Shared publicly  - 
Personally refuse to file patents. That's the only way to ensure they aren't weaponized.
Casey Barker's profile photoMathieu Lonjaret's profile photoLinas Vepstas's profile photoTyler Tallman's profile photo
Personally refuse to file patents. That's the only way to ensure your brain doesn't melt while trying to explain the invention to the patent attorney.
+Brett Slatkin Let's assume I invent something, and I want to make sure that this invention will always be freely usable by everyone. Since I obviously can't rely on the good will of everyone, If I don't patent it myself to have control over it and make sure it stays "open", how do I make sure nobody else can patent it either? How do I protect it against "malevolence"?
If we're talking about software I suppose the equivalent would be to GPL it, but what if it's an idea or something concrete?
(Disclaimer: I know nothing about such procedures obviously...)
I think that refusing to file on behalf of your employer is inherent in Brett's statement, yes? If you invent something and want to give it away and are unencumbered by contracts with another party, what's to stop you and what is the downside? Sure, one could argue that participating in a broken system is to some extent approving of it, but it beats getting you life sued out from underneath you.
+Mathieu Lonjaret: At one of my previous employers, an "invention disclosure" form had one of three outcomes: Apply for a patent, register it as a trade secret, or publish it in an Intellectual Property Disclosure journal. Nobody actually subscribes to these IP journals, but they're designed and operated to count as "published" in the eyes of the law. This guarantees you (and everyone) the right to use the invention by preventing anyone from patenting it.

The nice thing was, we engineers got some kind of bonus for the disclosure regardless of the outcome. (Well, there was a 4th outcome: Rejected. No bonus for those.)
+Mathieu Lonjaret, +Casey Barker, the sad thing is, even when prior art is well documented or even published in a disclosure journal, the USPTO will still often grant a patent for it. they've gotten better at catching those applications recently, but they're still not great.

worse, once a patent is granted, in practice you need a costly, time consuming, unpredictable court case to overturn it. not many people or companies sign up for those. :/
+Ryan Barrett -- Very true. If +Mathieu Lonjaret wants to defend an invention, publication is the best (and cheapest) way to go, but it's not going to stop a troll from wasting his time.

Still, patent wars are akin to real wars: If you put your weapon out there in litigious battle, there's a chance you'll lose it. So if you can point to a publication that could result in a judge wiping out the patent, a smart troll will steer clear. They'd rather keep it in their stockpile and use it for cross-licensing or bulk deals, rather than lose an expensive piece of war material.
Thanks to all for the explanations/clarifications. Just for the record, the invention was completely hypothetical ;-)
Well, as +Casey Barker points out, many employers offer explicit awards for filing an "invention disclosure", often from hundreds up to a thousand dollars. For many employees, this bonus provides a real incentive to write up even the lamest ideas. I think "patent hamster wheel" is the derogatory term: 'leap, run, leap, ooh I got an idea, stop, write, run leap run...'

As +Ryan Barrett points out, prior art is not investigated. I was particularly pissed off when a "master inventor" (>25 patents) described a particularly juicy one he got recently, double-bonus, as it was deemed a core technology: I had implemented the same thing in an open-source project some years earlier, in GPL'ed code. He's a nice guy, but when we talked about it, he brushed it off -- what's the point, call the patent attorneys, the management, and say "oh never mind"?

Patents are worded so that they have a long list of claims, sorted from very broad claims to very narrow ones. So even if my prior open-source work invalidates some of the broad claims, well, maybe some of the narrowest claims still stand. Pointless to investigate prior-art, until you absolutely need to, in court. Its really really hard work to investigate prior art. much harder, by orders of magnitude, than coming up with the original idea.
Add a comment...