US grants patent for anti-gravity device: FAO our expert

Any thoughts?
Should this not be retitled "Device for the Mass Extraction of Urine "

US grants patent for anti-gravity device via TheInq

patft.uspto.gov 6,960,975

 

Well you may laugh. But I watched a program a few weeks ago where a guy made a "kite" like object levitate. No props or any other form of normal propulsion what so ever. Just the supply of electricity at a very high voltage in to some wires that were attached to the top of the flat "kite". So far no one really understands how it works. But work it most certainly does.


GTM

 

I saw that! It was cool wasnt it?

genuine anti-grav.

 

I also saw that (the US patent, that is). One of the arguments voiced against the USA adopting the same filing system as the rest of the world is that quality of US patents will fall. Some of us believe that the quality of US patents has not only hit rock-bottom but has also commenced digging. It frustrates the life out of those us who struggle to get patent cover for worthwhile things in the USA, and who see this rubbish getting through.

It must also be remembered that there is no requirement that an invention actually work as described. The requirement is that the invention show "utility", that is, that it has a practical use, and a simple statement of utility suffices ("this invention is useful for..."). It is not the job of examiners to check whether it actually works - they check it for novelty and inventiveness against what has previously been done. The checking of working or otherwise belongs to the competition who, the Patent Office correctly believes, is much better equipped to do that than it is.

However, to be fair to the USPTO, my personal suspicion (and it is only that) is that this has been allowed through because nobody's ever going to try to enforce it. The USPTO is required to turn a profit. To do this means that the applications have to move through the office fairly briskly, which in turn means that the amount of time that an Examiner can spend on a case is limited (the USPTO is short of Examiners and has an enormous load of new applications). Therefore granting a rubbish patent like this (a) gets it off the Examiner's desk, freeing him or her for more useful things, (b) brings in more income (the applicant has to fork out the final fee for grant), and (c) its granting is going to bother precisely nobody - as it doesn't work, who cares? And even in the unlikely event of someone else getting it to work, the patent is unenforceable, because it lacks enablement, in that it doesn't tell the skilled person how to perform the invention. In any case, if the final fee is paid and the application proceeds to grant, it'll probably vanish when the first renewal fee ($US800) comes up 4 years from grant. And if the patentee is so silly as to pay that, the 8 year fee ($2000+) will surely stop him in his tracks.

There is a British precedent for this. There was once a British patent applicant called Arthur Pedrick, whose name is a legend in the profession. In the 1950s and 60s, Arthur put in an unbroken stream of spoof patent applications on all sorts of amazing subjects (Google on the name and be amazed). Arthur was not only technically astute, but also an ex-UK Patent Examiner, so he knew all the tricks. In the end, the Comptroller of Patents ordered his Examiners to let anything of Pedrick's through without raising any objection.

 

Must agree as their is quite some unease at the way applications for some bits of software has been treated. Thanks Tones for your thoughts on the subject.

 

Although at least software per se is patentable in the USA, be that a good thing or a bad thing. It's not in Europe, courtesy of the EPO's requirement for a "technical effect". If you can combine your software with a figleaf of technical effect (i.e., the software actually does something other that direct a computer to work), the EPO will grant it. However, if you have, e.g., financial software, you have no chance in Europe. The EU tried to put together a Directive on the patenting of computer software, but it failed earlier in the year, and we're back to technical effect.

 

This might work but you never know.

 

All software can ever do is direct a computer to work.

There are loads of European and UK patents that cover algorithms, http://v3.espacenet.com/textdoc?DB=EPODOC&IDX=GB2387008&F=8 for example. I don't see any 'technical effect', this is a description of processing data. There's no difference to, say, some sort of solving or reporting algorithm in financial software.

Paul

 

yes I saw it as well Ion wind sounded believable to me but I'll believe anything. If he could find a way to get rid of the wires it'd be great. Otherwise we'd have to put that electrified mesh that dodgems use everywhere to travel on it.

 

Sorry, Paul, but all European computer-program patents MUST have a "technical effect" of some sort. It is usually something that the computer then directs something to do, e.g., a milling machine. It may be buried in the text somewhere, but it must be there. And the EPO sees a big difference between financial and technical software. It's a simple fact of life that everyone in the profession has to live with.

Here's a good description of the EPO position from a leading US law firm:

http://www.ladas.com/Patents/Computer/Computer.EPOJP.html

and here's the UK position:

http://www.mondaq.com/article.asp?articleid=34792&email_access=on

Note that, in the UK, in the UK, software which shows a technological effect, and has the capability of being applied in industry, will be treated as patentable subject matter.

 

I'm pretty unimpressed with the patent system. The patent I referenced above includes the text,

IOW any old PC. It's software, an algorithm, not a machine. It should have been published in a research paper but instead an ex-government agency has constrained research into speech recognition and related areas. Very annoying.

Of course this type of patent is unenforceable. The only way to tell whether I'm using their algorithm without permission is to disassemble my software, which I think may be illegal... It certainly would be in the US.

Paul

 

Wrong on both counts - a discovery order from a court (awarded on presentation of reasonable grounds) will allow examination of all documentation not the subject of legal privilege, and that would probably include inspection of the software. The USA is especially big on discovery. It would be difficult to enforce, but not impossible - and given the sort of penalties a US court can throw at you (including treble damages), it would be a brave man who would try it on any reasonable scale. (A low-profile individual might get away with it).

I personally think that patents are inappropriate for computer software, which to me is more properly the subject of copyright. However, the patenting of software started in the USA (where the definition of "invention" is much more vague than elsewhere), and the problem has spread. I would like to see a special intellectual property right specifically for computer software (the way we have plant variety rights for stable plant varieties), but the genie is out of the bottle and there's no getting him back in.

 

You couldn't get 'reasonable grounds' without disassembling, or otherwise infringing my copyright in some way.

I don't see a difference between a software algorithm and a machine, both can be equally inventive amd equally general. Copyright doesn't work because I can simply express my invention in another way and avoid copyright infringement. The problem with patents is that they are awarded for tiny little bits of invention that anybody who put their minds to it could have produced, and those bits of invention can be rather vague ideas rather actual working inventions. The bar is too low. They end up being ammunition for commercial disputes between multinational corporations.

Pharmaceuticals may be a rather different matter, I guess you cannot patent a drug without knowing what it's might be good for?

Paul

 

As far as I know you can't patent a drug if it has natural ingredients in it which is why, even if the natural stuff works better, pharmaceutical companies use man made alternatives that often have more side effects.

 

It is probably possible, unless you succeed in maintaining a wall of complete silence. I'm always amazed how word gets around in the chemical industry. I don't work in the computer industry, so perhaps the folk there are better at keeping their mouths shut, but somehow I doubt it.

This is the nub of the argument - it turns entirely on what is meant by "invention". Different countries have different definitions, and those definitions are interpreted differently by different national courts. Even so-called European patents are interpreted differently. In the most notorious case, Epilady, the UK High Court and the German Bundespatentgericht came to exactly the opposite conclusions on the same set of facts. The courts have a different opinion from you, and unfortunately what they say goes.

If it's an invention, it can't be the subject of copyright! And you'd have to make it noticeably different - if you took a substantial portion of a copyrighted work, you still infringe. This would apply to a computer program - if you took a substantial chunk of a program, you infringe - and that includes the case where you never produce the program in written form.

This can be the case, but not always. There are certainly trivial "inventions" (as we've seen, a notorious feature of US practice), but the Europeans are harder. And in the chemical/mechanical field where I work, the bar is rather higher than I'd like!

Pharmaceuticals are like other patents, in that the invention must have utility, so you've got to say what the thing is useful for. In the case of pharmaceuticals, there is an even bigger hurdle - the national licensing authorities, particularly the US FDA. The documentation (details of clinical trials, analyses) to get a pharmaceutical approved for a particular indication (treatment of a particular disease) literally fills rooms - it doesn't take forever, it merely seems that way. On average, a new pharmaceutical will get on the market 12 years after the patent application is filed, by which time most of the patent life is gone. This is why pharmaceutical patents are uniquely extendible. Oh, and if you find your drug has another indication, you have to go through the whole thing again for that indication.

 

Oh yes, you can. If you isolate a substance from nature and place it at the disposal of the public, you can patent that substance, even though it existed forever. This is a feature of European-type patent law. In the USA, to the best of my knowledge, you can't do this as this would amount to "mere discovery".

If it's exactly the same substance (same molecular formula, same stereochemistry), it will always work exactly the same, be it natural or synthesised. Any difference in effects will come from other natural materials that are there with the desired natural substance.

 

So, legally, there are many ideas that cannot be protected by copyright and cannot be patented, but are, nonetheless, useful and have a commercial value.

Here's an invention for determining the greatest common divisor of two numbers, there's no danger of copyright or patent infringement because the prior art is at least 2000 years old.

The gcd of two numbers can be found by repeatedly subtracting the lesser from the greater until the lesser becomes 0, at which point the greater represents the gcd of the original numbers.

This is the equivalent of the patent I pointed to up thread. Copyright cannot protect the idea, patents shouldn't be applicable, all that's left is trade secrecy.

Paul

 

Certainly.

First up, it would fail at least the European test for being an "invention" (Art.52(2)(a) EPC).

The idea, no, a physical representation of it (such as a computer program), yes. And copyright would subsist in that program, even if the idea were 2000 years old. Remember, prior art is a patent concept that counts for nothing in copyright - the only relevant question is, is it an original (non-copied) work? If it is, it automatically has copyright under the Berne Convention.

aka the "Coca-Cola method" is always an option in many cases. It has the disadvantage that, if somebody else discovers the same thing later, you can't stop them. Worse, there are circumstances, especially under US law, where they can stop you.

 

So the European patent I pointed to earlier, from Qnetiq, which used to be DERA a part of the UK government, is implicitly invalid?

More patent fun. IBM have EP0649144, which again is basically about software or data processing, and it contains the following,

There are many more soporifics where that comes from, I wonder what they were trying to hide?

Paul

 

Without reading the whole thing I couldn't comment, but that wasn't a granted European patent you had, Paul, that was an unexamined published British application. At this point, you see them at their broadest scope, before the patent office starts chopping them back. There is an equivalent EP application, but it's having a tough time (I had a quick peek at the EPO's Epoline on-line Register).

Goodness knows - glad it's not my field (although I know plenty of chancers in the chemical field).