thinking about what serving the public actually means, and just lose the part about trying to be a "business." At this point, that would be the most useful "reform," in my opinion.
What the applicant would do to avoid complete rejection (and avoid paying your added rejection fee) is just to modify the claims so as to avoid all prior art...
The problem with your scheme is that it chills potential innovation, which is what the original patent idea was all about--grant a monopoly for a few years in return for disclosure of good stuff, heretofore unknown.
What the
THe reason that the USPTO does not search google is b/c most of the USPTO examiners who work on s/w patents are EE's, who may not know enuf background on s/w stuff to be effective at searching google for some patent on web design....so what the uspto needs is more comp sci oriented examiners, but the uspto hiring process is slanted towards engineers....
Sig:
Navy nuke sub lifestyle?
Well, they're off to a bad start on reform if they going to try and take away the technical jargon. That will simply make the process too vague and would allow it to apply to much more than it should be.
The answer is not to dumb things down, it's to hire people that can understand the technical jargon in the first place.
I recently applied to become a patent examiner. I met one of the supervisors who encouraged me to apply, after hearing about my background. Several months later, I learn they're not hiring unless Congress gives them a bigger budget.
This is my sig. There are many like it but this one is mine.
you have no idea do you? The main problem facing the office is pendancy, that is, in some technologies it takes 4 years from the date of filing before a new application reaches an examiner because of the increased number of filings. Hiring 2000 examiners will reduce pendancy because there are only so many cases an examiner can do.
http://www.popa.org is the patent examiners union. Read their critique.
Bring back the old version of slashdot.
WRONG, you didnt read the whole thing didin't you!
There would be a major change in the examination process. Examiners would no longer search for prior art, it would all be submitted by the applicant and the search would be preformed by a private search firm. The examiner then would then determine patentablility via that search.
There are several problems with this:
1. Patent examiners know what prior art is out there because they deal solely in a specific technology area (and are intimatly familiar with those sets of patents of record). What happens if the examiner knows of prior art not listed by the applicant? (This has yet to be addressed.)
2. There is a conflict of interest by the private search firm. They are being paid to find prior art that the applicant really doesn't want them to find.
3. You are removing a function of government by eliminating searching.
for more info, http://www.popa.org
thats the examiners union, im not a member by the way.
Bring back the old version of slashdot.
What Rogan does not explain is why the fees for claims should increase exponentially with the number of fees. If your linear relationship is off, reset the slope dammit! It isn't right to punish an inventor because his invention is complicated.
Fees should probably be higher, the USPTO should be allowed to keep all its fee income and pay the examiners accordingly or supplement the research resources the examiners have. The fee probably should be higher for longer patent applications. (Yes, it is the same fee for 10 pages of description and 1000 pages of description now.)
Patents are granted unless the PTO can find a reason not to. ("A person shall be entitled to a patent unless --" 35 U.S.C. 102) Shorting the examiners of time and resources necessarily leads to bad patents getting out.
Part of the reason that patent application fees are low is that the U.S. Patent system is a "winners pay" system. After the patent is allowed you still have to pay a publication fee, an issue fee, and then maintenance fees at 3, 7, and 11 years. That way, small inventors can take their shot but not bear the full cost unless they get a patent.
What Rogan won't confront, is that the darn points system, which basically is a system where you require the hamsters to run a little faster each year, is not the right management tool (at least in a vacuum). The only thing that supplements it now are "customer satisfaction surveys". Mind you, having the USPTO be polite and responsive is ok, but please notice that QUALITY OF ISSUED PATENTS just does not enter the picture. Please, FIX THE METRICS. If it takes money, which is must, then raise fees. Don't just raise fees to hand money over to Congress though.
A general response to the original thread:
The Patent Office sees about a fifth of 1 billion dollars in fees that it makes for the government.
It seems pretty clear to me having (up-until-recently) worked at the Office, that the PTO is marked as a "cash cow" for the new Homeland Security Office. And why wouldn't it be when it makes all that money for the Feds!?!
I may be a minority on slashdot, but I don't think there is anything wrong with IP in general (whether it be copyrights, patents or trademarks. Yeah, am I definately biased because that is where I make my living. BUT the current system of extending patent and copyright term protection just isn't effective for the fast moving world of computer software and business methods. I think (or maybe, I hope) that over the next several years, the Feds will figure out that software and business methods are essentially without value after a few years on the market and reduce the term protection for those patents to a "reasonble" four or five years (I say "reasonable", because that seems reasonable to me based on what I know).
Thus, the inventor gets a chance to make his money when the invention has real value and the rest of the tech industry can spend the next five years coming up with a "better virtual moustrap."
-A
The only way to get rid of ridiculous patents is to require a minimum of (documented) effort spendt on inventing (in my opinion minimum 12 months of work).
Rationale: If something is so easy to invent that it only takes a week there is absolutely no reason to grant an exclusive worldwide monopoly for that.
Bonus benefit: a single inventor will have little problem documenting all his/her spendings and effort while big multinational companies will probably have some degree of undocumented effort.
When you are sure of something, you probably are wrong (search for "Unskilled and Unaware of It").
You may have some bizarre definition of "not making money". From a 4/01 News.com article:
In other words, the USPTO takes in more money than it spends. The extra goes to other government programs. Now maybe you don't call it "making money", but that's exactly what it is. And the top-level poster is correct that the patent office could be spending more of its revenues checking applications....from an earlier post of mine:
I know how to do away with all of this patent nonsense from here on out.
I'll make a machine that will approve or reject patents, and store them on microfilm. I'd like it to look like something Terry Gilliam would animate. A huge throw switch for accept/reject. An elephant on a treadmill for a source of power. Two rubber stamps, one for approved and one for rejected. A huge bellows to dry the ink. A massive series of lenses, mirrors and candles to reduce the image down to microfilm size.
Then, I'll patent it. If it gets rejected, I'll keep changing components until it passes. Replace the bellows with a cage of pigeons and a box of popcorn and resubmit.
Once I get my shiny new patent, I'll wait one week. Then I'll tack on the words "with a computer" and resubmit. We all know that the magic phrase "with a computer" makes a new patent. Ask Jeff Bezos - he'll tell ya.
Now - it'll be illegal to use a computer to store or approve patents. It's my idea now. The entire process will have to be done by hand. If you want a patent search...well the patents number around the 4,700,000 range. If it takes a minute to read a patent, then it'll take about 20 man years to prove it's original. By then it won't matter.
And just in case the government gets any funny ideas about "prior art" - well we know those lawsuits aren't ever won. Look at Wizards of the Coast. They managed to patent card games for chrissakes. Even though prior art of all kinds exists *cough cough* Steve Jackson *cough*.
But, I'm a reasonable guy. If they press their case strongly enough I'd be willing to settle out of court. Just pay me a nickel royalty for every patent in your database and I'll be okay with that.
Weaselmancer
Weaselmancer
rediculous.
When the government is found to be doing a poor job at something (as it always is, due to its nature) the standard response is to charge more for it-- obviously it must be lack of money that is the problem. So they raise fees.
This is how we have the situation where the average single person pays over %50 of their income in taxes, and still doesn't get adequate fire coverage, adequate roads, adequate health care.
Any REAL patent reform would require taking it out of governments hands and putting it into an entity that has an incentive to provide a good job. For thats is why government sucks-- it has no incentive to do an adequate job, and so it doesn't.
Structure an entity such a way that it makes more money with good patents and its unprofitable to issue bad patents and you'll then have a good patent office.
Until then, the patent office will continue to give the socialists among us an excuse to complain about how "All property is theft".
Yeah, and you guys panned the ipod too: http://apple.slashdot.org/article.pl?sid=01/10/23
1) The airplane.
fortunately for WWI, the government stepped in. Without patent protection, the airplane industry innovated like NO OTHER industry in history within a span of 10 years.
2) The revolver
Colt's patent completely killed all pistol innovation for the period of his patent. All improvements to his design were squashed. The revolver remained unchanged until the patent expired
If you care to look, history is very clear on this. Patents may cause a single spurt of innovation, but ALL innovation on a given invention ceases from the time the patent is given until the time it expires. This is an intentional side effect of patent law.