The embarrassing part of the patent system is not how long it takes but the way patents have been perverted from their original purpose.
The point of a patent was that you would get an exclusive license to your invention for 8 years. In exchange, you would provide detailed instructions on how to duplicate your invention so that after 8 years, other people could build it. It was supposed to spread innovation. It was a trade you made with the government: they would protect you, you would inform them.
Instead we have patent trolls who build nothing and use patents to extort money from those who do, actively hindering development of new products. The patents granted these days do not include instructions on how to build a device, they just say "I do it this way so now you can't". It's a disgrace.
If it's possible for somebody else to copy your invention without reverse-engineering your product, then you shouldn't be granted a patent.
If it's possible for somebody else to copy your invention without reverse-engineering your product, then you shouldn't be granted a patent.
No, you should be awarded the patent. It's the person who reverse-engineered your invention and did it a different way who shouldn't be found as be infringing. They should also be granted a patent for instructions on how to make the same invention a different way, this way both become public knowledge and once the patent runs out (right now that's basically never) someone who reads both may be inspired to again make the same invention, perhaps combining both into a more optimal device.
I see what you mean, but I was referring more to the situation where merely the existence of the product is enough to show how it's done -- e.g. the infamous "1-click" patent, which doesn't describe a technology but instead a method.
My point was that a patent was supposed to be a trade: monopoly for blueprints. If nobody needs your blueprints, then why are we granting you a monopoly? That should be the yardstick, and the bias should be towards denying patents rather than granting them.
> It's the person who reverse-engineered your invention and did it a different way
What is 'a different way' though? Can I create a different implementation of MPEG4? Or is MPEG4 an implementation of 'video encoding' or 'video compression?'
In one instance, I can just find another algorithm for encoding/decoding MPEG4 files/streams. In the other instance, I need to create an entirely new format.
That patents now last 20 years is completely absurd in the modern world. Anything that is 20 years old has either long ago become ubiquitous, or is long forgotten.
Even 8 years is too long these days in many markets (esp. technology) but would be somewhat more acceptable.
Anything that is 20 years old has either long ago become ubiquitous...
That's kinda the point of the modern IP playbook: to get everyone to pay license fees for your technology.
There are plenty of patents that only make money in the final years of their term. For example, RSA was patented in 1983 but didn't come into widespread use until ~1995 and the patent expired in 2000; if the patent lasted 8 years we can assume that they would have made no money at all. (Of course, since the inventors of RSA were academics it's likely that the patent value provided no incentive towards the invention of RSA.) LZW is a similar story.
I heard the patent office wants to modernize their system, but they can't because there's a patent already covering 'a system to electronically manage patent submissions...'
They just have to negotiate a royalty agreement with the patent holder.
It would be a clever denial-of-service attack to start filing patents that would preclude the USPTO from "practicing" its work without negotiating agreements with patent holders.
The US government has -- and has been known to use -- the ability to exempt itself or its contractors from patent liability for specific purposes (typically national security).
I know you're joking but Section 1 of the UK Patent Act 1977 as amended (it's been 5+ years but I'm pretty sure it's in 1(1)) includes the requirement for industrial applicability.
AFAIR I did make an objection under that clause once out of hundreds of applications. No one can issue patents, by law, apart from the Patent Office and so the patent could not be applied by the holder - also there are morality clauses (which again only very very rarely could be applied, again IIRC I made a casefile note on consideration of a morality objection only once but didn't object). Finally the Gov. has the option to use any application in times of national crisis - I think I remember something about radar patents being used in WWII but that the inventors had been compensated.
Lastly if you're sitting on a patent and won't license it and aren't using it yourself the Comptroller can force you to license it under terms they consider fair.
The fact that any modern government agency processes information in such a backwards way is embarrassing, no matter what that agency is doing. If you want patent reform, campaign for it directly, rather than relying on how slow and tedious the process is.
The point is that Obama isn't embarrassed about the patent system because it is broken. He's embarrassed because the patent office's infrastructure is behind the times. There's a distinction.
It's there, on the lines, not in between them. The flaws that he's pointing out are, in fact, inefficiencies for which I am thankful, for the exact reasons dpcan mentioned.
"Believe it or not, in our patent office -- now, this is embarrassing -- this is an institution responsible for protecting and promoting innovation -- our patent office receives more than 80 percent of patent applications electronically, then manually prints them out, scans them, and enters them into an outdated case management system," Obama said a today's Forum on Modernizing Government.
To be entirely fair, it's theoretically possible that (if they didn't have to do these things) they could examine the average patent more carefully on the same budget.
It's unlikely, though. What would actually happen is an entirely different matter.
If the process gets faster, it will directly affect the speed that which big business can get their patents approved.
So, if he wants the patent process to go faster, then he wants big business to be able to get their patents faster (as is true with any patent filer actually) and therefore, a==b==c
Having a natural aversion to the kind of idiocy described in the article does not mean that you're specifically for the dumbest patents that are granted by the system.
Personally, I'm against inefficiency and against stupid patents. One is an administrative issue, the other is a policy issue.
"This is one of the reasons why the average processing time for a patent is roughly three years." [emphasis added]
For an average patent, what percentage of those 3 years is spent on printing, scanning and entering? If it's one day, then that's the maximum time that could be saved. Profile twice, optimize once.
Let's say it takes a mere day to process a patent application the roundabout way (printing, scanning, etc.). However, the system is near or above 100% capacity because total throughput is slow. Even if they have enough staff to match the needed throughput they still have a 3 year backlog. So even though they can process 1 year of patents in 1 year, it still takes 3 years to make it to the head of the line. Now, if you cut down overhead you'll be able to get rid of that 3 year backlog and get to processing current applications in closer to real-time.
When I worked in the UK Patent Office, now called UKIPO, the patent "processing time" was a measure of the actual work time spent on each application by an examiner. In the UK we did both search and examination and followed the application through to grant - which is different to the US and EPO I gather (or was) - the time spent doing this is referred to internally as PP (patent processing time).
When I left I was working mainly on IPC code G06F (computing). This field had the highest time allowance as it had the most complex/time consuming applications to search and examine. Search would take about 1 day and examination approximately the same other case handling would bring this up to 2½-3 days. That is processing time by the examiner is under 3 days usually over about a 3-5 year period (there is also "formalities" ensuring all the sections are there, the right line spacing, the right fonts, payments, etc., done by others). Remember this was the highest allowance of time.
Some applications would take far longer it generally goes by the number of claims made. Some would be a lot shorter - for example earlier in my career I did "displays" and nearly every other search in one heading there was for "automatic number plate altering systems for trucks to show the cabs license number on the trailer" - once you'd done a couple of such searches it could often be a quick review of more recent prior art and then cut and paste the citations adding pertinent comments. I'd say however that most applications had some novelty of some sort that could be gleaned to gain a patent by a good agent but of those not many would represent enough of a product to be worth financing (except for the mega corps like HP/IBM that would, it seems, deal with bundles of patents to broker entry to a given area of tech).
It depends a lot on what field the patent is in, but less than 20% of utility patents take 5+ years to issue. It looks like the average time has been about 3 to 3 1/2 years recently.
The point of a patent was that you would get an exclusive license to your invention for 8 years. In exchange, you would provide detailed instructions on how to duplicate your invention so that after 8 years, other people could build it. It was supposed to spread innovation. It was a trade you made with the government: they would protect you, you would inform them.
Instead we have patent trolls who build nothing and use patents to extort money from those who do, actively hindering development of new products. The patents granted these days do not include instructions on how to build a device, they just say "I do it this way so now you can't". It's a disgrace.
If it's possible for somebody else to copy your invention without reverse-engineering your product, then you shouldn't be granted a patent.