How Patent law really work?

[Pawelk198604]

I wonder if i inventor invent something and not patent it, and make his/her discovery available for everyone does something greedy other can patent it.

 

For exemple if someone invent drug for Cancer or HIV and not patent it just release it for general public can other patent it?

 

For exemple i know that Marie Curie-Sklodowka (because she was born in Poland not in France:D ) And her Husband Pierre Curie decided not to patent their method of production of Polon (Named in honor of the Polish, homeland of Marie ;-) And Radium(named for radiation and glow of that element, which Marie and Piere liked to watch in dark :D)

 

So does discovery need to be patented, or just can be released for anyone?        

[kurja]

You can not get a patent for anything that isn't a new, previously unpublished invention. The invention also needs to have an application, you can't patent an idea or a discovery that doesn't do anything.

[KSK]

No - once something is in the public domain, most countries won't let you patent it. Rule of thumb for aspiring inventors - patent first, publicize afterwards. Some countries (the US most notably, although the rules there have got a lot stricter recently), allow a so-called grace period in which inventors can disclose their inventions but still get a patent.

Also, you don't need a patent to do anything. Patents aren't permission for you to use your invention - they're a legal right to stop other people using your invention. That can work in all sorts of ways.

For example, I create an HIV drug. I plan to make that drug available at cost to developing countries. But I've just invested several million dollars in developing my drug and I don't want my competitors to get a free ride off my research.

Another example. I invent a cancer drug and it also turns out that a very similar drug can help cure baldness. I want to make my cancer drug available to the world for free but I don't mind making a fat profit selling my baldness cure.

In both cases, I'll probably want to file a patent for my drug. In the first example, my patent will give me the right to sue my competitors to stop them using my drug in their own research. In the second example, one patent would probably cover both drugs and I don't want my competitors undercutting sales of my baldness cure and stopping me subsidising my cancer drug.

Yes - these examples are a bit contrived but they make a serious point that patents can be (and are) used for what people would normally think of as the public good. They're not necessarily all about patent trolls or companies blocking technological progress for nefarious reasons.

Edited May 19, 2016 by KSK

[PB666]

39 minutes ago, kurja said:

You can not get a patent for anything that isn't a new, previously unpublished invention. The invention also needs to have an application, you can't patent an idea or a discovery that doesn't do anything.

Oh damn, and i was about to patent spamming. 

[sevenperforce]

A patent is a lot like a contract between an inventor and a government. The inventor agrees discloses how his new idea works, and the government agrees to enforce a 20-year exclusive use arrangement for the inventor. This benefits the inventor, because he is now (theoretically) guaranteed two decades of potential profit, and it benefits the government, because after those 20 years have elapsed, the invention becomes available for public use. Everybody wins. 

Another option is to keep your invention a trade secret. If you simply never explain exactly how your invention works, then no one else can reproduce it. Coca-cola did this with their recipe for their eponymous product. Trade secrets, unlike patents, give you the ability to profit from your invention for far longer, as long as no one figures it out. 

You can't patent something that someone else invented and produced, even if you came up with the idea independently, and even if they never patented it. If something is already being produced, it is considered "prior art" and so the government will not consider it to be patentable; they are only interested in patenting ideas that otherwise would remain trade secrets.

In the past, ownership of a patent was based on who invented it first. So if Carl's Crab Cafe and Alice's Awesome Eatery both file a patent for bacon-wrapped seaweed at the same time, but Carl has emails proving he invented bacon-wrapped seaweed first, then Carl would get the patent and Alice would not. But just a few years ago, the USPTO changed the rules. Now, the patent is awarded to whoever files first, regardless of whether they actually invented it first. This encourages people to file sooner rather than later, sharing their invention earlier and resulting in more progress.

If you create an idea and decide not to patent it, but just begin production, your competitor may actually still be able to apply for a patent. However, if they attempt to enforce their patent, you can show that you were already PRODUCING the invention, making their claim prior art and getting their patent thrown out. You can also get their patent thrown out if you can show that the idea was already published long before either of you came along. That's what happened with Blue Origin and SpaceX; BO took out a patent for barge landings of first stage rockets, then sued SpaceX for trying to land their rockets on barges. SpaceX pointed out that the Russians had published plans to land rockets on barges long before, and BO's patent was tossed. 

Finally, if you don't mind other people using your invention but want to keep someone else from patenting it, you can file a patent directly into the public domain for free.

Disclaimer: not a patent attorney, but I did stay at a Holiday Inn Express last night. 

Edited May 19, 2016 by sevenperforce

[softweir]

3 minutes ago, sevenperforce said:

You can't patent something that someone else invented and produced, even if you came up with the idea independently, and even if they never patented it. If something is already being produced, it is considered "prior art" and so the government will not consider it to be patentable;

...

if Carl's Crab Cafe and Alice's Awesome Eatery both file a patent for bacon-wrapped seaweed at the same time, but Carl has emails proving he invented bacon-wrapped seaweed first, then Carl would get the patent and Alice would not.

Of course, the owner of the "prior art" has to be able to prove how long they have been using the patentable idea. You implied that when you mentioned Carl having emails proving the date, but I feel this point needs to be made explicit! No proof, and there is a danger that the patent may be upheld.

[sevenperforce]

11 minutes ago, softweir said:

Of course, the owner of the "prior art" has to be able to prove how long they have been using the patentable idea. You implied that when you mentioned Carl having emails proving the date, but I feel this point needs to be made explicit! No proof, and there is a danger that the patent may be upheld.

It also depends on whether we are talking about the old rule (first to invent) or the new rule (first to file). For example:

1. Carl writes an email to his chef dated January 1 mentioning an idea for bacon-wrapped seaweed, but they don't make any. On May 1, Alice comes up with the idea for bacon-wrapped seaweed and applies for a patent. Under the old rule, Carl could apply for a patent after Alice, citing his January 1 email, and he would get the patent. Under the new rule, Alice would get the patent because she applied first.
2. Carl has an interview with a newspaper on January 1 and mentions that they will be adding bacon-wrapped seaweed to their menu, but they never do. On May 1, Alice applies for a patent on bacon-wrapped seaweed. Carl points out that he already published the idea in a newspaper four months earlier, and so the patent is thrown out and both Alice and Carl are free to sell bacon-wrapped seaweed.
3. On January 1, Carl begins selling bacon-wrapped seaweed to a small number of his customers, but doesn't put it on the menu. On May 1, Alice applies for a patent on bacon-wrapped seaweed and puts it on her menu. Carl sues, alleging that he came up with the idea first. He loses the lawsuit, because he can't stop Alice from making it since he never applied for a patent, but Alice won't get her patent because Carl was already making the stuff.
4. On January 1, Carl files a patent for bacon-wrapped seaweed. On May 1, Alice begins selling bacon-wrapped seaweed. Carl sues for patent infringement. Alice's lawyer finds a trade magazine from 1981 which discusses the possibility of bacon-wrapped seaweed, proving the idea is "prior art", and Carl's patent is thrown out, leaving both parties free to sell it.

One critically important point is the difference between #1 and #3. In case 1, Carl merely came up with the idea. Under the new rules, this won't keep Alice from getting a patent on it if she comes up with the idea independently. In case 3, Carl not only came up with the idea, but developed it and began producing it. In this case, Alice can't get a patent on it. This difference demonstrates the value of actually developing and producing an idea. Anyone can come up with an idea, but you have to actually do the work (either by producing it or by going through the patent process) or it's really not worth much.

It should also be noted that after any of the above cases, Peter from Peter's Pleasure Palace can come up with a method for producing bacon-wrapped seaweed, and that is a new idea which can be patented "on top of" the old idea, regardless of whether the old idea is patented or prior art or anything else.

[wumpus]

WARNING: Just because you patent something, there is no guarentee that somebody doesn't come by, make an improvement, and patent that improvement.  This happens in any system that awards patents on inventions, because this is what inventions really *are* (read/watch some James Burke if you don't believe me.  That's his schtick).  In modern systems built to encourage patent trolling (more by the vested status quo fending off disruptive tech than the official "patent trolls"), expect any interested parties to patent up all possible implementations.  At that point they have an effective patent on it and you have nothing.  An example:

Lempel and Ziv invent dictionary-based compression.  They create LZ-77 and LZ-78 (happened in 1977-78 obviously)

Welsh adds an improvement to create LZW.  Sperry files a patent (patent applied for in 1983, published in 1984)

Spencer Thomas write "compress" that implements LZW (1984), and in the process fixes a few things to make it possible.  Apparently it wasn't public domain (not that such is possible any more), but I've never heard of a license being required.

"Compress" becomes the heart of compression, going on to power .Z,.gif, .pak, .zip (with a huffman backend) and others.  Result: Sperry-Univac owned *all* of it [patent granted in 1985, and has since expired], thanks to the LZW patent.  While I'm not sure how much they managed to grab out of the whole sordid mess, it certainly damaged the LZ systems in favor of other less encumbered systems.

Edited May 19, 2016 by wumpus
added dates

[KSK]

No guarantee at all and that's a part of how the system is supposed to work. Party A invents something, Party B either has to come to an arrangement to use Party A's tech, or invent around it, ideally improving it along the way. Net result - gradual ratcheting up of technology. In practice, what tends to happen is a series of bandwagons. Party A invents something, Parties B, C, D, E and F invent useful improvements to it, Parties G through Y patent ever more bizarre workarounds to A's patent but sticking with similar technology. Party Z comes up with something radically different and the cycle begins again.

It's also not necessarily true that interested parties can patent up all possible implementations leaving you with nothing. If your patent is worth the paper it's written on, you should be in a pretty good position at that point, since some (or possibly al)l of those fancy-dancy improvements are based on your original technology which is protected by your patent. So before said interested parties can actually use their patents, they need to come cap in hand to you for a license to your original patent.

Trite example. Party A files a patent claiming a "container suitable for holding water comprising a handle and a pouring spout." Party B then files a patent for an improvement claiming a "container suitable for holding water, comprising a handle, a pouring spout and an electric heating element." Party B's superior water holding device still falls within the scope of Party A's patent, ergo, Party B cannot sell their improvement without infringing Party A's patent. Party B (ideally) signs a cross licensing deal and both parties go into business selling electric kettles.

In general patents are supposed to describe at least one way of implementing an invention such that a suitably skilled reader can go away and duplicate it. Maybe not duplicate it well but produce a functioning version of the invention. Your story - whilst I'm not doubting it for a moment - instantly raises one question - was the Sperry patent actually valid? If Thomas 'fixed a few things that made it possible', my immediate thought is that the Sperry patent probably wasn't enabling and could have been challenged on that basis.

Personal opinion - US style software patents are out of whack anyway. Not because they're intrinsically a bad thing, but because unless the patent offices is very careful (and historically it hasn't always been terribly careful), a software patent can end up being far too broad and covering the entire problem to be solved, rather than a particular set of solutions to the problem. It's a tricky balance at the best of times - narrow patents can be essentially worthless because they're too easy to work around, whereas broad patents can stifle progress entirely by handing an entire technical field to one party - but for a long time I think software patents fell squarely into the 'too broad' camp. Typically, the USPTO has now swung entirely the other way and getting any sort of software patent - even for a sensible invention - is an absurdly whimsical process.

Edited May 19, 2016 by KSK

[wumpus]

1 hour ago, KSK said:

It's also not necessarily true that interested parties can patent up all possible implementations leaving you with nothing. If your patent is worth the paper it's written on, you should be in a pretty good position at that point, since some (or possibly al)l of those fancy-dancy improvements are based on your original technology which is protected by your patent. So before said interested parties can actually use their patents, they need to come cap in hand to you for a license to your original patent.

The OP was asking about what happens if you don't patent something.  My example showed that after multiple improvements and necessary details were worked out, only one was patented (and not by either original inventor).  The employer who patented the employee's invention grabbed everything.

Pretty much all [US*] IP law requires you to nail down claims for everything you might want to protect (at high legal fees).  This gives you standing to fight in court (at much, much higher cost) against anyone who might jump your claim.  If you don't get that initial claim in, you can't do anything when somebody else claims all effective use.

* this isn't just the US.  There was a recent trademark grab by an anti racist/violence group grabbing the name "Soldiers of Odin" out from some other group using it.

http://yle.fi/uutiset/woman_trademarks_soldiers_of_odin_name_for_unicorn-themed_clothing_in_anti-racism_protest/8886818

[KSK]

Uh-huh - I'm aware of the costs. Managing an IP portfolio and budget is part of what I do for a living. Most of it goes on attorney fees.  And I think my examples showed that there are good reasons to patent something even if you don't intend to make a profit from it. 

The trademark case is interesting. Could go either way I think, if the SOO decided to oppose that trademark registration.

I'm not aiming this at anyone in this thread but IP is a really bad topic for internet discussions in my experience. This one is a whole lot better informed than most that I've seen. The devil is invariably in the details when it comes to IP - it's not an easy topic to shoehorn into a glib analogy, although that doesn't stop everyone from trying, usually with varying degrees of misconception.

Edited May 19, 2016 by KSK

[magnemoe]

5 hours ago, wumpus said:

The OP was asking about what happens if you don't patent something.  My example showed that after multiple improvements and necessary details were worked out, only one was patented (and not by either original inventor).  The employer who patented the employee's invention grabbed everything.

Pretty much all [US*] IP law requires you to nail down claims for everything you might want to protect (at high legal fees).  This gives you standing to fight in court (at much, much higher cost) against anyone who might jump your claim.  If you don't get that initial claim in, you can't do anything when somebody else claims all effective use.

* this isn't just the US.  There was a recent trademark grab by an anti racist/violence group grabbing the name "Soldiers of Odin" out from some other group using it.

http://yle.fi/uutiset/woman_trademarks_soldiers_of_odin_name_for_unicorn-themed_clothing_in_anti-racism_protest/8886818

This is trademarks who is something totally different, trademarks is about brand and brand recognition. 
Its also far more fuzzy than patent law

 

13 hours ago, sevenperforce said:

It also depends on whether we are talking about the old rule (first to invent) or the new rule (first to file). For example:

1. Carl writes an email to his chef dated January 1 mentioning an idea for bacon-wrapped seaweed, but they don't make any. On May 1, Alice comes up with the idea for bacon-wrapped seaweed and applies for a patent. Under the old rule, Carl could apply for a patent after Alice, citing his January 1 email, and he would get the patent. Under the new rule, Alice would get the patent because she applied first.
2. Carl has an interview with a newspaper on January 1 and mentions that they will be adding bacon-wrapped seaweed to their menu, but they never do. On May 1, Alice applies for a patent on bacon-wrapped seaweed. Carl points out that he already published the idea in a newspaper four months earlier, and so the patent is thrown out and both Alice and Carl are free to sell bacon-wrapped seaweed.
3. On January 1, Carl begins selling bacon-wrapped seaweed to a small number of his customers, but doesn't put it on the menu. On May 1, Alice applies for a patent on bacon-wrapped seaweed and puts it on her menu. Carl sues, alleging that he came up with the idea first. He loses the lawsuit, because he can't stop Alice from making it since he never applied for a patent, but Alice won't get her patent because Carl was already making the stuff.
4. On January 1, Carl files a patent for bacon-wrapped seaweed. On May 1, Alice begins selling bacon-wrapped seaweed. Carl sues for patent infringement. Alice's lawyer finds a trade magazine from 1981 which discusses the possibility of bacon-wrapped seaweed, proving the idea is "prior art", and Carl's patent is thrown out, leaving both parties free to sell it.

One critically important point is the difference between #1 and #3. In case 1, Carl merely came up with the idea. Under the new rules, this won't keep Alice from getting a patent on it if she comes up with the idea independently. In case 3, Carl not only came up with the idea, but developed it and began producing it. In this case, Alice can't get a patent on it. This difference demonstrates the value of actually developing and producing an idea. Anyone can come up with an idea, but you have to actually do the work (either by producing it or by going through the patent process) or it's really not worth much.

It should also be noted that after any of the above cases, Peter from Peter's Pleasure Palace can come up with a method for producing bacon-wrapped seaweed, and that is a new idea which can be patented "on top of" the old idea, regardless of whether the old idea is patented or prior art or anything else.

Thought first to file was applied to multiple 19th century patents like the radio.

Production methods is another case. 
I also support that coming up with an idea is not enough, you need an prototype
One issue of many idiot patents around, mostly software patents on web page interfaces is that it has been previously use.