Slashdot Mirror
Inside Look at Patent Examination
KingFatty writes "This article written by a former patent examiner describes patent application as a matter of luck when it comes to the competency of the examiner. "Every examiner starts with his or her first patent application after receiving just two weeks of training at the USPTO Patent Academy, where he or she learns the basics of the Manual of Patent Examining Procedure. Will your patent application be examined by that newly recruited examiner? If so, will the examiner's supervisor (supervisory patent examiner or SPE)[be] sufficiently skilled in the art in which the patent application is classified?" Gives insights as to the problems with the US Patent and Trademark System."
Why aren't patents exposed to peer review, like academic articles are? The invention (if it is) will be protected by the patent pending laws while it's reviewed.
It's quite clear to me that most of the problems in our patent system were caused by Albert Einstein.
Now patent clerks everywhere do half-assed work, their brains busy trying to come up with the next great theory of time, gravity, and light.
There are only 10 types of people: those who understand decimal, those who don't, and, uh, 8 other types I forget.
I feel the largest problem with the patent system is allowing people to patent "ideas" without a working prototype. I think this can and will lead to intuitive innovations being passed over because the manufacturer would have to pay patent fees to the "inventor". Total BS.
If the "inventor" didn't have the balls to put his/her money where their mouth is, then they don't deserve shit! Let's not punish companies willing to put forth money into a great idea by making them pay some lazy ass moron who payed a small patent fee to patent his idea of clapping to turn on a light.
From the article:
"The salaries of entry level patent examiners presently range from $32,819 to $70,959. Overtime is strongly encouraged after several months of experience is accrued, and it is not unusual for a junior examiner with three or four years experience to make more than $100,000 annually with overtime and bonuses."
Sounds like they're trying pretty hard to entice people to become patent officers, because the pay scale seems abnormally high for a governmental job. Starting salaries up to $70k per year? Geeez.... Or maybe the salary range is required because the job sucks so much.
Then why the hell do companies have to write 'Patents Pending' on their products for decades? You'd think that incompetency of this sort would speed up the process.
"Did a search on Google for some of the keywords in their description, but only a few thousand webpages came up... Patent Granted, next!"
Mr. Smith: I've come to patent a theory. I call it "Smith's Theory of Relativity".
Einstein thumbs through the papers.
Enistein knocks out Mr. Smith, and runs.
"If we let things terrify us, life will not be worth living."
- Seneca
This is exactly why I keep my anti-grav generator under wraps.
When we graduated my wife received a job offer to work as a patent examiner. Despite the fact that she has a degree in computer science I do not believe it would prevent absurd and obvious invention patents from going through the system. As best as I can tell the examiners are limited to prior patent applications and official publications in searching for prior art. Little room seems to be reserved for common sense. Quite thankfully she did not accept the job offer because I didn't want to move to D.C. and the pay wasn't that great for someone with that degree.
I wonder what kind of technology patent examiners use. Probably ancient PCs running Internet Explorer 3.0 and a cached version of AltaVista from 1994.
Every examiner starts with his or her first patent application after receiving just two weeks of training
So, what is the turnover rate at the USPO?
What percentage of the examiners are seasoned examiners of patent applications etc.
What percentage of the examiners have worked there for less than 2 years?
Howdy Doodly Doo!
Anybody want some Toast?
I was talking to a lawyer about patent applications once . . . his comment was that they are always rejected on the first round, so it was best to actually not submit all documentation in the first round so that there would be additional documentation to submit during the second round. His claim was that patent employees are overworked and often underskilled (because they are required to have such a broad breadth of knowledge) and if they don't know a lot about the particular technology, they would reject the application based on some general ground rules . . . if you gave a lot of extra data, etc. in the resubmittal it would often go through. I don't know if this is a common practice and IANAL, but I think its more than a bit sad that a lawyer would have such a cynical view . . . and perhaps even more sad if the system is actually deserving of this cynical view.
Sorry to disagree with the article, but the best way to get an inside look at a patent examener is with a blunt object.
Yeah, I know its a hard job.. but even the worst fuck-up gets it right once in a while.
The solution is simple. Pay the experts what they are worth to do the job. Just like they should with teachers. Its never going to happen.
There is way too much money to be made in having IP control over an uneducated populace.
Norman Cook's Ode to Sl
One thing I've thought about recently as a help to the patent problem is putting a cap on the number of patents granted each year. If only 25,000 patents are awarded per year (for example), that would in theory cut down on the number of crap patents. Atleast they would have to think a little longer about what really deserves a patent, and many people would be discouraged from filing if they knew there was a good chance it wouldn't get approved. Just an idea...
Generally, bash is superior to python in those environments where python is not installed.
Jeez, how many words do I have to use to say a pot with holes in the lid??:}
Harpo Tunnel Syndrome--my wrist feels funny.
"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;"
- US Constitution, Article I, Section 8.
Good luck getting that changed anytime soon.
What?
In the early days of the patent office, they used to require a working model. I think they should return to this requirement. In my experience, ideas themselves, which can be patented, aren't nearly as valuable as the implementation of those ideas. This would probably also cut down on a lot of bogus patent submissions.
"All examining positions require at least a bachelor's degree in computer science, physical science or engineering"
Apparently a bachelor's degree ain't what it used to be...
--Stephen
Did you ever notice that *nix doesn't even cover Linux?
The real problems for other manufacturers are the submarine patents, where the inventor keeps the application alive for 15 or so years, tweaking the application. Since there is an application pending, all other applications for the same thing get denied. An example would be single-chip-microprocessors. Since everyone in the industry tried to patent it and were denied, they assumed that it was not a patentable idea. Big surprise when the submarine came up to sink the industry. When I went to a police academy, we were taught never to assume anything. The saying goes, when you assume, you make an ass out of u and me.
The big problem for the software industry is that there have been enterprising crooks filing patents based on obscure theses and books, hoping that no one notices it was plagiarized. The patent examiners are not stopping duplicate patents now, they want the user fees, and to the devil (err the courts) with the details.
Examiner: Let me see... [thinking] mmmm... `One Click Shopping'... what on earth is that!!!... Oops, shouldn't show that I didn't understand anything about it... passing things this way and that... there's something about saving... should be a cool way to save money... it's a tough one to figure out, and quite lenghy too... they won't write all this for a trivial one... better approve it and get rid of it... [after an hour of empty gazing] Yes, that is a very innovative one!
We already have lots of patents for perpetual motion machines, apparently.
Now that the Patent Office has kindly placed a lot of patents into their database; they should begin the process of writing programs which help them to eliminate patents based upon (if nothing else) their documentation.
;-)
:-)
I would think the rejection of Patents would fall into a few categories: 1)Bad Spelling, 2)Bad Grammar, 3)Bad Idea. Once these are past - THEN - the object could be compared against pre-existing documents.
I know there are probably millions of documents which might match a new document but, like Google, they could be pared down based upon how closely they look like another invention. Google has how many servers and how many pages to look at and decide upon? Yet you get feedback in only a few seconds. Why not base how the PTO works on Google's model? The PTO could even sell priority for first consideration to companies.
Truly though, since every invention has to be written up, and since we already know that millions of pages can (and are) rejected by a good search engine - this is what the PTO really needs.
Which brings up the question: How many patents are filed for items which are not really in that category? Like an idea for a game which might overlap how something is done in the real world. Would that patent be enforceable because it was already given out? Or does the fact that the patent is filed in different categories mean that you can patent something which has already been patented - but in a different category? So a toothbrush used by a robotic servant (and is a part of the robot itself) can be patented separately from a regular, toothbrush which is held in the hand?
If the former - then that cuts down on the number of patents which has to be searched. Otherwise, all of the patents have to be searched. Which is why it becomes so hard to determine the merits of a given patent. A good search engine though - could shave a lot of time off of how long it would take to determine the merits of any given patent.
This does not, however, eliminate the need for people who are smart enough to know that a flashlight is still a flashlight. No matter if it is called a flashlight or an object which emits a beam of light. This too (the phrase "an object which emits light") should become a part of the search engine. Or, in other words, you create a relational database which relates single words to phrases so the search engine can make better judgement calls.
Just a few thoughts.
Someone put a black hole in my pocket and now I'm broke.
The people protect patented IP rights by funding the PTO and courts in which IP cases are tried. Patents ought to be filed with a tax on their owners as a percentage of returns on the IP. So there's no barrier to filing, but cashing in on the protected asset pays for its protection. The owner's IP assets:income ratio would be applied to the IP, and say a 0.01% tax would be applied to the IP's share of the revenue. The taxes would be spent on a small administration of patent process managers. The actual examinations and tests would be outsourced to certified private American engineering firms. Thus the engine of American ingenuity would be harnessed to drive around the seatbelts, airbags and ABS brakes of the entire infrastructure. Until then, the IP holders are getting a free ride.
--
make install -not war
Q.
Insert Signature Here
Some 4 years ago, in my previous company we submited a software patent (the investors insisted we apply for lots of patents) that most likely landed on a first-timer inside the PTO. It was clear that he did not had a clue about computer architecture, when we challenged we got routed directly to his supervisor who was probably pretty pissed of about having to do his subordinate work. As far as I can tell the examiner algorithm is: 1) read patent application, mark words you don't understand. 2) goggle such words 3) if a match on any, reject all claims, type a letter and cite all goggle hits as previous art. 4) upon challenge from inventor, call your super 5) super is overloaded w/ work and will concur with me (examiner) so keep rejecting patent claims. Keep collecting extension fees. 6) the inventor will find a recognized person in the field and make him do an affidavit. Now I get that he has money and really wants his patent, so accept some of the claims at random. 7) the inventor will change the wording on the other claims and resubmit. Collect more fees. 8) keep faking knowledge about the subject and accept 50% of the claims that had been reworded. 9) inventor has gotten by know 80% of the claims accepted. You wont hear from him again. 10) go to next patent
Note to self, submit patents in 24-point font.
Karma: It's all a bunch of tree-huggin' hippy crap!
See 35 USC 154, specifically section (a)(2).
The term of a patent is 20 years from the filing date assuming it is granted. It used to be 17 years from the issue date, but they changed it precisely because of that reason.
"I have a porkchop, you have a porkchop. I have a veal, you have a veal".
The Patent Bar is the hardest fucking test I've ever taken (and I passed 1st time, BTW). I'd rather take 5 state bar exams than take the USPTO exam again.
Example question (paraphrased):
Below are 5 obscure sentences taken from somewhere in this New York Yellow Pages sized book (the MPEP). We have inserted the word "not," or changed the word "or" to "and" in 1 sentence. You have 2 minutes to figure out which sentence we altered.
It is a sadistic test that tests nothing about your knowledge of patent law. It merely tests your ability to find a needle in the haystack known as the M.P.E.P.
That should at least give them some incentive to not pass questionable patents or even opaque ones.
Free Software: Like love, it grows best when given away.
Interestingly enough, Einstein was influenced by patents he saw. You see, at the time everyone was submitting patents for ways to syncronize clocks. This led Einstein to think about the nature of time and light and such, thus leading to special relativity. I'm sure someone can post a link or more information.
By INQUIRER staff : Donnerstag 15 April 2004, 14:03
*A CONFERENCE* chaired by the Federal Trade Commission, the National Academy of Sciences and the Berkeley Center for Law and Technology kicks off in California in a few hours time.
The idea is reform the patent system without stopping innovation, but the industry is along there in the shape of giants Intel, Microsoft, Symantec and others, and we're sure they'll try to pursue their own agenda.
There will also be representatives from the European Patent Office and the US Patent and Trademark Office.
Luckily, they are counterbalanced by legal organisations and academics.
The agenda for the workshop is here http://www.ftc.gov/bcp/workshops/patentsystem/pate ntsystemagenda.pdf, and
there's a very lengthy discussion document about proposed reform on the FTC
site, here http://www.ftc.gov/os/2003/10/innovationrpt.pdf.
The FTC document proposes a number of changes to the current system including a new admin procedure to challenge a patent's validity without having to go to law, allowing courts to find patents invalid on the preponderance of evidence rather than clear and convincing evidence, and the limiting of awards of "treble" damages.
It's pretty clear that for many large companies, patent actions have become a wing of marketing. Kicking off big patent cases can tie a smaller competitor up for years as the painful battle continues through the courts. Let's hope sanity prevails. But don't hold your breath for that.
Day 1: Thursday, April 15, 2004 at the Bancroft Hotel
1. 12:00 - 1:00 pm, Press Conference
- Mark Myers: NAS & Xerox Corporation
- Commissioner Mozelle Thompson: FTC
1:30 - 3:00 pm
Overview of the Patent System and FTC Proposal for Reform
- Susan DeSanti: Senior Policy Analyst, FTC
- Prof. Peter Menell: BCLT & Boalt Hall School of Law
- Prof. Robert Merges: BCLT & Boalt Hall School of Law
Day 2: Friday, April 16, 2004 at the Bancroft Hotel
8:00 - 8:30 am
Opening Remarks
- Dean Designate Chris Edley: Boalt Hall School of Law
- Robert Merges of BCLT and Boalt Hall School of Law
- Mark Myers: NAS & Xerox Corporation
- Commissioner Mozelle Thompson: FTC
8:30 - 9:40 am, Non-obviousness Panel (Reinvigorating the Non-obviousness Standard)
- Rochelle Dreyfuss: New York University
- Rebecca Eisenberg: University of Michigan
- Ron Laurie: Inflexion Point Strategy, LLC
9:45-11:00 am, Opposition and Post-Grant Review Panel
- Robert Blackburn: Chiron Corporation
- Prof. Joe Farrell: Economics, UC Berkeley (CPC)
- Bronwyn Hall: Economics, UC Berkeley
- Dietmar Harhoff: European Patent Office
- Steve Kunin: Patent and Trademark Office
- Prof. Robert Merges: BCLT & Boalt Hall School of Law
- Douglas Norman: Eli Lilly
11:00 - 11:15 am, Break
11:15 am - 12:45 pm, Litigation Panel (Including Presumption of Validity)
- Mark Janis: University of Iowa
- Mark Lemley: BCLT & Boalt Hall School of Law
- Lynn Pasahow: Fenwick & West
- James Pooley: Milbank, Tweed, Hadley & McCloy
- Matthew Powers: Weil Gotshal & Manges
- Arti Rai: Duke University
12:45-2:00 pm., Lunch
2:00 - 3:45 pm, Industry/Institutional Issues Panel
- Carl Shapiro: Haas School of Business, UC Berkeley (co-moderator)
- Commissioner Mozelle Thompson: FTC (co-moderator)
- Robert Baechtold: Fitzpatrick Cella Harper and Scinto & AIPLA
- Robert Barr: CISCO
- Bart Eppenauer: Microsoft
- Sean Johnston: Genentech
- Jay Monahan: eBay
- Ron Myrick: Finnegan, Henderson, Farabow, Garrett & Dunner
- Kulpreet Rana: Google
- Robert Sacoff: Pattishall, McAuliffe & ABA IP Section
- David Simon: Intel Corporation
- Herb Wamsley: Intellectual Property Owners
3:45-4:00 p.m., Concluding Remarks
Commissioner Mozelle Thompson
Open Standards Portal
"why should I share my idea?"
Why not? If you aren't going to implement it, why not share it? Right now, it costs thousands of dollars to put through a patent application. If you think you might get a patent, you have a strong incentive not to tell anyone (hard to prove that they didn't have the idea first if they file first). Thus, the current system provides a strong incentive for people who do not have the resources to get a patent to keep quiet.
How many patents are actually sold that way? How many individuals pay thousands to get a patent, find a manufacturer, get the item made, and sell enough to recoup their investment? Don't forget to include the opportunity cost (you could have been flipping burgers at McDonalds) of all the man hours of finding a manufacturer and distribution channel.
If this was a valid path to riches, then all those invention submission companies would pay to get the patents for you and split the proceeds. Instead, they want you to pay $6000 to them to patent the idea and "try" to find a buyer. Instead of making money off licensing, they plan on making money off gullible "inventors." Even with the ability to cherry pick the good ideas, apparently patents can't even provide a reasonable return with any consistency.
Note how IBM and other big companies use patents. To protect their research? No, to keep other companies from patenting things and holding them up for ransom. They invent things to use them and patent to keep others from keeping them from using their invention. The guy who ran IBM's patent licensing division was considered innovative for coming up with the idea of licensing some of these defensive patents for some extra cash. Even so, IBM still does not make enough from licensing to cover the research costs.
The point of invention is not to license the patent. The point is to improve your own product. Most inventions will occur in engineering on existing products.
Look at the 1-Click Shopping patent. Would Amazon have implemented 1-Click Shopping without the patent? Of course they would have. It enhances their business by improving their customers' experience. Even if they knew for certain that it would lead to all their competitors adopting it, they still would have (i.e. no competitive advantage).
The only area in which patents are a requirement for distribution is pharmaceuticals. The expensive trials required to get FDA certification cannot be funded without patent protection (or equivalent).
My cousin's wife has an idea that she would like see implemented. In the current system, she keeps quiet in the hope that someday she'll be able to patent it. Without patents, she would have no incentive to keep it quiet and would talk it over or send an email to the appropriate manufacturer in the hope of getting a few free samples (it's a disposable product, so individual items are very cheap).
Bizarre, I've never heard this angle. Most references I find are related to claims in Galison's book "Einstein's Clocks, Poincaré's Maps: Empires of Time ". Everyone else seems to regard this as a bit apocryphal. Still interesting, though.
From the linked book review:
The young Einstein was not, of course, employed in the academic world but in the Swiss Patent Office. And Switzerland, as we know, was a centre of invention and innovation in clock technologies. The patent office at Bern was a clearing-house for new timing technologies, and Einstein's job afforded him a veritable grandstand seat from which to become acquainted with new electro-technological advances.
Slashdot quality declines as the number of hot grits posts decreases. - Provolt's Law, Apr-09-2005