Do people patent ideas. By that I mean, for instance, say I have an idea for a computer program that would be very useful for just about anybody in the world, could I patent simply the idea, or would I have to find a computer programmer to write the program?
You always read in books that so and so had an idea so he made it/did it/started it, but they never actually say how (s)he did it…
Absolutely not.
35 U.S.C. Section 112 provides that you have to supply the public with (1) a written description of the invention claimed; (2) in sufficient detail to allow one of ordinary skill in the art to practice it without “undue experimentation;” and (3) including the “best mode” of practicing the invention known to you at the time of the application for patent.
Abstract ideas aren’t patentable.
So – “I claim a rocket belt for personal flight to the moon” – not patentable (unless you’ve done some serious 'splainin).
Which isn’t to say some crap patents don’t make it out of the system. But, they’re not supposed to.
I’m pretty sure there are loopholes, Huerta.
I remember hearing that the person who owns the most patents in the US has never made any of them. He would “predict” a likely invention–he was (is?) very good at synthesizing ideas, and predict probable developments. Then he would patent this hypothetical invention, and file it. When someone else came along and actually invented and manufactured the thing, he would wait for them to show a profit and then swoop in and sue them for infringement. Evil, but smart.
Is that myth? Where was I likely to have heard this story?
lissener, the only ‘loophole’ is that if an inventor can give an enabling written description showing that he “had possession” of the detailed implementation of the concept, such a “constructive reduction to practice” (“reduction to practice” being one of the requirements for “invention”) may suffice (as long as he satisfies the other requirements of the statute), even without having built a physical embodiment of the invention.
But someone who’s just predicting, “Hey, wouldn’t X-Ray vision, or 3-D television, or personal rocket belts, be valuable?” – that doesn’t cut it. Now, there are instances (probably too many) in which the Patent Examiner lets these through. The dirty secret is that the Patent Office is pretty ill-equipped, and in practice does not try, to verify if the inventor has constructively or actually reduced to practice.
You probably heard that true story about the late Jerome Lemelson or the not-yet-late Ronald Katz, both of whom amassed hundreds of millions of dollars by aggressively asserting dozens of patents on (some would assert) speculative, vaguely-worded, “predictive” “inventions.” Lemelson was particularly notorious for amending his applications during pendency to sound more like some newly-emerged and valuable product than his original (in some cases, decades old) patent disclosed.
So it’s possible, but it’s not supposed to happen (and, both Lemelson and Katz are roundly considered lowlifes).
The holder of the most US patents is IBM. And they do a whole lot of making and selling of products based on those patents.
Sounds like your story is an urban legend type.
Did the teller ever give you an actual name for this ‘person who owns the most patents’?
Should have made it clearer: The “possession” or “reduction to practice” can be totally within your head or on paper. It’s naturally much easier to substantiate, and probably more common, when the inventor has built and tested the invention to proof-of-concept stage, but the possibility remains of “mental” reduction to practice – but it still has to include the same level of enabling detaill (which you thought of or wrote down).
Heck, just look at the RIM-NTP debacle. Someone posted the patent paperwork in a link on a previous thread, and the patents were… ummm, let’s just say I was confused as to where the real patents were.
The late great Richard Feynman wound up with patents for nuclear powered submarines and nuclear powered airplanes. He had also suggested a nuclear powered rocket but some other joker had already suggested that first. He actually didn’t know he had the patents for a while. See “Surely you’re joking Mr. Feynman” for details.
I hope no one here is surprised to learn that he didn’t build any of these.
(Oops. Forget to add this.)
IBM is a company not a person. Patents are issued to people. (Who can then assign the rights to companies, e.g., their employers.) Also, the articles about this a couple months back made is clear that searching for patents by person is not all that easy. Not everything has been suitably computerized at the USPTO. So the champion is only tentatively annointed.
Yes, that’s just the problem with a lot of patents in, especially, the electronics process/software/algorithm fields. The $64 question is (ought to be) did the NTP inventor clearly envision, and clearly describe and claim in his application, the details of a system sufficiently analogous to the RIM system that it is just to allow the NTP patents to be deemed to cover the RIM system? (Put aside that NTP’s patents got torn apart in Patent Office re-examination proceedings that were proceeding simultaneously with the trial).
Lemelson (whom I mentioned above) notoriously filed vague, prophetic patent applications on “machine vision,” then exploited the ability to file “continuation applications” (basically, a follow-on application claiming the benefit of the earlier one’s priority date) in which he drastically amended the language of the patent claims, seeking to cover laser barcode scanners. When you realize that he was trying to do this based on (and claiming priority from) some applications dating to the 1950s, whose progeny he had kept stringing along through the 1980s (no joke), when lasers had actually come into being, you see the kind of mischief that can be brought about by cynical gaming of the system by those who were never truly “in possession” of the essential technical attributes or insights embodied in the devices that they accuse of infringement.
Shunpei Yamazaki has the most US patents, at 1432. IBM of course has a lot more.
Although you’re not supposed to patent ideas, it happens all too often:
Universal machine translator of arbitrary languages - Star Trek-like universal translator
Space vehicle propelled by the pressure of inflationary vacuum state - warp drive
On the other hand, I think they didn’t need to worry about it for this one: Method of concealing partial baldness - they patented the “comb-over”
Ok. Let me be more clear. I recently thought of a software application that would be really useful, however, I cannot write it. Could I describe in detail the idea, and patent it?
Then again, should I even be talking about a Patent? I’m just wondering how I could realize this, since I can’t do it myself.
My understanding is that in most countries, the inventor is accreditted, but the owner (often the employer) is right there on the application form. Otherwise, how could you trust your employees to agree to assign down the track?
Arthur C. Clarke is the often quoted as the inventor of the communications satellite, but didn’t think to patent his sci-fi idea. So the next time you think of something neat like bouncing a radio wave off an orbital device, go and talk to a patent attorney first.
Software can be protected by copyright (in the US you have a Copyright Registry, something I’ve only ever seen otherwise in Taiwan and Indonesia), but otherwise you can (depending upon the jurisdiction) get a patent for software, or for the business method behind the software.
I’ve actually built the items in the patents I’ve received, but I’ve seen some patents which I serriously doubted had been beuilt (because they wouldn’t have worked). Nevertheless, even the goofy patents looked as if they could have been built – nothing beyond the scope of the technology up until then would have prevented it.
In his book Perpetual Motion, Ord-Hume claims that the Patent office started demanding the inventors of proposed perpetual motion devices construct working models of what they wanted to patent. It was supposed to whittle down the enormous number of Perpetual Motion patents they had to deal with.
“Build a working model” sounds like a pretty good idea for all proposed patents – it would have eliminated some of the sillier ones I’ve seen. It’s my understanding that the Patent Office did demand this for a time. There a large collection of such patent models at the library at Steven Institute of Technology in New Jersey. I don’t know when this requirement lapsed. Nobody asked me for a model.
The US Patent Office is regularly ridiculed worldwide for being absurdly lenient in what it allows to be patented, so I can’t really comment from its perspective.
In the UK and EU, a patent specification (ie. simple description, drawings and “claims” - the exact features you’re saying are different to everything else before it) can be objected to on grounds of insufficiency, ie. if a person “skilled in the relevant art” cannot literally build or execute the patent in the real world based on your description. One specific objection occurs when the patent is claimed “merely by result to be achieved”, such as a spaceship which goes faster than light: you;d have to tell the skilled person how it does that.
Now, it may be that somewhere you’ve made an engineering mistake or somethig and your patent won’t work: that’s OK, you’ve just effectively been granted a monopoly on useless junk which nonbody else will want to exploit anyway. So you don’t necessarily need to know every detail of how to make it in toto, but you must disclose the essential details of the features you’re saying have never been done before. (There is actually a specific clause in the UK Patents Act which forbids perpetual motion machines for “violating known laws of physics”, but that’s really just a convenient excuse for throwing them out quickly - they could be treated like other cases, I suppose.)
In the US, probably - they grant pretty much anything and let the courts sort out invalidity objections later. In the UK and EU, probably not: the programme must have what’s called a technical feature, ie. it must make computers in general work better somehow, rather than simply allow humans to do something more quickly or easily. I tmight well be that your programme is better covered by copyright.
Copyright will protect the code only, and will not prevent someone from achieving the same (exact) functionality by independently develpping different code that achieves the same pfunction.
A patent for a software algorithm, on the other hand, would exclude all others from practicing the same algorithm (or, substantially similar ones).
Most jurisdictions outside the U.S. (well, Europe and Japan, the most important ones traditionally) are fairly hostile to software patents “as such.” Last year the EU rejected efforts to liberalize this policy (the “antis” are led mainly by people who think the U.S. experience with software patents is insane and don’t want to duplicate it).
To the OP: It depends (never a fun answer).
What do you mean by “I could describe it in detail?” That’s the key (and goes to all the requirements, such as enablement, etc., that I discussed).
Example:
“It would be cool to write a software program that would predict tomorrow’s lottery numbers. First it would calculate which number would win, then it would print it out in purple ink, then I would win.” Nope.
“It would be great if word processors could improve your writing. First you would write a paragraph, then the software would figure out how to make it sound better, then it would change it, then you would go on to the next paragraph.” Nope.
However: “I cannot write a single line of code – that’s not what I do. However, I know that the person of ordinary skill in the software world is readily capable of providing code that will perform timer-clock functions [true]. And I know that computers can perform basic arithmetical computations [true]. And I know that random-number generators are well-known and easy to provide in computers [true]. And I know that computers can store image files [true]. And I know that e-mail servers can be programmed to auto-send messages [true]. So . . . I claim a method of sending a random number of images of Carl Weathers to my grandma every X seconds, wherein X is an increasing Fibonacci sequence.”
You haven’t built it, but you’ve mentally conceived it and described it in terms that someone who did understand the art could likely implement without “undue experimentation.” Is it new? I hope so. Is it “non-obvious?” Tougher call, as (if the Patent Office is looking hard enough), there should be a ton of other applications that, while not doing EXACTLY what you claim, perform similar automated-message-dispatch functions. These are independent hurdles to patentability. But, you’ve probably cleared the written description, enablement, and best mode hurdles, all without ever building a model or writing any code yourself.
In many other countries, the owner or employer is listed as “the applicant.”
I am always suspicious of claims that some sci-fi writer “invented” X. A lot of these ‘inventions’ were likely more in the nature of “wouldn’t a rocket belt to the moon be cool?” than of something that was enablingly described (I have no knowledge of whether Clarke could have explained the details of implementing com-sats in sufficient detail – perhaps he could have).
I have read one of his early stories about communication satellites. He is just talking about the concept os using the satellite as a platform for having a radio transmitter broad casting to what was below it. In his story people are worried that the soviet union will destabilize the US by broadcasting porn to the US and corrupting people.
I am always sort of skepticle of a lot of these stories about people pattenting obvious things. People read the abstract and think that is what is pattented. You need to read the spcific claims to see what is being claimed.