Hi,

I have been reading up on the difference between a patent and copyright (as it applies to computer programs). I don't think I'm fully grasping the idea. Can someone give me a real life example? Is this correct...

If I write a program to add two numbers, I have a copyright on the actual code instantly. If I get a patent, I have writes to the process of adding two numbers.

This doesn't make sense to me. How can someone own a patent on a computer idea like that? Wouldn't, in essence, everyone be violating this? How does this pertain peticuallry to Linux.

Thanks.

Ryan

You've got the right idea. Of course in your example you couldn't really obtain a patent on the process, since there are about 4000 years worth of prior art. But let's say you came up with an algorithm that tells a computer how to identify a chair or a table or a sofa. This is a very complicated problem, and a solution would be very powerful. I this case a patent would mean only you and people who obtain permission from (in the form of a license) would be allowed to create software that uses this algorithm. Once the patent expired (after 17 years, I think), then any old joe could use your algorithm in their software.

The idea behind patents is so that the person who developped a technology can get a good return on their research investment. If you spend six years of your life working full time on something like this, that costs a lot of money, so you want to be sure it's worth if for you. It would suck if some big company took your product and made their own version, then used their own brand name and marketing clout to push you out of the market of a product you invented in the first place, wouldn't it?

Sorry about posting this on the wrong board. I wasn't sure exactly where it should be.

If patents are a good thing, then why are so many Linux professionals against them?

Ryan

They are not a good idea because in the above example, for instance, you would have a copyright of the code you wrote to implement the algorithm, which is good, no-one should be able to copy it if you don't want them to.

However, if you were allowed to patent your idea no-one else would be allowed to write *any* algorithm to distinguish a chair from a table, thus stopping innovation in that field and giving you a monopoly. (You should make money because you have the best algorithm, not because you have the only one)

Google around for software patents and try to imagine where we would be today if they had all been granted.

With software people or companies are not developing technologies, they are using an already establshed technology in different ways to perform diffrerent tasks.

You may as well give an author of a novel a patent on a love scene or whatever, thus ensuring that only they could write about love scenes.

As far as software is concerned, copyright good, patent bad.

Nicely stated ahh

what determines whether or not someone copied my code? i mean, say the two programs are both written in the same language, there are bound to be similarities even if they are each written in isolation. i'm not a programmer, so that's part of my question too: how easy is it to tell if something is copied, or unique? who gets to decide, and what are the guidelines? and what sort of enforcement is there? penalties?

I'm afraid I have to take issue with this. The demonizing of software patents is getting a bit extreme. It's simply not true that "no-one else would be allowed to write *any* algorithm to distinguish a chair from a table". A patent would protect a process/implementation of an idea. If I were to patent an algorithm that identified a chair from a table, then nobody could use that algorithm without agreeing to a patent license/whatever. That does not stop someone from developing an entirely different algorithm to identify a chair from a table. In the patent, I would have to describe in some detail this patentable algorithm. Maybe my algorithm uses video input from an array of video cameras to accomplish the task. If someone comes along and develops an algorithm based on scanning the object in question with a grid of lasers, they definitely aren't violating my patent.

The key focus the anti-patent groups are yelling about are popularized, patented algorithms. For instance, the GIF algorithm was patented, and everyone raised bloody hell about it because the GIF format became popular. There was no lack of other, competing image compression algorithms (jpeg). Nobody would have cared that the GIF algorithm was patented if nobody used it. But since it became popular, people felt they needed to support it. That's the whole point of a patent on anything. The good patents are desireable by the masses, which help compensate (and encourage) the innovators. Is it possible small-time software development houses can't afford a patent license? Yes. Is it possible that a small manufacturing company can't afford a patent license to improve their manufacturing process? Yes. Why is the software development firm's situation any different from the manufacturing company? The popular response to the manufacturing company would be "Tough cookies, make due without it, or innovate on your own." I don't understand why the same shouldn't apply to the software firm.

Are all software patents good? No. Are all software patents bad? No. Some patents seem downright idiotic (like Amazon's single-click ordering). I don't know all the details of all these questionable patents, but you have to trust that unreasonable patents will be nullified when challenged. Regardless of popular opinion, the courts are not stupid. They rely on input of professionals to help guide their decisions.

isn't that quite expensive?

I'm sure it can be, but you also have to understand that other companies are not going to look favorably at another company's questionable patents. It's not always going to the the David and Goliath situation that many anti-patent groups want to depict (the poor, single developer versus the huge media conglomerate). Suppose Microsoft wants to get into the online retail business. Then, all of a sudden, Microsoft has a vested interest in challenging Amazon's one-click patent. Yes, Microsoft might simply pay the patent license to start, but eventually the accounting department is going to ask: "Why are we paying them for a frivolous patent license? That money could go directly to our bottom line." The patent gets challenged (by Microsoft itself, or a group of online retailers), and then a court decides if:

A) The patent is unreasonable and should be stripped
B) Clearly defines what situations the "one-click ordering" patent covers (and hence what it doesn't cover)
C) The patent is legitimate, and you have to "pay to play' - in other words, it is a reasonable patent

For A or B, the door opens to implement what the patent described either exactly as it was described, or through different, unprotected methods. This is a marketplace form of checks-and-balances. Never underestimate the motivational force of greed. It may not happen as fast as small developers want it to happen, but that's just life.

However, let's look at the David and Goliath situation. This poor, single developer can force the issue regarding a patent because (as the anti-patent groups like to imply) the developer is so poor, (s)he is "judgement proof". That's a term my brother uses to describe a situation where even if a fine is handed down against the developer, there is no way they can pay it; it's a worthless judgement beyond making a legal point. Why is this a good thing? Well, the developer has "nothing to lose". The case draws attention to the patent, and companies begin examining what they are doing that might be found in violation. If there is a company that finds themselves in such a situation, they may very well "come to the developer's aid" in order to strike down the patent before the patent-holder decides to go after the company.

I honestly don't know if there is any possibility of jail time for patent infringement. If there is, that changes the situation; it's no longer "nothing to lose". However, to my knowledge, I've only heard of monetary judgements against people that infringe patents.

This would not be a software patent, as video cameras are not software. The software patent would be something like:-

Creating a binary representation of three dimensional objects for the purposes of categorizing, comparing and differentiating between the objects.

This would then cover videos, lasers or someone typing in a description.

From http://petition.eurolinux.org
The above are just a few examples of software patents.

Imagine a world where a very, very rich software company, already convicted of illegal business practices and yet strangely unpunished, bought out all these patents. Scary!

As for a patent lapsing after 17 years, what use is that in the computing world? Do you want 1988 technology now? This is ludicrous.
This just shows how unsafe patents are.

Are we to rely on companies with questionable business practices to police them? Who would be rich enough to challenge Microsoft's patents (and note here it depends on how wealthy you are, not how right you are!). If the patent is unsound it shouldn't be granted, but it seems even Dark_Helmet doesn't have confidence that this would be the case.

What would happen is a few rich companies would end up owning the patents, they would scratch each others backs (i.e. you use my patent and I'll use yours, as already happens) so would incur no license fees. The rest of us would have to pay (probably extortionate) fees. This would stifle innovation (and probably be the end of OSS).


Hmm, most people don't consider bankruptcy "nothing to lose"

have you even herd the case of SCO? (as a popular example .... others ain't hard to find)

SCO: IBM stole millions of lines of code from us, and stashed it in Linux
IMB: ummmmm, no we didn't
SCO: those lier's stole our code!
JUDGE: SCO, were is the proof?
SCO: we refuse to say, as doing so would alert IBM to what codes we own, and thus allow them to take it out of Linux
JUDGE: you have 3 months to disclose the lines of code IBM contributed to Linux that you own
..... 3 months later ..............
JUDGE: were is are the "stolen" codes?
SCO: we refuse to disclose the code
JUDGE: you have 5 months to show the codes
........ and the same pattern goes on (BTW is that case still open?)

i think that is good proof the courts are not the smartest people in the world (probably can make chimps look like mathematicians) (at least in the US)

and if you don't buy the "courts are stupid" argument .... then how about they are greedy wh**** (and the proof is in their symbol) ...........

- a half-naked girl weighing gold (i do believe you can inf err what the "*"'ed out word was from here)
- the girl is blindfolded (yes, either she likes that kind of stuff, or shes being non-discriminative (most gold wins!))

honestly ..... having the courts decide is almost the same as giving patents away to the "big" businesses ... even if you won, they would appeal and keep the case in court till the patent either wears off, or becomes irrelevant ..... ... . . . not to mention the money that would be lost (which would probably make you bankrupt anyways) ... . ..... for small to single people, patents would force them to either buy the license (not a option for OSS .... few, if any, projects could afford the prices) .. or force the project to close


as the example ahh put up says the holder of this patent would not only have a patent over the specific process, but also could have a patent over any process to identify 3D objects using binary (used in all modern computers ... its the 1's and 0's)

software patent are patents that patent ideas, and when people apply for them they usually make sure the wording is only as specific as needed (so the patent covers as many ideas as possible)

tho, patents are a good thing, but only actual patents on actual devices ... not patents on general ideas, or patents to force people not to copy ones device (it impedes innovation (buy creating scare, and having companies believe if they try to "innovate" they will run into a patent), and creates a scary environment (companies have been known to sit on patents (purposely allow people to infringe it (like the patent on 3D environments (covers almost any 3D game, and 3D meddlers), then sue them all once time is right, and reap in the money)

You're not understanding the point I'm making. In patenting an algorithm that uses an array of video cameras, you are patenting an algorithm that analyzes a specific dataset. The data given to you from the video cameras is entirely different than the data given by a grid of lasers. The hardware itself is not part of the patent, but rather the means of processing and analyzing the data provided. That is most definitely different, and clearly one algorithm would not violate the other's patent.

I don't agree with this either. That line of argument falls directly in line with "scare tactics" people typically attribute with politicians trying to push specific legislation. This point is also argued from the "hindsight" position. "Imagine a world where this extremely popular technology had been patented! The sky would be falling!" In all truth, we don't know what would have happened if the technology was patented right at the beginning. Patents can serve as a deterrent to use. Perhaps the technology never would have caught on. Perhaps someone would have created a competing standard because they didn't want to pay patent license. Or perhaps, everyone would have agreed that the patent is reasonable, the technology is super cool, and the patent holder should be rewarded.

I have to disagree again. I never implied Microsoft was invulnerable. The truth is quite the opposite. I don't have a specific website to link, but there was a relatively unknown (to me) company that sued Microsoft for patent infringement based on their video technologies. They won a decision on the order of $200 million or so. It was posted, front page, on Slashdot. Microsoft is not immune. IBM is not immune. No single company is immune when one, or a group of other companies decide to take action. This is the exact same method of checks-and-balances employed by the US federal government with each company serving to "check" each other competitor. All "government sucks" and "all politicians are sleazy" comments aside, this method has worked.

Most (if not all) states have legislation in place to prevent you from having vital assets stripped. For instance, in Texas, you cannot have your house taken from you (unless you default on a mortgage or home loan - something tied directly to the value of the home). Retirement savings (401k, Roth, etc.) cannot be touched either. So, it's not as though this developer will become destitute and homeless. They are free to begin developing on their own again.

Note: I didn't say it would be easy. If this developer has pinned all of their hopes and dreams of financial independence and affluence on implementing this fantastic algorithm, and can't figure out an alternate way to implement it then they chose poorly, and they need to find another project. Or they can roll the dice and push the issue into the legal realm.


Yes, I have. The SCO problem is not patent related at all, but copyright. Which, incidentally, is evidence that the copyright-only-software-development model ahh suggested earlier already is seeing problems (in terms of enforceability). Proving copyright infringement between code is virtually impossible if anyone puts any effort into obfuscating the code. First, you need a reason to convince a judge to have company A reveal their source code to you (so you can check). Then you have to examine the source code looking for any infringement. If it's been obfuscated, then you can only point to similarities in the code. "Well, they called three functions that do kinda the same thing that mine do, and they use data structures that look kinda the same as mine, and the result is kinda similar to mine." Unless it is a gross infraction there will be no clear decision either way. Going to a copyright-only system of software development, for all intents and purposes, removes any real, meaningful protection for people's software development work.

Back to the SCO case... The disagreement is over breach of contract between SCO and IBM, and that IBM misappropriated copyrighted code of SCO's into the Linux kernel. There are no patent issues with this case. However, it does reinforce the idea that companies will monitor each other. This isn't a case of a single developer taking on a huge company. It's letting the cash-endowed corporations slug it out.

Also, to address the courts not being stupid item, I have to assume you've been keeping up with the case every so often. Recently, Slashdot posted an article where the judge in one of the cases (don't remember if it was the SCO-IBM, or RedHat-SCO, or whatever) basically insulted SCO regarding their total lack of evidence. If that's what you were referring to, I don't see how that bolstered your point. Ok, so the judge is giving them more time. How is another X months going to help them find evidence that doesn't exist? After this case is over, and SCO's claim is tossed, are you sure you'll hold the same opinion? Or will you look at the granted delay and say, "Man, that judge was smart. By granting SCO even more time, he took away the we-didn't-have-enough-time excuse from SCO to force an appeal."?

I'm sorry, but I just can't discuss this point. I am not trying to be rude, but every political or economic reply I have ever read you post indicates you are convinced that every politician, every policeman, and every other individual in power is corrupt, greedy, and out to screw the public every chance they get. I'm not saying I can prove you wrong, but having that mindset makes it impossible to have any intelligent discussion. I also believe those perceptions are out-of-step with those held by the majority.

i was only speaking in a general manner at the whole, and not towards the individual .............. (which is funny when you relies that a room full of different people will always look different then the people that make the room up)

and no, my opinion about the SCO case will never change because its been dragged so long I'm sure ill miss the ending (thats what its all about .... i love endings)

I get that, but you are missing the point. When you apply for a patent you want to to be as broad as possible. If you developed this software you would allow for input from more than one type of device, even if it was rudimentary and virtually useless. This would stop anyone else from developing useful software and you would be free to continue providing buggy software with no incentive to improve it.

This is not a scare tactic, it is a proven business tactic of Microsoft who have bought out or destroyed companies simply because they want their technology, or don't want their competition. Now your imagination may lead you to think the sky would fall in, but I would suspect most peoples would lead them to think that competition would be all but illiminated. For 17 years!

Created a competing standard? Isn't that an oxymoron? You seem to be forgetting that a system is already in place that covers this. The copyright holder is free to license others to use their code, and thus be rewarded. This system has worked well so far, and some people have made a tidy sum out of it.

This comment wasn't about patent infringement, it was about the granting of inappropriate patents. It seems the system relies on people with a vested interest trying to disprove others rights to hold a patent, and not on the patent office. Are you saying that all you have to do is apply for a patent and it should be granted, then it is up to someone else to spend time and money proving you shouldn't have it? This is what seems to be happening, and is unacceptable in my view.


Again, you miss the point. The developer does not have "nothing to loose".

Going to?? Where have you been?

The copyright system is in place, working, and people have made lots of money with it's protection.