3 ms·
NO. It's the fact that the actual inventors of an idea can get some of the value from their idea instead of the one lucky company that finally manages to get an
by nnq 11y ago
NO. It's the fact that the actual inventors of an idea can get some of the value from their idea instead of the one lucky company that finally manages to get an idea to market and make profit from it. Software's is full of brilliant ideas that brought close to zero benefit to their inventor. People like Alan Key or Doug Engelbart, and other much younger "idea people" that few even hear of, deserve(d) to be billionaires just as much as Mark Zuckerberg.
I do believe that software patents should exist, but I believe that they should only be granted to individuals (maybe to non-profit foundations), never to companies! And they should work in a way that nobody can use a patent to prevent you from using his/her algorithm, but only to force you to pay a minimum amount (set at contract creation) per N users (if you're a P/SaaS company).
The software industry is creating tons of value, but I'd argue that less than 1% of it goes to where it should: it just piles up the food chain in the hope of being later redistributed by philanthropy!
And yes, writing such a law would be f HARD, you'd have to really train a new generation of software-expert-patent-clerks and software-expert-lawyers and judges for it to work and don't do stupid things like patent trivialities that would then stifle everyone else.
Oh, and I really believe that patents should not even be about benefiting society. They should benefit the "idea creators", shifting some of the immense wealth from the "idea distributors" and "idea marketers". As long as there are still problems unsolved in the universe, ideas are always in short supply! A problem can only be still unsolved because: (a) we have no idea how to solve it, or (b) we have no idea how to solve it cheap enough. And the whole "we only need to sit and implement properly the ideas we already have" is crap! Idea-space is infinite, so you can always randomly come over a new idea that allows you to do the same thing you're already doing but at 0.1% of the cost and that instantly changes everything.
Let's say Joe Web comes up with a new idea for building CRUD web apps 1000x faster and cheaper. The saved $$$ from this will end up being shifted to other things. And let's say that 0.01% of the saved money gets shifted to something like cancer research. (Or maybe the shift was more subtle: Joe's colleague Moe got fired from making CRUD apps and then went up to learn to program protein molecular dynamics simulation software that later enabled a breakthrough discovery in medicine.). And at some point this research produces a result. Then Joe Web absolutely deserves some 0.000000Y% of the benefit that society got from a cure for cancer. And that minuscule percent could be a lot! And we have no way of distributing this profit back to idea creators right now. Unless we grant software patents to individuals. (And no, the fact that he simply "benefited from the cure for cancer" is not enough. Maybe he was already extra lucky and neither he nor anyone who cares about would ever have been affected by this disease anyway.)
Even if people create ideas even without reward, it doesn't mean they shouldn't get that reward.
In the end society should serve the interests of the individual, not the other way around! This is why we've created society in the first place and why we continue to put up with it!
- icebraining 11y agoYou say that people like Doug Engelbart should benefit more from their inventions, and to achieve that we should grant patents only to individuals or non-profits. But a patent for Doug Engelbart most famous invention - the mouse - was filled, and it was assigned to a non-profit (SRI, a nonprofit research institute), and yet it seems neither he nor the institute got much from it. As for Alan Kay, from what I can tell, he doesn't have a single patent with him as the inventor (which is always a person, an employer would be the assignee). So how could he benefit from your law? I don't disagree that it would be fair for inventors to benefit from their inventions, but I don't frankly see how would your proposal actually achieve it.
- nnq 11y ago- first example: that's because a patent shouldn't simply expire if no royalty is paid for it. (As an aside: the clock should start ticking for the first time someone starts selling a product implementing it. And patent royalties should also be paid with mandatory delay of at least 3 years or something: so, for example, an entrepreneur can start a company that uses a patented technology, even if now he can't afford to license it, and only have to start paying for it after he had some time to start making profit.) - about AK: all his ideas were software related and quite broad. Also, I would guess the he recognizes that allowing "broad software patents" indiscriminately with the current laws would be a total disaster, so he's mostly against software patents, obviously. But in a different legal context he's probably think completely different. I don't know that much about patent laws, but the parts that I do know of seem totally backwards to me. Like nothing makes sense the way it is, you'd have to rewrite everything from scratch and retrain all the lawyers and judges. Probably because the laws were invented for easily-buildable physical inventions like the steam engine a 100 years ago. And they have no concepts that make sense for "informational products". You can't look at the behavior of people in the past and say "considering their behavior, such laws would not have benefited them". The simple fact that a law exists and is written in certain way can completely alter the behaviors of people because they know they now can take advantage of it, so they behave differently. I am personally against software patents NOW, btw. In the current legal context they would be nonsensical an harmful. But change lots of the context around them and they could be a very good idea for the reasons I mentioned in the previous comment. Probably the solution would be to invent a new concept of "soft idea patent" that has no relationship whatsoever with the concept of "classic patent" and for which none of the current patent laws apply. Then write from scratch the laws for these new "entities". And "digital computer software patents" would be just one of the subclasses of "soft idea patents", each with their separate laws and regulations.