The USPTO Is Asking Developers For Some Pointers On Fixing Software Patents

The world of software patents is a strange, confusing, inefficient place. There's a pretty serious question as to whether or not software patents should really even exist. Now, the USPTO is making moves to at least step in a better direction by calling out to software developers anywhere and everywhere for advice on fixing the whole thing up.

The call for advice will take the form of two round table events in Februrary, one in Silicon Valley, the other in New York City. Seats are limited but pretty much any developer can attend if they RSVP by email and include some credentials. The events will also be webcast, and any developers/spectators who can't/don't want to attend can also just send in written comments.

The USPTO is looking specifically for advice on "how to improve clarity of claim boundaries that define the scope of patent protection for claims that use functional language," but also stresses that "group consensus advice will not be sought." And it stands to reason they wouldn't be too keen on the idea of no software patents at all. Still, it's a good acknowledgement that the systems is a bit borked, and the USPTO seems to be making an effort to ask the right people for pointers. We'll just have to wait and see if anything good comes out of it. [USPTO via Groklaw]

Image: USPTO


Comments

    The world would be a better place without software patents, in all honestly.

    As a non-programmer, I realise my opinion carries less weight than it would if I could program.
    However, to me it makes the most sense to allow software patents, but only on the code itself. not on the results of the code.
    This means you can patent a METHOD for doing something, not actually doing something.

      As a programmer, I like your thinking and I agree, however, it gets a bit trickier when you realize that a lot of code is done the same way, drawing this line is the difficult part.

      I would personally suggest that algorithms and binary data (images/videos/sounds etc..) can be patented, not code.

      I would also suggest that they look at how complex the algorithm is as to whether it is allowed or how long the patent lives for.

      By doing this, I can't simply copy and modify the code and be in the clear as I would be copying the algorithm. It does however, give me the freedom to make my own using my own methods.

      Think about the MP3 codec, they spent a lot of time figuring out how to do the algorithm, however, quite a lot of people can program it, and in different ways with different efficiencies.

      Something like Facebook, Twitter or Instagram? Sure, you can make your own version, however, you can't copy any images, sounds or logos as this is the marketing of the site, which is what has sold the site.

      Finally, thinking about the mobile phone OS, you shouldn't be able to patent "swipe to unlock" or anything stupid like this, as anyone can think of it in two seconds.

      Imagine if someone patented the "double click" or an "Icon" or the "desktop"

      These are why these patents are stupid and stifle innovation.

        I'd also like to point out that a lot of developers can't afford patents, so they work on their code, only to get a large company patent it which stops them releasing or selling it.

        I should have really said that algorithms get patented and binary data gets copyrighted.

    one thing to do is that unless it requires complex computations to do so, methods of doing things on a computer (or on the internet, or in a video game) should only be patentable if one could also patent doing the same thing physically.

    e.g. the patent on minigames that are played while the main game is loading, you can't ban people from doing other things that may include say, doing a move in chess, physically while a game is loading, and games contained within a larger game are also not patentable, so why should a combination of the two be patentable?

Join the discussion!

Trending Stories Right Now