LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   General (https://www.linuxquestions.org/questions/general-10/)
-   -   Can software patents be open source friendly? (https://www.linuxquestions.org/questions/general-10/can-software-patents-be-open-source-friendly-483865/)

General 09-15-2006 03:28 PM

Can software patents be open source friendly?
 
John Doe comes up with a radically new idea; almost as innovative as 1-click shopping. His new software allows computers to read human psionic brainwave patterns, allowing people to interact with machines by channelling tightly focused and concentrated, happy thoughts. This new software has the potential to radically alter the way in which we play FreeCell and Tut's Tomb. John patents the idea and immediately receives hundreds of offers from leading software companies. On the one hand, John is greedy and wants to make a fortune from his idea so that he can retire to some tropical, island paradise. On the other hand, John has been a Linux user for many years and would like to see his idea take shape in the open source realm, where poor, basement-dwelling Linux users all over the world, would benefit from his discovery and live out the remainder of their days as human vegetables.

How can John Doe profit from his idea, but at the same time, keep it in the open so that it can become an integral, non-compromising component of GPL software?

pda_h4x0r 09-15-2006 03:41 PM

John Doe should use a BSD license first, and when he's rich enough he can switch to GPL.

Dark_Helmet 09-15-2006 06:02 PM

Receiving a patent on an idea means the invention's owner gets an exclusionary right regarding the invention. That is, the owner gets to prevent anyone from using the invention. The owner of a patent can then issue licenses to whatever groups he/she wants. There is no requirement that a license be sold. So a favorite group can get a license for free. Similarly, the invention's owner may simply not pursue an infringement claim against another person using the invention. It's up to the invention owner to find infringement and to sue alleged infringers. So wanting to give open source people preferential treatment isn't a problem.

The question of making money doesn't hinge on patent issues but on economics. The question is really asking, "How can I make money on something when I've given it to some of the public for free?" Look to Red Hat and other distribution companies. They make their money through support and service. A crafty patent owner might issue patent licenses to corporate entities with a clause that says:
  1. [Corporation] may use the patent for free during the length of this license for purposes of software development so long as the software development is provided under an Open Source license (including, but not limited to, GPL, BSD, etc.) to all who ask.
  2. [Invention Owner] is entitled to receive X% of fees charged by [Corporation] for activities including, but not limited to, support, maintenance, and service for products using/incorporating the patented invention.
That seems like it would work in theory, but I'm not a practicing patent attorney.

General 09-15-2006 06:56 PM

Regular patents are written in English for the most part, but John Doe is not smart enough to be able to understand or write using the legalese of software patents, perhaps because English has had only about 30 years to develop the poetry necessary to describe the virtual world. John Doe is going to have to pay some lawyer to write up his patent, but he is currently a basement-dwelling Linux user with little money. John Doe imagines that his idea could potentially spread from computers to PDAs, cellphones, bluetooth devices, bank machines, etc., but John doesn't even know if the public would even want to use his brain reading software, even for something as rudimentary as a game of Solitaire.

Is there some way that John Doe can release his FreeCell prototype, coded in Python, into the view of computer users to receive public feedback on its feasibility, before investing his time and money into buying a patent, without losing his right to later patent the idea?

NOTE: John Doe believes that the mechanism behind brainwave reading is rather simple and that there are a variety of ways in which these could be implemented. As such, even if John Doe posted only a video on YouTube of his software in action, others would be able to quickly duplicate its construction or find other methods of receiving and translating brainwave patterns for use in their own software.

Dark_Helmet 09-15-2006 10:40 PM

Quote:

Originally Posted by General
Regular patents are written in English for the most part

Don't be so sure. A patent's "claims" section defines the scope of the patent. The wording of each claim is tediously precise. Specific phrases have special legal meaning. The rest of the patent application (the description of the invention for instance) is more or less in "plain English" but is arguably (from the property rights of the invention owner) less important. Courts look solely at the claims section when hearing a patent infringement case.

It's not likely a regular joe will be able to understand the implications of the language used by the claims or the USPTO's reasons for claim rejection (which will happen at least once). A patent attorney is worth the time and money. Using an attorney is not required but as a practical matter, it's extremely difficult to handle without one.

In the US, an inventor is granted a little wiggle room regarding when a patent application should be filed. First, there's a built-in "grace period" of one year. That is to say, if an inventor starts distributing the invention to the public in some form or another, the inventor must file the application for a patent on the invention within a year or lose the ability to patent the invention. I can discuss the policy behind that if you're curious, but first and foremost, realize there is a time limit.

However, US case law recognizes an "experimentation" exception to the one-year grace period. If the invention is genuinely offered to the public to get feedback/performance data about the invention, then that doesn't start the one-year clock. Understand though, a court won't likely be fooled by an inventor saying "Oh, well, I put 'testing' in the source code--it was experimental use." There must be an established behavior indicating that purpose. For example, limiting distribution of the invention to a small testing group, evidence showing that the group provided some real information, evidence showing that information was used to improve the invention, etc.

What I'm referring to is very generalized. A patent attorney can explain the application filing requirements in detail.

As a side note, the drawback to the one-year grace period is that once the invention is in the public stream, anyone can start copying the invention. Once the inventor gets a patent, he/she can send a "cease and desist" letter to anyone that continues using the invention. However, nobody will have any liability for making/using an invention prior to the grant of a patent. The inventor cannot sue anyone for copies made prior to the grant of the patent. Most inventors will put "patent pending" on physical items to discourage copying during the one-year period. And of course, putting it into the public stream makes it possible for someone else to claim they had the idea first. If that happens, documentation of the invention's creation date, work done, etc. to show which person was actually first to invent will help a court resolve such a dispute. Filing a provisional application with the USPTO is one way to establish a paper trail for that purpose. Again, a patent attorney can give details.

pda_h4x0r 09-16-2006 12:00 AM

John Doe could also design a "shared source" licence that will allow non-profit groups (i.e. LUGs, non-profit distros like Debian, etc) rights to see, compile, and modify the source, but not distribute or share the code without John's permission. The licence could be phrased such that any derivative works of John's project would be subject to the same licence. The license may require the groups to sell the software for a fee and give a set percentage of the fee back to John (this would help both parties). For-profit groups would have to purchase a licence to accomplish the above (again, it would help both parties, and for-profit groups have the capital to market John's software). Once John gets rich enough, he could set a date for the software to become GPL-ed (in fact, the original licence should include a clause that would allow John to change the licence's terms at any time).

This is kind of what I was getting at in my original post (I was just in a hurry and did not think my wording through).

A cheap way for John to demonstrate his copyright without immediately filing for a patent is to put all the source code, documentation, experimentation, etc. into a box and mail it to himself. The official postmark date on the box could be used in court to demonstrate that John had the idea first in the event of a copyright infringement. That way, when John gets rich enough, he can hire a patent lawyer.

Dark_Helmet 09-16-2006 12:54 AM

Implementing a licensing scheme is one possible way to do it, but I don't think it's the best fit. Here are my thoughts:

Copyrights and patents serve two different purposes. Copyright protects a specific work--whether it's a painting, music, or a piece of software. That specific work cannot be copied without the copyright owner's permission. A patent protects an idea. The copyright protects one specific implementation whereas the patent protects all implementations of an idea.

For example, I could have a copyright for a program that sorts data. If someone writes a program that produces the exact same result, but uses different internal data structures, my copyright has not been violated. It's only if someone uses my original work as-is or with minor modifications.

On the other hand, if I patent the idea behind this sorting algorithm, then any program that uses any implementation of my patented sorting algorithm could infringe my patent. It wouldn't matter what internal data structures were used, the language, or anything. If I can show the third party uses the concept I patented in any way, shape, or form, they are infringing.

An inventor interested in protecting the invention wants the greatest amount of power possible because power equates with control. And control is what will ultimately lead to money-making arrangements.

Quote:

Originally Posted by pda_h4x0r
That way, when John gets rich enough, he can hire a patent lawyer.

But again, the inventor only has one year from releasing the invention to the public before the inventor loses all rights to patent the idea.

pda_h4x0r 09-16-2006 07:59 PM

Quote:

Originally Posted by Dark_Helmet
Copyrights and patents serve two different purposes. Copyright protects a specific work--whether it's a painting, music, or a piece of software. That specific work cannot be copied without the copyright owner's permission. A patent protects an idea. The copyright protects one specific implementation whereas the patent protects all implementations of an idea...if I patent the idea behind this sorting algorithm, then any program that uses any implementation of my patented sorting algorithm could infringe my patent. It wouldn't matter what internal data structures were used, the language, or anything. If I can show the third party uses the concept I patented in any way, shape, or form, they are infringing.


Right, but if John Doe chooses to create a licensing scheme, he could phrase it so that any works derived from or inspired from his work would be subject to his same license (similar concept as the GPL). This way, his partners couldn't stab him in the back. Granted, this doesn't stop a 3rd party from replicating his work from scratch, but it may buy John a window of time to get enough money to file for a patent before the 3rd party can beat him to it.

John may also just wish to take out a loan to pay for a patent lawyer, and repay the loan when he gets rich through royaltees.

Dark_Helmet 09-16-2006 10:17 PM

Ok, I see what you're getting at. I guess we're both looking at that one year window differently. I'm hesitant to say that an inventor would be able to secure enough money in one year starting from the "unknown project" stage. And I think you're looking at it from the perspective that the invention would be so intriguing that it would catch fire so to speak.

There might be some complications in trying to restrain, as you mentioned, works "inspired" by the invention. I don't know that copyright could extend that broadly.

Regardless, I have a feeling "John Doe" is the same as the proverbial "friend" everyone has when they visit a doctor.

So, my suggestion to General would be: talk to an attorney about your situation. Patents and copyrights both come from the same area of law: intellectual property. An attorney for patents may very well have a working familiarity with copyrights (or know another attorney that does) or vice versa. Call the attorney and ask how much a consultation would be. They aren't going to charge you for the phone call, but they will charge for the consultation of course. I can't imagine you spending more than an hour for a basic consultation. You'll get reliable insight on:
  1. whether your invention is eligible for patent or copyright in its current state
  2. what options you have available (patent/copyright and pro-con of each)
  3. some of the law's basic requirements to protect your rights to file for a patent or establish a copyright.
If this idea is as valuable as you think it could be, I imagine it's worth spending a relatively small amount of money in the near-term to protect the long-term potential.

And telling the attorney(s) about the invention in the consultation is safe. The attorney-client privilege kicks in and the law requires attorneys to act in the best interests of their clients or face malpractice suits (which could lead to disbarment).

cyb0rg777 09-17-2006 08:58 AM

I was thinking about the same thing a few months back .I just said heck with it and posted my idea in my blog .I don't have the time or the money to work on it.Maybe one of you will pick it up .

edit :www.electro-tech-online.com seems to be down .What my idea was was an eye position sensor using infrared reflective object sensors to detect the pupil instead of the usual methods like using a camera.


All times are GMT -5. The time now is 02:10 AM.