Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Patents The Almighty Buck

Securing a New Idea for the Public Domain? 20

piotru asks: "I live in Japan and am implementing a new idea in software. I was told that it deserves patenting, but I intend to make sure that it remains in public domain (preempting anyone else from patenting it). What can I do to make sure that no one else patents my idea? Here are some choices I have come up with: 1) Patent and apply free license; 2) Apply for the patent, then drop it (will be published with time stamp, and more than likely will be costly); or 3) Copyright the implementation (will have a time stamp, but will it remain compatible with the GPL?) Any ideas?" Some of these issues were touched on in an older story, but the basic premise remains a difficult issue. How would you go about releasing an idea to the public domain while making certain that close software couldn't co-opt it out from under you?
This discussion has been archived. No new comments can be posted.

Securing a New Idea for the Public Domain?

Comments Filter:
  • by Tom7 ( 102298 ) on Monday August 02, 2004 @12:49PM (#9863382) Homepage Journal
    In order to use the GPL at all, you must copyright your software. So just copyrighting it will of course not be incompatible with the GPL!

    Just time-stamping it is supposed to work, although it seems that they'll grant patents for anything, regardless of prior art, these days. It might make sense to try to publish any obvious minor variations on the idea, too.
    • Very true. But you don't have to do anything to copyright a creation. Copyright is automatically extended to an original work.

      Proof and enforcement aren't automatic (and if you don't have proof or don't enforce it, the copyright is essentially void), so you do have some work to do (at the least, get a copy digitally signed or get a hardcopy notarized), but that is reasonably simple and affordable.

      It wouldn't hurt to have it registered, but unless I'm remembering incorrectly it is not necessary.

      I definite
  • by TheRealFoxFire ( 523782 ) on Monday August 02, 2004 @12:52PM (#9863398)
    All you need to do to make sure the idea isn't ever patented and used against your wishes is to make your idea strong prior art.

    To do this, describe your idea in detail (as if you were going to patent it), but instead of spending the large sum of money to patent it, spend a smaller amount getting it officially notarized by a notary public.

    In addition, you may want to timestamp a digital copy (Verisign will do this for a small fee), and make it available online.
    • by jmac880n ( 659699 ) on Monday August 02, 2004 @12:59PM (#9863447)

      A better way would be to get the idea published in a trade journal (an ACM or IEEE journal sounds pretty good). In addition to firmly establishing it as prior art, it will also make the idea available to others.

      If the idea is really significant, then these journals "should" be willing to work with you to get it published. That's what they're there for, after all.

  • by gl4ss ( 559668 ) on Monday August 02, 2004 @12:56PM (#9863430) Homepage Journal
    ..somewhere where you can prove the date later on, preferably with a working model/implementation. after that you'll be having hard time getting a legit patent for it even if you tried it.

    oh and why you think copyrighting it is gpl incompatible? what do you think gpl depends on if not copyright law? all gpl'd stuff is copyrighted..

    besides, if you just write it it will be copyrighted by you anyways.
  • by cybermancer ( 99420 ) on Monday August 02, 2004 @12:58PM (#9863441) Homepage
    Go through all the effort to make a patent disclosure, then make sure there is a public record of it. What you could do is also mail a copy of the disclosure to yourself via certified mail.

    At least in the US once a public disclosure is made, if a patent is not filed within one year then the invention is not patentable. As long as you can later prove that you made said public disclosure (this the certified mailing) then it will be un-patentable by others.

    If someone else tries to patent it after your disclosure and before the year then that is where the complete disclosure you made comes in. If you can prove that your disclosure is for the same invention as theirs then you successfully provided prior art.

    Shouldn't cost you much.

    The main draw back is that someone else could patent an improvement on your invention, and there would be nothing you could do to stop them. They would control that improvement, and if the improvement is a logical progression from your invention then it would effectively block your invention.

    Ideally we need an organization like Creative Commons to hold public patents. It would be important that the organization could not have their portfolio purchased. If they had some funds to file for patents, and then the patents would be freely licensable. The catch is the license would have to disallow licensing to anyone who leverages an improving patent on those patents.

    Since an improving patent is useless without a license on the patent on which it improves then an improved invention could not be manufactured. The only reason a business would have to file such a patent would be to block someone else from using the technology, but in doing so they would rob themselves of access to the entire patent portfolio of the holding organization.
    • I had heard that mailing a copy to yourself is not a good way to time/date stamp material.

      Such "sealed, mailed envelopes" usually don't hold up in court as they are attacked by the opposition as possibly tampered with . . . there is no true standard for doing this securely . . . in fact in the US, there is no postal requirement that an envelope be sealed in a way that it cannot be reopened . . . one could mail empty unsealed envelopes to oneself and then fill them with material later.

      In other words, a p

  • All you really have to do, of course, is make sure your software gets really famous. That way, if anyone ever patents it (which they will, of course), the victim can just refer to your program during the trial, after a busy morning of reading Slashdot. There really should be an official office for registering prior art, though... One shouldn't have to turn to Slashdot for a task that could be performed by the government.
  • You can do so in the knowlege that I know how to use Google, and that I understand what a patent, copyright, prior art, and a distribution license is. Before you consider what the best way to distribute your program might be, you probably want to educate yourself in the very basics of the field first. Hacking patent and/or copyright law is going to be difficult if you don't know how it works.
  • I recently heard about something called "dedicating" a patent. Apparently this means owning a patent and then declaring it for public use. (google for "dedicated the patent")

    I'm not sure this is something you'll actually want to do, given the costs of patenting.

    Establishing prior art should be simpler, just send your idea to one of these companies that are required by law to keep records of all their communication (or use a lawyer or whatever the standard way of establishing priority is).

    On the other han
  • by north.coaster ( 136450 ) on Monday August 02, 2004 @05:19PM (#9864876) Homepage

    Most of the responses have missed the point of the question. The mere act of publishing an idea somewhere will not prevent someone else from patenting it; publishing just gives the original inventor an opportunity to challenge a patent. The poster wants to prevent an offending panent from ever being issued.

    The best way to prevent someone else from patenting the idea is to make sure that the invention has been disclosed in a way that any reputable patent attorney will find it during a prior art search. Most patent attorneys will not be searching the archives of the ACM, SlashDot, or Usenet. Instead, they will be searching for existing patents.

    So the best way to disclose an invention is to file a patent application, and request that the patent office immediately publish the patent application. Yes, there are some fees involved, but this is the only wat to make sure that patent attornesy can find your disclosure. You can then abandon the application if you don't pay any additional fees.

    Go to the US Patent and Trademark Office web site [uspto.gov] for more specific information about the process, fees, etc.

  • Post the idea on a website to get donations from people that think the idea is worth protecting for FOSS. If you get enough donations to get the pantent then license it with a FOSS like license and start a FOSP movement. If you do not get enough donations then maybe the idea is not worth protecting or at least the FOSS comunity has had there chance. Donate the extra money to EFF or FSF or some other FOSS group.

  • on as many usenet and wiki sites as possible. The time/date stamp on your message isn't proof by itself= but combined with several million other messages within a single week, it becomes easy to defend against the patent with prior art.
  • patent claims are couched in descending levels
    of abstraction, so that if a more general claim
    proves indefensible, a more specific claim is used
    as a fall-back position. Copyright is like the
    most specific possible patent claim formulation,
    and does nothing to defend the invention as an
    abstraction. We NEED a public-interest patent
    foundation to fund these applications.

Order and simplification are the first steps toward mastery of a subject -- the actual enemy is the unknown. -- Thomas Mann