Slashdot Mirror
Linus says Patents are a real problem
After Michael Widenius of MySQL denouncing patents,
now it's Linus' turn. During
a panel at LinuxWorld, Linus
said: "I agree that the patents as they stand now are a
real problem. Something should be done to make changes to
the laws, either making software patents a non-issue by not
granting them or making them unenforceable. I don't know
how to do that in the next nine months. Start writing your
letters now and if they get enough, maybe they won't think
you're a crackpot. We need a level-headed person to be an
advocate for this." RMS's comment that "three years
ago, Congress passed a law that doctors aren't restricted
by patent infringements in surgery" shows that writing
your representative might not be a complete waste of time.
You can lookup
your representative here. Thanks to Jean-Paul Smets
for this one.
Yes, but you can apply for a patent on anything that you claim you invented up to one year ago. One particularly nasty scenario:
1. Some dude invents a Really Cool Idea, and throws it out to the world as GPLed software.
2. Evil Corporation notices the RCI, and wants to steal it.
3. Evil Co. grabs the code, gives it a superficial facelift, and goes to the patent office. Evil Co's lawyers claim that they've been working on the RCI for several months.
4. Evil Co. gets the patent. The USPS doesn't even give it much thought, since each patent inspector has a quota of patents to grant, not a quota to inspect.
5. The inventor can't claim prior art, since he publicly admitted he thought of the idea months after Evil Co claims they thought of the idea.
6. The inventor can't mount a legal challenge because he is poor compared to Evil Co., and can't afford the lawyer bills. Inventor hopes for some lawyer to offer to work pro bono, in the hopes of getting a civil settlement against the large Evil Co., by somehow proving that EvilCo didn't come up with the RCI when they say they did. Good luck.
I'm going to be bugging my local Linux User's Group (LUG) to start a letter-writing campaign. I recommend you do the same, but there is an important equation to remember:
1 handwritten letter
= 10 form letters
= 100 phone calls
= 1000 emails
= 10000 chain letters
When elected officials evaluate grassroots support for a movement, they take handwritten letters pretty seriously. Form letters, email, and chain letters tend to get ignored because they don't require much commitment on the part of the sender(s).
Get some local LUG'ers together for an afternoon to write some letters. It's okay to give them ideas, but let them use their own words. Focus on quality, not quantity.
A patent provides a nice carrot in some instances to get 'innovation' going. However, one must also realize that the whole point of IP is to promote invention, NOT to line the inventors' pocket. So, if there exits some field for which patents aren't really that useful or actually a detriment, they have no real place. Software development for the most part is just such a field. It moves far too quickly to be burdened too much with copyrights let alone patents.
The patent systems as originally created in centries past are far too pedestrian to do more than hold back a field of endeavor where a mere few years is an eternity.
The time scales of patents and software simply are not compatible.
Besides, no carrot is really needed from the government as software development prior to patent abuse clearly indicates.
They only serve to allow companies to hoard and not actually invent.
Patents are for encouraging invention, not merely lining the pockets of some hoarder. When outside encouragement is not meaningful anymore, then neither are patents.
You mean, like Microsoft has been?
The problem is that as new technologies are patented, free software is effectively locked out of them for 20 years (it used to be 17, but an international treaty makes it 20 now). The owner of a patent can charge us a license fee that effectively makes free software not free any longer. The patent owner can also refuse to license the patent, and they can sue us for an infringement that has already taken place - so I could be sued for principles that are used in my free software today.
My god. That's just AWFUL! I can't believe that commercial software vendors would have the freaking NERVE to keep a free-market economy going, while driving hobbyists out of... business???
Give me a break. Patents are patents. God help people who actually have JOBS if any of this LUNACY that RMS and Linus are theorizing were to come to fruition.
Get a job.
Patents aren't supposed to be granted for inventions that would be obvious to someone skilled in the field of the invention. It seems that if we could just lobby for a sensible definition of "obvious" as it relates to software patents, it would eliminate the frivolous patenting of XOR cursors but still provide patent
incentives for coming up with something truly innovative and original.
For the "patents are evil" crowd, patents are around to encourage inventors to publish their ideas so that eventually, after the term of the patent, they go into the public domain. Withou them, valuable ideas would be kept secret by the inventors.
I'm not an IP lawyer, but I have one, and I've had to deal with this stuff. The answer to the question about dude2 is that, yes, s/he is legally constrained by the patent even if s/he "just got the idea from this article." My company just went through this exact situation. A patent applies whether you get the idea by reading the patent application, an article (whether it discloses the existence of the patent or not), or if your dog told you.
To break a patent you need to be able to prove "prior art," which means proving that the idea was publically available prior to Evil Co's patent claim. Oh yeah, you should have plenty of money for lawyers too, even if you're clearly in the right.
While I agree the the US Patent Office needs to take a good look at how it issues patents (esp. for SW), I am not one of those "all SW MUST be OSS" types who cannot stand competition. I do think that most patents given today are for stupid, obvious *ideas*, but we must not get rid of all SW patents completely. The Patent office needs to start taking a closer look at patent apps, verifying that they protect valid, origional, thoughtful, _objective_ algrothyms and processes; and not "look and feel" or "hey i switched the order of these two things" types of patents.
We _must_still_ protect the legitimate ideas of inventors and programmers with the ability to patent thier ideas. Let's suppose that I came up with the next super-cool K-rad encryption algorythm. Sure I may be able to get a product out first, but what's to prevent a powerful, industry leading company from *integerating* _MY_ idea into thier next release, muscling me out of the market. What of OSS? Well, if you are in the right, and will not compete w/ the owner of the patent, what's to prevent the owner from providing a linked binary (like kernel video drivers) satisfying all involved. If a OSS coder can't get this, then stop crying and COMPETE w/ him.
I will agree that the USPO needs reforms, like having experts in the respective fields go over all patents, verifying the idea; and maybe shortining the length of patents; i do not think that we should get rid of all SW patents. We are a free society w/ a capitalistic economy, and we need to be able to protect the individual's rights, and his ideas- because they are the only things that provides him w/ his income.
Reid G. Ormseth, Esq.
4qk9ormsethr@marquette.edu
(i'm at work- so no cookie)
Patents and copyrights are not part of free market ideology. They are "artificial" government intervention in the marketplace in order to achieve a perceived social good (encouraging innovation and artistic effort).
You may think this intervention is effective or ineffective in achieving its goals, you may think those goals are good or bad, either way it is not a free market method of achieving those goals (whether that's a point for it or against depends on whichever dogma you favour).
I agree wholeheartedly. I am working on developing a GPL'ed streaming media system, including codec, and can pretty much count on infringing patents.
p ression.html
My question is whether GPL'ed code should be protected as expression under the 1st amendment or not. It seems to me it should be, but I may be biased. From reading about cases, it appears that the US courts are inclined to see code as expression (which is why it's copyrightable), though it's not so clear how much protection they're willing to give it. There's that recent case where a judge ruled that source code is not protected expression, but that's being appealed (I'm referring to the cryptography case).
Lynn Winebarger
http://rapscallion.math.indiana.edu/free/FreeEx
I work at a real company. Surprised?
If there were no software patents out there, we probably wouldn't exist. Over the last ten years, we've sunk tens of millions of dollars into R&D of our core product. A couple of years ago, someone managed to reverse-engineer (or perhaps just disassemble) our software, figure out how it worked, and put a competing product out on the market. One programmer, pretty much, who really knew nothing about the field I'm in. He thought, I guess, that we wouldn't sue his pants off and then kick him in the butt. Fortunately, he was wrong.
Not all software is like ours, yes. But at least some of it is. And if he'd been successful (he was selling his for about 25% of our pricing) he could have easily killed us and our only real competitor off. (It didn't help that his user interface was like a zillion times better than ours was.) And that would pretty much get rid of the entire source of innovation in our specific product field.
And yes, I'm sure some free software guy would eventually get tired of not having our software. And I'm well aware that just about everybody here thinks that Free Software is by default better in quality than anything commercial, and that one or two guys working evenings and weekends could probably do better than our $x-million-dollar-team. But, well, you're dead wrong. In fact, I'm willing to bet that in however many years it takes before our primary patents run out, the state of the art for free software that does what we do will be based on those 17 (20?) year old patents, while we will be 17 (or 20) years ahead.
a) Would you care to count the number of patents owned by individuals. And now count those owned by corporations. I think you'll find an awkward imbalance there.
b) Have you tried to patent something? It's expensive. It's expensive to maintain a patent. It's expensive to defend a patent too. A friend of mine spent 10 years working on an artificial heart pump. He's a heart surgeon. He wanted to patent his invention but could not afford the fees, etc. Guess what? It's now being manufactured by someone else who's getting all the benefit.
Patents do not defend the little guy. You are wrong.
a. So, that's not to say that it doesn't protect individuals. companies have more $$ to spend on the research leading up to patents. as far as defending a patent, most lawyers work for free on lawsuits if you loose, take a share if they win. if you're in the right, then take them to court and let the system work.
b. so lets lower the fees to allow the *little people* to apply for them too.
you did nothing to disprove my origional point that SW patents (in principle) are good. lets reform minor points, not throw away all SW patents.
Reid G. Ormseth, Esq.
4qk9ormsethr@marquette.edu
To Mr. 'NOT':
Ah. So patents are part of a 'free-market' economy, huh? Patents are PROTECTIONS from a free-market economy. They keep the big boys from stealing the ideas that the little boys come up with, and they keep the jackals from effectively 'stealing' the entire R&D budgets of the real innovators. In a true free market society, as defined by Adam Smith, these things are self-correcting and we don't need patents for them. Of course, Mr. Smith had never seen or even really thought about the consequences of a free market, but it didn't stop him from making a lot of dumb and so-vague-as-to-be-irrefutable comments on them.
To Mr. Michael:
Funny you should mention Microsoft. Microsoft is one of those cases that patents were DESIGNED for. They didn't exactly work well, of course, in this instance. See, Microsoft doesn't make any innovations. They steal ideas from other people who innovate, and then use them. Cuts out the R&D budget, cuts down on wear and tear on brains. Just the sort of predatory practice that patents were designed to prevent. Now, of course, it hasn't worked in the case of Microsoft. They basically either can afford to buy all their competitors, and get the patents as well, or they can put enough pressure on the competitor to make them cross-license the patents. (C.F. MS steals Quicktime code from Apple, Apple pretty much proves it, MS threatens to stop making IE and MS Office for the Mac, Apple caves and crosslicenses all of the patents)
Every once in a while it amuses me how many arguments here are between two people who don't know much of anything about what they're arguing about.
Of course, the original post that these two were in reply to was pretty bad also. Heaven forbid that someone actually come up with something creative and new, themselves, and then not let the Free Software guys use it. How dare them?
--Adam Lang
thalen@cs.pdx.edu
Tell me about it :)
Well! A *REAL* company! What does your mission statement say? Something like "Real Company will provide medium quality proprietary software with a difficult to use interface to the non-free computing community. By holding numerous patents, selling at inflated prices and suing our competition, we will maximize the thickness of corporate management wallets."
... One programmer, who really knew nothing about [marketing]... his user interface was like a zillion times better than ours ... And that would pretty much get rid of the entire source of innovation in our field." It just sounded funny.
And folks, here is an exclusive for you. This is how this message actually was parsed by my mind: "Someone managed to reverse engineer our software
Now without knowing what his product did, and whether or not the "hacker" mentioned actually liberated their code, or simply their ideas, it's hard to call this one. But I'd say that I have the right to discover how programs on my computer work, by any means at my disposal.. disassemblers, system call tracers, etc. Otherwise, my computer will know more about how the programs work than I do. I'm smarter than my computer, so if I want to know, I'll find out. And once I have the knowledge of how a program works, I can write my own. And if mine is better than yours, apparently you will sue me. You suck.
Someone start a petition on http://www.ethepeople.com! bet we can get a good 200000 signatures w/ the /. effect :)
If the US switched to a first-to-file (instead of the -to a european mind- ridiculously east to fake first-to-invent) system like Europe, then at least some of the disadvantages of software patents would be eliminated. Of course, if the US patent situation gets to silly, Europe will just fail to recognise US patents ( Europe's already rather lax about enforcement in the software and electronics industries (not that that's a bad thing)), and you'll get more and more Free Software development coming out of Europe.
P.S. Sorry about any format strangeness in this message, I'm on a 'doze box
Richard: I'm in favor of it. I'm not anti-business, I'm anti-some-businesses. A parallel might be the effort to prevent business that make paper from dumping poisons into the river. Efforts to stop them weren't anti-business or even anti- the paper industry. I'm against companies that poison good will.
Someday in the next century, the computing industry will stablize, and 20 years will seem just as long as it does in the automotive and every other industry.
When the Cotton Gin or the Steamboat was invented, I'm sure that 20 years seemed like forever. Did they change the rules because Cotton Gin or Steamboat engineers thought that their shit was so revolutionary that their competition had to have less stringent patent rights than everyone else? I don't think so.
Patents, as they relate to "common" inventions (ie, hard stuff - like a new form of a resistor, or something similar), are, IMO, a necessity for individuals and business to protect their assets (namely knowledge). Most "hard" devices patented have a shelf-life of 20 years, and sometimes longer (the gasoline engine hasn't changed much since it was developed 100 years ago). Patents allow the inventor to recoup his investment in developing the device by forcing competitors to find some other way. However, the patent must expire after a resonable amount of time in order to prevent a monopoly on the invention from forming.
It has been said that it costs a lot to get a patent. Actually, to apply for a patent is relatively cheap - it is the hiring of lawyers to write the final draft of the patent (and unless you have actually read a patent, you don't know how damn confusing they are) and engineers to design the drawings - in addition to the fees that must be paid to the patent office. Even with all of that out of the way, you might find that there have been a "prior invention".
Even if you have a patent on file, it still may not be enough to keep someone from abusing it, even for a "hard" item. Case in point - my girlfreind's brother-in-law patented a device designed to sit in front of the air-conditioning vent on an automobile, to hold drinks, to keep them cool in the summer, hot in the winter. His patent was broad enough to cover every variation. However, look on many automobiles (Toyota 4Runner is a favorite), and they have a slide out device _right_in_front_of_the_vents_ to hold drinks! To this date, he hasn't received a dime, though he patented this in the early 80's. As another example, check out the Levitron controversy...
It is very expensive for little guys to set up and maintain patents. It used to be that anyone could go to the patent office, fill out a bit of paperwork and pay a fee, and get a patent. Lawyers, draftsmen, etc. were not needed - all that was needed was a few bucks and a prototype.
Software patent protection times need to be much shorter than 20 years - in many cases a year would suffice, but for some technologies, 5 years might be better - with varying degrees in between depending on the nature of the software.
Software is at once the same and different to other patentable "hard" devices. In theory, software is mechanical - ie, a Turing machine can be used to emulate another Turing machine (similar to how a program can emulate the hardware and code of a Nintendo). However, because of this flexibility, ten people writing code to solve a particular problem will, in most instances, write ten _different_ pieces of code.
There is only so many ways to make a gear - but there are near-infinite ways to create a word processor. Should the idea of a word processor be patented? Yes. For 20 years - no! 5 years? Probably not. 2 years? Maybe. 1 year? Certainly.
If there was some patented algorithm that this one programmer used in his competing product, why did he have to reverse engineer or disassemble your companies product? You see, patents are a _public record_. This means that any member of the public can find out how your product works without having to do any disassembly. So, that makes your story sound pretty bogus to me.
On the other hand, if nothing about this software was patented, then what business did you have suing? You admit that his software does the same job and has a better user interface, and is cheaper. So, it sounds to me like you used courtroom tactics to crush a competitor. In any case, I doubt any of what you said is true. More likely you're one of those people who don't understand that it's dishonest to present hypothetical examples as if they were fact.
I wonder how much Transmeta would be willing to pay Linus if his entire work product were handed over to Transmeta's competitors free of charge? Not much -- but that's basically what would happen in a patent-free world.
I know the left wing of the Linux community thinks everyone should work and live like Stallman (that is, for free, and in a cardboard box), but that doesn't result in an efficient employment of capital. If a company is going to spend money developing technology, they don't want to have to give it away to their competitors on a silver plate. If they have to give it away, they aren't going to spend resources to make it.
Quoth the clueless:
... One programmer, who really knew nothing about [marketing]
Well! A *REAL* company! What does your mission statement say? Something like "Real Company will provide medium quality proprietary software with a difficult to use interface to the non-free computing community. By holding numerous patents, selling at inflated prices and suing our competition, we will maximize the thickness of corporate management wallets."
I see, I see. Let's take this line by line.
>Real Company will provide medium quality proprietary software
Now, if you read carefully (something you may or may not be capable of, but are obviously not interested in) I just said that our interface sucks. Now, I consider UNIX's interface to be completely pathetic. I use it because it is powerful and useful. Could you consider that it may be so with our software as well? No, because you're a pri- uh, software bigot. Or, to put it another way, yes, the fellow improved our interface. But our interface is only the smallest, most easily coded part of our actual software.
>with a difficult to use interface to the non-free computing community.
I don't recall saying that our product isn't available on Linux. Just that it isn't free.
>By holding numerous patents, selling at inflated prices, and suing our competition
Inflated prices? Well, I'm sure you're a good judge of what an appropriate price for our software is. Amazing that one man working out of his house and stealing our [multi-million-dollar (total, not per year)] R&D budget can undercut our prices by so much, isn't it? I mean, really, he has to make the CDs and the documentation and spend all of several months by himself, reverse engineering the software and then recoding it. That's almost as much as the (11 or so years of) work our company [5-20 people] did on originally coming up with the algorithms, refining, testing, refining, testing, recreating, testing, refining... etc, right? So the prices we're charging for our [$100] software just MUST be unreasonable.
Or maybe we just should have come up with the RIGHT algorithm the first time, huh? After all, it's easy to do that if you're writing Free Software. (Feeding the hungry masses is easy too; shocking that none of the FSF people have bothered so far.)
>we will maximize the thickness of corporate management wallets."
Ooh. Look at this! It's that rare beast, someone who is derisive about the search for wealth in America, a man who is concerned about something other than a quick buck. Or wait, no, sorry... he's a libertarian. He would much rather see the guy steal from us and make a quick couple million and retire in ease. After all, stealing is a libertarian virtue, right?
Or really, I guess he'd just rather we spent the millions of dollars and then went under, and our source was released onto the web, because then he'd get our software, but he'd get it for free. (Not that he doesn't now... bet he has Windows and ten thousand dollars of pirated software on his hard drive. But that's neither here nor there.) Of course, then he'd never get anything MORE from our company. But, in this country, shortsightedness is a positive virtue.
>Someone managed to reverse engineer our software
Funny, yup. What I said? Nope. He knew nothing about our branch of mathematics or probability theory, nor anything about neural networks, nor anything about any of the other fun stuff that we have experts in. He knew about reverse-engineering, obviously. And he knew about interface building. And he knew quite a lot about marketing, or else he wouldn't've been a threat.
>his user interface was like a zillion times better than ours
You're not bone-stupid enough to actually think that all there is to innovation is the user interface. I know this because you can spell, after a fashion. So I'm just going to assume that you're being an ass and move on.
>whether or not the "hacker" mentioned actually liberated their code, or simply their ideas
I'll let you rest at ease. He bought a copy of our software, so far as was determined, and never saw our source. And agreed to a license that said he wouldn't reverse engineer our software. And then did.
Perhaps you simply think that software companies shouldn't be ALLOWED to license someone's use of their product. That would at least be consistant. Sort of. Except that you don't have a problem with the GPL... in fact, I bet you'd be happy to see someone sued who was selling software with GPL code in it as non-free-source. No? Yes?
Funny, though, you agree to a license every time you buy a book. It's called the US copyright code. You don't have to buy the book, if you don't agree to the code. But if you buy the book, you're not supposed to photocopy it and sell it to other people. I wonder, is this license as onerous to you as the one on a piece of shrinkwrapped software?
>But I'd say that I have the right to discover how programs on my computer work, by any means at my disposal.
I don't have a real problem with that, personally. Of course, I do think that user licenses do actually mean something, legally speaking. But hey, I don't care who knows how it works. After all, IT'S ALL IN THE PATENTS ANYWAY. Let me say that again. IT'S ALL IN THE PATENTS. EVERY LAST BIT. If you want to know how our software works, go read our patents. (Admittedly, it does take a little more knowledge in our field to understand how to apply the information if you read it from a patent form. Like, you need a half a clue.)
>And once I have the knowledge of how a program works, I can write my own.
Ah. Again, we're back to 'I have a right to everything you've ever thought of.' I guess you could build a self-consistant world around it. It would make the US a much less egalatarian state than it is now... since ideas are no longer worth money, the only people with 'capital' are the ones who control actual property, resources, etc.
People who are born with little property would find it much more difficult to drag themselves out of poverty, since any idea that they have can instantly be used freely by anyone with more resources once they reveal it. (Or maybe they should just hide their idea? At least that way it's still theirs.) The rich get richer and can do anything they want, and the poor stay poor. Sounds pretty Republi-tarian, to me.
(Please, please, please don't try to refute that until you've actually sat down and thought about it for at minimum 30 seconds. Tick. Tick. Tick. Tick...)
>And if mine is better than yours, apparently you will sue me.
Er, no. If you stole our ideas, we sue you. Whether it's a good product or not. Although if you stole our ideas WELL, it pretty much has to be a good product, though it might have a lousy interface.
See, I personally think ideas are MORE important than actual things, and that they should be valued as such. Someone who comes up with a good idea should be rewarded for it, both with money and with fame (if it's good enough.)
You obviously think that ideas, being non-material and infinitely copyable, aren't important at all. And don't give me the bullshit about ideas being too important to allow to be proprietary. That plays in Dover, but it doesn't wash here.
Basically, if any two-bit interface artist can rip off software that costs millions to develop, who is going to spend millions? If any photocopy artist can rip off a book that took five years to write, who's going to spend five years to write a book? Maybe you think free software is God, but I can tell you right now that we've never had any competition from free software people. And if we did, assuming they didn't reverse-engineer our software, they'd be fifteen years behind us. And MAYBE they'd stomp on our patents, but probably not. There are always a lot of ways to solve any given problem. We patented one.
If you want to compete with us, exercise that three-pound blood-cooling organ in your head and COME UP WITH YOUR OWN.
Bastard.
This one is easily answered. He didn't understand the math or statisitcs involved. All he could do was basically wholesale copy bits of the algorithms over from our code to his. If we'd failed in the patent infringement lawsuit, we would have tried for code copyright violation. However, since to my unsure knowledge there haven't been any suits based on the non-exact copying of ASSEMBLY LANGUAGE that have ever been won, it would've been pretty chancy at best.
Luckily, we could prove that his software violated our patents, and, well, voila.
The thing is, you don't need to UNDERSTAND how something works in order to COPY it.
Sorry; I can't tell you where I work. At least one of my managers would be exceptionally displeased with me. Same reason I didn't post my name. Post here, I'll respond.
If you knew the business I'm in, you'd understand. Let's just say there are certain parallels between us and continuous voice recognition technology. Does that frame it any differently?
You are extremely rude, son.
Nobody can stand on the street and make faces,
and suspect to make a living from nothing.
Realize that there are other ways of making
money in our society, than to write
proprietary software.
I would prefer to get paid for working on free
software, like for example the people down at
RHAD, or some of the FSF people (not including
Richard, who is a full-time volenteer programmer).
You can make a lot of money from support, but
if you only care about money, you should at
least not mock those people who work for free,
and who don't charge you.
So please get out of our way, and shut up.
On benefitting consumers: maybe. But if there weren't patents, this piece of software would never have existed in the first place. And even if it had, there would be no more versions of it in the future, since HE (by his own admission) didn't know enough about the algorithm to improve it any, let alone come up with a new and better one. Which benefits consumers more there?
As to the interface thing, I can answer this one with complete confidence: no, we're not using his interface changes. We've come up with a whole new interface for the next version, and IMO it's actually just slightly WORSE than the one we have now. I think it's a point of pride with these people. I will NEVER understand Windows programmers.
And that's at the student rate for members. $10 for me, a student non-member. That's not really what I would consider a reasonable fee. Especially since this sounds like a several page paper. Now, I know you may feel that I'm some kind of commie cheapskate, yadda yadda yadda for thinking you should be able to get information like this for free, but frankly I don't care. I can find plenty of people to disagree with me for free, and I'm sure their arguments are just as good.
Yup, I'm a coward. I'd like to keep my job, at least until I find another one.
But see my reply to my main article on parallels to, for example, omni-voice continuous voice-recognition software.
Hey, are you Ivan Tchackev (badly spelled, I know, sorry). The guy who's always ranting in Jesse Bersts ZDnet forums?
As for why we thought he reverse engineered our stuff... well, we were sure, and the judge agreed. I'm not going to go into the legal details, because that's moot; the decision was, he did.
As for the algorithm, there are a lot of parallels to omni-voice continuous voice recognition software. Begin to see?
Aso for your comment about my statement about FS being 20 years behind: I may not have stated it correctly. I *only* meant that it was that many years behind in OUR NARROW INDUSTRY. Personally, I consider a rehash of UNIX to be both 20 years behind and 20 years ahead of the current state of the art of commercial software, depending on what part you look at.
And as the bit about the guy beating the pants off of us... HE DIDN'T HAVE TO DO ANY R&D! How many times do I need to say that? If you were handed the whole voice recognition engine for Dragon Dictate, and all you had to do was toss off a quick user interface, could you sell it for $20 a pop? YES! Can Articulate Systems do so? NO! Why? Because they literally spent millions of dollars on developing their software. Software engineers aren't cheap, unless they're developing Linux.
If you take away the patents, you take away our R&D department, and you take away ALL R&D in the area, pretty much. He was still 100% free to do research in the area and come up with his OWN algorithms; there are LOTS of different algorithmic approaches to each problem. He didn't; he stole our approach.
I take exception to the word overpriced. I know slashdotters think nobody should ever have to pay for software, but you still seem to find it surprising that someone who steals all of our R&D can undercut us by so much. Doesn't that strike you as the least bit disingenuous?
And if you want to see a non-competitive environment, look at the environment that would abound if all incentive to do research were done away with.
As it is, our environment is at least somewhat competitive. We have one main competitor, who have been tuning their algorithms almost as long as we have. Their software sells for less and has a better interface, and ours is more expensive, faster, and our algorithms are better [more versatile, more accurate, more efficient]. But we have to keep hopping to stay ahead of them.
If the environment were really non-competitive, do you really think we'd still be spending hundreds of thousands (possibly over a million; I'm not privvy to the purse) every year on this stuff? Why?
So, when I use the word sex, do I mean gender, or
fornication?
It's called context, Dumbass!!
Actually, you're wrong. An algorithm is an invention. It's a new way to do a specific thing. You said it yourself, mathmatical, not artistic. Copyrights are for artistic differences- writing, poems, pictures, etc. Many different ways of expressing the same thing. Software (in general) falls under this category- two programs can do the same thing in the same way, yet still express individual differences-- as long as one is not a line for line copy of the other. New algorithms, on the other hand are inventions. Fifteen programs doing the same thing will always use the same algorithm. Furthermore, algorithms are not necessarily line-for line copies of one another. It is not a chance expression of individuality, it is a static expression of math, an IDEA. Like somebody said earlier, patents shouldn't protect a chance or obvious answer (copyrights should). Patents protect new and origional ideas.
One guy was able to build a work-alike of our software. He didn't steal our SOURCE... whether he stole our assembly is another matter, and isn't part of the public record. (I don't know much more than is available from the court papers.) However, without an exact match of assembly, I don't know of a copyright case that has been won on the simple copying of code, so patent was all we had to get him with. Besides, if Fred can produce a condensed version of Jim's new novel that has the plot and characters intact, why can't he produce a condensed version of our application with algorithms intact? Major grey area, copyright-wise.
More consistant code, yup. Produced faster, yup. For less cost? Well, DO YOU COUNT OUR R&D IN THAT COST OR NOT? Come on, did you just fall asleep at the switch? Without our company, he could never have produced this; he didn't even understand what he was copying, nor how it worked. He just copied it and added a user interface. This is fair? This is good? No, no innovation there.
Yes, his interface is a zillion times better than ours. And yes, that's nice. But without our algorithms, it would just be a pretty program that did nothing. I can do that. (And if they weren't working in Windows, I'd volunteer to.) But I can't understand the stuff these guys are doing, either, and I've got a fair bit of background in it.
And yes, an unusable user interface is a real hinderance. But I'd say it's no worse than UNIX's user interface. Awful learning curve, but once you know the app, it works. I use UNIX, too.
The state of the art of our stuff is based on three things: our patents, our competitor's patents [no, not the guy I'm talking about... another company who does the same thing we do] and a bunch of university papers that feature mainly impractical, unuseful approaches with the occasional paper that could be cool if it had any actual information or had been carried through to an actual algorithm. Meanwhile, we're spending hundreds of thousands (possibly more than a million, but I'm not sure) a year on advancing the state of the art.
The patent does cut us off from a huge talent pool, although few have the PhDs in the appropriate subjects to be useful to us. However, in order to make money doing it, we're happy to take that limitation. And, if we're willing to invest 50 hours a week doing this, then you paying $100 for a piece of software that you're going to use weekly at least for the next five years doesn't seem like such an unreasonable thing. To most people, anyway.
That is exactly the point: to force big guy to pay for the idea rather than just steal it. If small guy doesn't want to sell, he doesn't have to.
-- James
While I agree that there should be some protection of R&D time, 20 years is several lifetimes in computers (how long did the 486 stay the leading edge?).
I would think that around 2->4 years would be enough time to recoup the R&D costs.
-- Keith
>No they don't.
If you want me to take this statement right here seriously you'll have to present some argument to the contrary. I have come up with two patents for the company I work for (and recieved a significant amount of recognition (and money) for them.) They were both new ways of applying really old mathematical constructs in a very narrow context.
If I'd desired, when I came up with these two patents I could have instead quit my job here and filed for both, then offered to sell them to my former employer. I considered it at the time, and went to talk to my boss. He told me what I could expect if I did that, and I weighed my options, and decided not to.
If I *had* done so, though, I could have sold them to several companies, used different licenses, and done a lot of cute things to protect my intellectual property. And, if I'd wanted to, I could have gone and talked with people about my algorithms before I licensed them, showed them what I had and what it would mean to their business, discussed them with my old college professor who said I'd never amount to anything, etc, and I'm protected.
Try selling a great idea that's not patented to IBM? They'll ask for details and then tell you they'll get back to you. Nobody buys a pig in a poke, and once you've told them what it is, why, they don't need to license it any more, do they?
>I'm fairly certain that if patent laws were to be repealed tomorrow, IBM would still be standing and you and I would be much better off.
I'll believe the bit about IBM standing. In fact, I suspect IBM might be one of the few that was better off. I don't buy the bit about me being better off.
You make some good points, though. It *is* damned difficult to build something great if what you're building requires a lot of patent/license hoopla. And that *is* bad.
However, it is also bad that if a company wants to reprint a picture of a work of art, they have to contact the piece's owner (or creator, depending on the contracts signed) for permission, and possibly a royalty fee, first. Why? Because it limits the proliferation of art. Because it puts limits on free trade and the free exchange of information and ideas.
But in this case the limitation on the public's desire to see art is acceptable (to me), because the person who created it is deemed to have the right to control when and if these pieces are used, INCLUDING the right to sell this right to someone else if they so choose.
The problem is, if I want my ideas to benefit me, I have to admit that other peoples' ideas should benefit them. Now, personally I believe that people doing things because they're cool and for no personal benefit is a great thing, and I've done a bit of it myself upon occasion. But at the end of the day, if I have a staggeringly exciting idea, I want to get some recognition for it, and if it's saleable I want to get some money from it. And without patents it's just plain unlikely that I would do so. Big companies can already too easily steal from brilliant individuals; I don't think we should make it easier than it already is.
--Adam Lang
I disagree.
I know of no definition of "free market ideology" that excludes property laws. After all, without ownership there can be no market, free or otherwise. Patents and copyrights (as well as trademarks, etc.) simply simply extend the concept of ownership to intellectual property. They are examples of government intervention only in the broad sense that all property laws are backed by the judicial powers of the State.
-- James
You are the man. You are spending zilions on R&D -
by answering _every_single_comment to yor posts.
"Patents are good! I'm good... You are all bad!
What, questions ? Oops sory, can't say anything
more... it's a top secret."
You've made a fool out of yourself my friend.
Jane, you ignorant slut:
It's even _more_ of a problem if you have a job, since you can suddenly go
out of business with no warning. Every time your write a program, you may
be "violating" countless patents. And there's no way you will have any
chance of _knowing_ about it until you get a letter from a lawyer, which
could be years later.
Oh Jesus, now you've done it. All these kids who can't even be bothered to help Mom with the dishes are going to tell you that a) you should work for free and b) if you were any good, you COULD work for free, and carry all your belongings in a garbage sack and use a borrowed computer and everyone would love you anyway. Bah!
If the guy ripped your code, then I'm glad you won.
But imagine this: what if someone looked at your product, and decided to
compete with you without reverse-engineering or copying your code? There
would be a pretty good chance they would infringe on your patent.
For example, it is impossible to make an MP3 encoder that doesn't infringe
the Fraunhaufer patent. It doesn't matter if you copy their code or not.
If, in the end, you create a file that an MP3 player can play, you're going
to infringe.
And what about the case where someone decides to get into your market, and
they didn't ever even see your program. I don't know if Dragon, for example,
has patents (though it's not hard to guess) but the idea of writing a program
that converts speech to text is pretty darned obvious. Granted, the methods
for doing that might not be. But if you tackle the problem and figure out
a way to do it, there's a chance that you're going through the same motions
that someone else did. If that someone else filed for a patent, you're
screwed.
Patents are going to make it so that for every hour of programming that you
do, you are going to have to spend ten hours of a lawyer's time just to
research whether or not you infringed on someone's patent.
This kind of stuff can raise the cost of software development from $50 per
hour of programming to thousands of dollars per hour of programming. This
is WRONG and it can hurt a lot of people.
the concept of 'intellectual property' is shady at best and indeed goes against most of the entire tradition of Western education.
Or DELL, or VA, or Intel, or any other corporate
type who is supporting Linux.
Hm. Some of these companies have cross-licensing
agreements on patents. I wonder what it would
take for Linux to be protected under one of
these existing agreements?
(In other words, if Intel has cross-licensing
agreements, and Intel is funding VA, could
Linux somehow gain some access to the cross-
licensed patents? This example probably wouldn't
work, but there may be a workable arrangement.)
Yes, they do. It's called "copyright." It means that you can't take their code or their binaries and use it without their permission.
Patents are an entirely different matter: they mean that you can't use any patented idea, even if you came up with it on your own, even if you never knew about the patent, even if the patent owner waited until you spent many hours finishing your program.
Software patents stifle innovation because they usually cover only a small part of any program; the program author must go through a huge database of vaguely-worded patents and compare each one to each line of his program to be sure he isn't violating any patents. That's an exhorbitant burden on any law-abiding coder, but the alternative is being slapped with a patent suit up to 20 years down the road.
Software patents also stifle the average small business or independent coder because those small groups don't have the money to apply for patents (so cross-licensing is out), nor do they have the money to directly license everybody else's patents.
These problems don't exist with copyright. You can write a program in the comfort of your own home, and be absolutely assured that you haven't broken anybody's copyright. You don't need to do extensive research to be sure; you don't need to deal with companies to use their ideas. You just make sure not to copy any code from anybody else's work without permission. Simple and efficient.
Software patents are nothing more than cheesy, anti-competitive shackles for programmers.
What is one glaring example of software patent abuse. One that even a congressman could understand.
Nice, well-thought out post, and I'm not really sure that I disagree; however since I'm an argumentative bastard :
From point 1 you'd have to conclude that all patents are bad. (Which maybe you do, I personally don't) After all, you can reduce any mechanical design to a model of physical laws and hence to mathematical truths. Of course conventional wisdom holds that there is a line somewhere beyond which a design for a plow is not the same as a mathematical description of a lever. Now, the issue of where to draw this line in the case of digital constructs is certainly much murkier, but that doesn't mean that the line doesn't exist.
-- James
Hey,
;)), or if he had still done so he could have been forced to license your library to sell with his product.
Don't mind some of the fruitcakes who replied to this.
Almost all of your points are valid.
Not everybody is a student that doesn't realize that papa is paying for their damn study by working for a company that patents software.
Some people need to make money today. It's amazing how these people don't seem to know where their food comes from. And that fancy Beemer they drive round Stanford.
And yes, of course you can put your nifty idea in the public domain, and open source and all that.
Then you can get a nine-to-five at MS to pay for the beer. And in your spare time you can work on your nifty idea! Wow.
Some people choose to secure their nifty idea, so that no one can steal it, get a loan and work the nine-to-five on their own nifty idea.
And that loan needs to be paid back, be it to a VC or be it to the bank, rest assured that you won't get a loan if you don't have some sort of security (patent?).
Of course there are patents that are ridiculous, and should have never been granted. It's obvious that something needs to be done about this. But patents are there for a reason, and why should they not apply to the software industry? Is the software industry an other country, or an other legislation? I don't understand.
Our company too has some patents, and we too release code we write as open source. Not the patented stuff, but other things.
And it seems to me that that's been paid for by the shareholders.
So I ask those complainers now: what is it that you want?
Should we stop doing everything and work for Mac Donalds?
One of the things I would also like to mention is that it would have probably been a better idea to separate the user interface from the algorithms in the first place.
If it had just been a library of some sort, maybe the guy would not have infringed and just used your library (what they call a win-win situation
If people who own patents where to only implement and ask royalties for that portion that's covered by the patent, that leaves the rest open.
Anyways, the last word is probably not said bout this one.
Yawn....
Breace.
You mean you spent Millions on patens, yet a lone one guy develops a better interface than you??
Well now, who is the incompetent company now... i fyou have so much money, why not design a propper interface , or is that somethnig that needs 23 forms and 6 managers to sign? or do you get your lawyers do design your interface?
Personaly, the guy could have released the source for free and his app and taht would have screwed your company up bigtime, he could have done this anonymously too.
Guys, there is the glorious ideal, the wonderfull daydream put on paper. Then there's reality. Companies are going to go on a patenting spree, and most of this stuff is going to sit on a shelf for years. Who does that benefit and how does that help the industry? It has all the class of a couple of 5 year olds fighting over a toy when each has dozens of others they can play with. It is one thing to protect hard labor on a particular product, but you can't put a limit on other's imagination. Who has ever had an ORIGINAL thought? All life is is building on what we see and how we react to what's around us, including software. There's 6 billion people on this earth. Don't think for one moment that your idea is unique. You just put your flavor on it. So patents do NOT do the industry a favor by "forcing unique ideas". In the stead, it severely reduces the available mind-pool on any given project or idea. Been to Russia lately? How much product diversity do you see there? Toilet paper is toilet paper. In the states you have an endless variety of products to wipe your butt with. So, do you want to be stuck with ONE pitifull word processor, or do you want to be able to choose one from many that has the particular features you require? Patents stiffle innovation and competition, and patent lawsuits, regardless of the patented item, usually end up locking development for years and effectively destroying the idea, whether you win or lose. (i.e. - by the time you get back to the development, your idea is old news and worthless...). What's next? Language patents? Syntax patents? One's and Zero's patents? Hell, if you're not carefull, you could be required to pay a fee every time you turn on your computer. It's all foolishness and counter-productive. This is not a matter of preventing the 'copying' of a product, but a matter of power and control. It's a matter of money and who has and who doesn't have. Someone said IBM prides itself on the amount of patents they have. To me, that's saying that they have more toys than your and you can't play with them and "mommy, he's looking at me". No one company should have that much power over an idea. That prevents the expansion of the idea and the evolution of the technology. It is better to be known for innovation and contributing the the mindshare as a whole than hogging little peices of paper that say someone else can't think of the same idea as you. That's arrogent, selfish and unethical. Imagine a patent on jaz. That would quickly rob us of the great variety of jaz players out there, and reduce the mind-pool - soon jaz will be boring and dull. One person's creativity cannot possibly contribute to an idea in the way that dozens, hundreds or thousands. That's what made Linux the best OS in existance. Next thing, OS's will be patented out of existance - who do you think would apply for that one? (hint - company name starts with M).
Mike
Thanks for the reference. I'm not an IP lawyer and I'm certainly willing to concede the point. However, the assertion I was responding to concerned 'free market ideology' which I interpreted as the theory of free markets in the abstract, not any particular implementation of the theory. The U.S. Constitution can place whatever intrepretation it wants on the issue, but the fact remains (in my mind at least, for what little that's worth) that there is no conflict between the concept of intellectual property and free market theory.
To clarify: I was not addressing the validity or nature of intellectual propert from a legal or ethical standpoint. Merely that, inasmuch as it is considered property, there is no economic difference between the property rights granted to intellectual materials and other property, at least as they relate to a free market.
-- James
And so do most of the developers working in one way or another on the Open Source movement. Many have jobs programming proprietory software. Their ideals may land them on the street, but that doesn't mean they are wrong. Patents go against free trade and competition. They basically say you can compete with me EXCEPT here, and here, and here, and so forth. Many patents sit on a shelf, never being developed on. How does that help business or the industry? Patents, as a whole, do more harm than good. They reduce the mindpool on any given idea, and prevent the evolution of technology. There's no such thing as a UNIQUE idea, so don't go spouting that garbage in reply. Our ideas and creativity are DIRECTLY related to our environment. I'd like to see a GPL on technology as well as software. God knows what kind of outcome such a thing would have - imagine hundreds and thousands of minds working on the same project rather than a couple...
Mike
Um, only if it isn't all that useful. If the software is necessary to somebody doing profitable work, the R&D will be funded and the code will get written. If nobody wants to pony up the R&D bucks, or work on it themselves, well, it ain't necessary.
Yes, R&D budgets will suffer if software patents went poof. Yes, innovation in some cash-poor niche market might slow. On the other hand, innovation across the board will benefit from the free exchange of ideas, and the niche markets will still fund whatever R&D they need.
not to belittle your gf's bro...'s work, but that invention seems to fall into "ovious" category. otoh, many things are obvious once you seen them...
You call him "son", but you definitely have the Magic Refrigerator ("never empty") mentality going on. So that makes you, what, fourteen going on fifty?
As for this "from each according to his ability to code, to each according to his affinity for phone support" bullshit, um, yeah. The scales are lifted from my eyes. Thank you, masked man!
And making faces on the street? There's room for one RMS, rather less than a hundred Red Hatters, and... what, a man-hating female tech reporter or two. The rest of us are going to have to get up and go to work in the morning to earn those damned inequitable green pieces of paper you've heard so much about from your working friends. Truly, life sucks, and then you die! Fortunately my livelyhood doesn't depend on a Ponzi scheme of replacing NT with Linux, and thank God for that.
This guy's hit the nail on the head here.
(And no, I'm not the original author guy, just another AC)
Just out of curiousity: Do you weenies put a little picture of Linus at the top of your bed when you're going solo with the palm sisters? It sure seems that way... The guy made a wanko micro-kernel that was derived from someone elses work. Oh, and pulled all the GNU software under his umbrella as well. What a HERO.
Just like to say that there are no Software patents in Sweden, The Swedish government did something good and threw out the EU directive to put patentes on software. Software copyright is all the protection it have, which i think is still quite adequate.
your lawyers were bigger than his lawyers.
Let us know how you fare when you go against somebody your own size and don't resort to cross licensing.
Restrict patents to individuals only.
Seems obvious!
I work for a living. If my company gave away its software it would make either no money or not very much money and at best. If my company made less money then they couldn't pay me as well. By being paid well, I can provide food and shelter to my family. I'd rather be paid well than have free software, and I think most people would agree. "Look honey, I got some more free software" I'd say. "But we're starving!" she'd reply. "Oh yeah, but look! This software is free!"
Just imagine that the company you work for decided to give away its hamburgers and french fires at cost. Would you be willing to work for free or for a lot less money to do so? I don't think so.
So, you LIKE the fact that there are more flavors of Eunuchs (and all its bastard children) than Baskin-Robbins has flavors?
/. seems to be proposing is VERY much a double-edged sword. Seems to me that even proposing such an idea is asking all those little entrepreneurial shops to play a quick game of Russian Roulette!
I get you, in theory. The reality is, though, the chaos that would ensue would plunge all but the wealthiest companies right into the crapper.
Small companies could come up with the ideas (as everyone here seems to insist they do), but then companyX could come along and point-blank CP everything, no cost, and probably do it better (at least throw more resources at it).
What everyone on
I have to say I agree!
Interesting, but unworkable.
First of all, who decides what adds to something andwhat is just window dressing? If someone patents 'voice recognition software: algorithm improvement by modulation down one octave', who's to say whether it actually helps or hurts the recognition? After all, he can't produce a full-fledged product JUST to file a patent, so of course his isn't going to be as accurate as ours... but it would be if he finished it, right? Probably?
Plus, there are a thousand little things, based on our technology, that we could do to improve things slightly. Should we patent all of them? That would be prohibitively expensive. To go back to the voice recognition example, we could do band-pass filters and our algorithm, we could do frequency modulation, we could do flanging, etc etc etc. (Hell, I don't know; I don't do voice rec.) If we had to patent each of these ideas before someone else (who can, remember, read our patents) and think of EVERY SINGLE way in which our technology could possibly be improved... nah.
So basically what you're saying is that ideas are inherantly valueless. Only the implementation of those ideas has value, and EVEN THEN, ONLY WHEN IT RESULTS IN A TANGIBLE PRODUCT.
Think about this for a moment. Is this really what you want to say? That, for example, only those with lots of resources should have a shot at getting more? That the ideas of someone who is poor, even if they benefit the world at large, are inherantly valueless and should benefit that person in no way?
Basically, you're creating the ultimate oligarchy: a state in which only those with capital can make more capital. The only thing that the people without capital have that is worth something in this day an age (besides luck, which is rather undependable) is (potentially) excellent ideas. Which, under your model, are free and even meaningless, since they should be a completely shared and free resource. Thus, those with the money keep the money, by copying the ideas of everyone else, and those with the ideas get the shaft. Heck, soon they won't even get welfare, in the US at least, and then we won't have to worry about them because they'll all starve to death in the streets.
There are pieces of software, sold commercially, that it would be unviable (UTTERLY) under your model.
Nice as it may sound to think that niche markets would fund whatever R&D they need, there are certain niche markets that aren't that organized. Imagine, for example, my earlier example. Voice recognition software. If IBM didn't stand to make a boodle selling it commercially, what would they have done? They could have funded it out of the goodness of their hearts, but I'm sure you'll agree that that's unlikely. They could have funded it for use in their own organization, but I doubt there's really much of that, and even if there were, they'd never have spent as much as they did on it.
Or would they send out a contract to every potential customer in America, saying 'If you'll send us $100 now we'll research voice recognition software and send you a copy, with full free source code, when/if we have a product?' Nobody in their right minds would do such a thing; software development is always risky, and spending money on a pig in a poke goes against the grain for most people.
Or would they send out a contract to people, saying 'If you sign this contract you agree to pay us $100, contingent upon us delivering you a workable piece of voice recognition software, upon delivery?'
Well, aside from the fact that a HUGE number would default on their contracts, thus requiring an immense amount of collections, since after all it's all available for free now so why should we be stuck paying for it... And aside from the fact that a large number just wouldn't sign in the beginning, figuring that everyone else would pay for it and then they'd get it for free...
...it's just like the commercial software model, except some people are stuck paying for the software that everyone else uses. Which, IMHO, is pretty obnoxious.
So I guess you haven't been reading for long, have you? Or are you just incapible of extrapolating from a general case to a specific? I already gave you a viable example: voice recognition software that doesn't require training for individual voices. I gave you every specific of the situation except the actual name of my company and the actual business that we're in (which would in turn give you the name.)
Maybe you should try reading with a mind that isn't quite so completely closed before you start commenting on who the fool is.
Or maybe you should just sod off and leave the openminded intelligent slashdotters be.
>If the guy ripped your code, then I'm glad you won.
:)
Code, no. Binary, probably. But you didn't hear it from me. (Wait, you don't know who I am. Never mind.
>But imagine this: what if someone looked at your product, and decided to compete with you without reverse-engineering or copying your code? There would be a pretty good chance they would infringe on your patent.
No! A good point, and true in a lot of cases, no! Dragon, for example, has patents on HOW speech can be turned into text. There are a hundred different methods of analysis... if Dragon has patents that are similar to ours, only TWO or THREE of those avenues are completely unavailable. And on ten or twenty you'd have to be pretty careful, but it could be done. The rest are totally open and unhindered, although some of them might or might not be actually viable.
>For example, it is impossible to make an MP3 encoder that doesn't infringe the Fraunhaufer patent. It doesn't matter if you copy their code or not. If, in the end, you create a file that an MP3 player can play, you're going to infringe.
See, but this is an awful example! What you should be saying isn't 'can I create an MP3 encoder that doesn't tromp on these patents.' MP3 IS the patent... it's a patent on how to compress a sound file. If you want a non-patented way to do so, you MAKE A NEW FORMAT. There are always 100 ways of doing something (if not more); look at your actual goal. Complaining you can't make an MP3 encoder is like complaining that you can't make a CD player with the name SONY on the front of it.
Now, MP3 happens to be popular. Largely because it's pretty good, I guess. But just because it's popular doesn't mean that everyone should have access to all of the patents. In this case, what you're doing is using Fraunhaufer's research (in audio file compression) to make your own product (an MP3 encoder) and you don't want to reimburse him for his research efforts. If you want to make an audio encoder, and you don't want to step on his patents, do your own research.
>And what about the case where someone decides to get into your market, and they didn't ever even see your program.
Patent searches are cheap. Like, practically free if you're willing to do the research yourself. You sit down and do the patent research, and you find out what approaches have already been tried, and then you come up with a new one. Or, in a lot of cases, you read two patents or three and you see something the authors didn't see and the patent search ITSELF gives you an idea for advancing the state of the art... quite possibly without even conflicting with the patents at all. This has happened to me once or twice. (Well, the first time the idea didn't pan out, so I guess once.)
And I dispute the lawyer thing. If you are decent in your field, you can do patent searches yourself. For example, any patent on voice recognition technology will have 'voice recognition' or 'vocal recognition' or at least both the words voice and recognition in it somewhere.
>This kind of stuff can raise the cost of software development from $50 per hour of programming to thousands of dollars per hour of programming. This is WRONG and it can hurt a lot of people.
Name five.
But seriously, I guess you have to weigh it against the alternative: that ideas are worth nothing at all by themselves, because as soon as you open a startup with a nifty new way of doing things, Microsoft reverse-engineers your software and slam-dunks you out of business. Yes, it happens even with patents, but let's not make it legal, shall we?
Quoth the Breace-man:
;)), or if he had still done so he could have been forced to license your library to sell with his product. If people who own patents where to only implement and ask royalties for that portion that's covered by the patent, that leaves the rest open.
One of the things I would also like to mention is that it would have probably been a better idea to separate the user interface from the algorithms in the first place. If it had just been a library of some sort, maybe the guy would not have infringed and just used your library (what they call a win-win situation
My reply:
We have a library, just as you describe. We charge out the yazoo for it, because people who buy it are basically competing against us, though generally only in some limited market. Buying it and paying royalties on it would have pushed his product up in price to the point where we were competitive again, though with his low overhead he probably could have managed to sell it for $75 a copy instead of $100. (Incidentally, the royalties aren't actually that huge, but the initial capital outlay IS.)
The interface sucks because
1) the programmers don't think that the interface is important, because they can do anything they want with it as fast as thought, because they know the user interface like the backs of their hands. THIS IS LIKE LINUX, YOU GIT! UNIX's interface sucks big brass balls in terms of usability; in that it's harder to learn than algebra was when you were in 8th grade.
2) The managment thinks it's fine, because they don't use our software at all and the windows kind of look pretty.
3) The spirit of QA was broken long ago, and they confine themselves to bugs like 'when I click here it crashes my machine and reformats my hard drive' and are often placated by the response 'don't click there'.
4) We don't have any human interface guidelines, nor do we follow any, not even the over-lax rules of Windows, let alone the guidelines set down by the most self-consistant user interface around. (And if you think that's Linux, try comparing and contrasting uuencode with gzip sometime.)
And yes, the guy could have released his source for free. He still could, he probably still has a copy. (He'd be sued big-time again, of course.) But WHY would he want to do this? He spent months and months of full-time work, living off his assets, just to reverse-engineer this program. Why did he do it? He did it FOR THE MONEY. Say it with me now: he did it for the money.
>My question is, from what you know, how easy/hard was it for the average Joe with a disassembler and some time to kill to decompile your product, to the point where he could make it into another?
:)
Very hard. This guy was a 40 to 50 year old guy who had cut his teeth on assembly programming and was damn good at it.
However, I'm not sure about the code obfuscator. It's an interesting idea (maybe you should patent it? Hee!) but I suspect that anything a computer could obfuscate, a computer could unobfuscate. And there are already algorithms to do that... they're called optimizers.
Any obfuscated code would also automatically have an adverse effect upon speed, just because of the methods you mentioned. Conditional branches, for example, take major boatloads of time from a highly-pipelined processor.
But it's still an interesting idea. Of course, it would REALLY encourage hacking into software houses and stealing code or algorithms, or even just plain espionage. (I know that, after hours, anyone with any talent at all could get past our one patrolling security guard if they were willing to cut some holes in some windows or come in through a skylight.) After all, if you can't prove they had your code, you can't sue them, even if they use the exact same algorithms as you do. Ever seen the game Shadowrun? Kind of a scary idea; corporations coming up with new ideas, other corps sending teams of spies in to find out the ideas, and maybe even to kill the innovator. I can actually visualize some companies doing this, if they couldn't be sued even after they brought the product to market unless the injured party could actually prove that source code or ideas were 'stolen.' Something that's practically impossible to do without hard evidence.
As for current IP law being unkind to some joe who does our research over again and comes up with the same results... well, true. And I'm not sure it is fair, as written. HOWEVER! However, he can, before he sinks any time at all into it, go and look up the patents, ON THE INTERNET NOW, and see what the state of the art is. Now, it's not simple, and it's not fast. But it is there.
And this can HELP him as well as hurt him... if he sees what the state of the art is, and sees five different approaches that work okay, he might see a sixth approach that would work better.
And if it's significantly different, he can get a patent on THAT idea, even if it DOES tromp on the patent of someone else's work that he saw. A derivative patent. And then he can take it to the company and say, look, I've got a way to improve your stuff. Shall I license it to you, or will you license yours to me? Yes, the company can say 'bugger off,' but they can never use his idea without his sayso, just as he can't use theirs.
It's not perfect, but it's better than nothing.
Quoth A.C.:
:)
If you patent something then the algorithms are public, anyone who wants to can read the patent application and discover the algorithms. It is VERY hard to discover algorithms by decompiling source code - much harder than reading the legalese of a patent application.
Reply:
Quite true in some instances. And in others, for example (rehash) voice recognition, unless you are a physicist, an acoustics professor, or some brand of mathematician, you're not going to understand a line of the patents. You won't understand what the math is, you won't understand how it works, you won't understand how to apply it.
Whereas if you see code that applies it (and yes, some people can read assembly almost as easily as you can read C) you can see how it works, if not why. And that, given time and effort, can be reproduced. (Or perhaps just copied from one binary to another; I don't know how he did it.)
>If the programmer decompiled the program and used that in his product then he could be sued for breach of copyright.
If he used the ACTUAL MACHINE CODE from our product in his, he is committing a breach of copyright law. If he used none of our code, but recoded it by looking at our code and then, say, coding in C as he read through the assembly, that is NOT a copyright violation. It is not idenitical to our assembly, and it is so highly unlikely to be identical to our C code that the possibility doesn't enter into real life. He is stealing our intellectual property, but he is NOT violating our copyright.
Even source code is hard to defend in copyright court, where you are supposed to be defending an END PRODUCT, not a manufacturing step. The distinction is subtle, but it is definitely there, and this defense HAS been used successfully before.
>Obsfucation products exist, you guys should use them.
Are you talking through your hat, or do you know something I don't? The obfuscation software *I* know of translates, say, C code into illegible gobbledygook. Not assembly. C. So that it can be compiled into a program without the person compiling it knowing how it works. But they can still reverse engineer the software. I've never seen a single indication that there is anything out on the market to make assembly harder to read.
>If an algorithm is genuinely difficult to discover, keep it secret, make sure everybody signs non-disclosure agreements, and you will make plenty of money before anyone can replicate it.
I love this idea, because it sounds so simple and is so wrong. A company like Microsoft can devote amazing amounts of money to chase something they KNOW is there, if it is perfectly legal for them to do so. If they invest the same amount of money in research, it's much less certain. But I can guarantee you that if it were legal for Microsoft to disassemble our code and sell identical versions, they would be doing so. A sizable outlay for 100% guaranteed results. Who wouldn't, if they had the money and the market force?
>Many of my ideas that were patented took about 5 minutes to come up with - any competent developer faced with the same problem would come up with these ideas.
My patents are of the type that any comptetent software developer who had an undergraduate degree in physics and several specialized computer courses on the subject matter that I'm working with, MIGHT have been able to come up with it after a bunch of trial and error. (I suspect my background in music helped too, though in non-obvious (i.e. non-audio-related) ways.)
Also, remember something. Modern probability theory, at least the basic stuff, is SOOOO incredibly obvious that anyone who looked at the problems would have been able to figure it out, right? So why wasn't it 'invented' until the middle of the 20th century? Hell, calculus has been around a thousand years.
As for software patents in their current form, I have to agree with you, but not because of how they work. Just because of how incredibly, sickeningly easy it is to get one. Used to be you had to have a specific way of doing a specific thing. Now you can get one for ANY way of doing a specific thing. That's evil, and should be destroyed. If someone wants to compete with my company, they should be able to do so, using their own algorithms to achieve the same end that we get. In fact, there is a company like that out there already.
We're better, of course. Nyah.
And won a pretty huge settlement, based on a patent dispute.
Now are you going to argue that Microsoft is an easy target? Or that Stac was bigger than Microsoft, or had more expensive lawyers?
Or are you just going to stop commenting on things you wot not of?
So you consider that in a free market you can define the ability to restrict entry to the market as a property right, give someone this item of "property" and still call the market free?
What characteristics do you feel a free market must have if you do not feel that giving one player total control over the market in a particular (patented) item renders a market no longer "free"?
I'm not saying that patents are automatically bad as a result, just that they are a departure from a free market (and, incidentally that opposing intellectual property laws is not an inherently socialist position by any stretch of the imagination, despite some of the comments about RMS).
OK, first of all I freely admit that I was engaging in a bit of semantic nit picking. Nevertheless, at the moment I find it more interesting than SQL so:
Remember the orginial premise that I was disputing was that patents are somehow inconsistent ("not part of" I think was the phrase) with free market ideology.
My point is that free market ideology is concerned with the free trade of goods, not with preventing monopolies. The only characterisitc that a market needs to be 'free' is unrestricted trade. A monopoly can definitely be a 'free' market, and in some cases monopoly is the natural result of a pure free market approach. It may be inefficient and undesirable, but it's not inconsistent with the theory of free markets.
My comparison to ownership rights was to refute the assertion that patent enforcement was an "artificial" intrusion by the government into the free market. My point was not that intellectual property was necessarily a good thing (that's another discussion) but that if you accept its existence, then enforcement of it is no more artificial than enforcement of any other property right.
-- James
Problem is, it didn't do Stac a hell of a lot of good. By the time the judgement came down, Stac was already pretty much out of business. The fact is, Microsoft is more than willing to pay out 100 mil to put a competitor out of business. It doesn't even make a real dent in their assets. But Stac is gone, and the only disc compression software out there for windoze is M$'s.
Didn't Stac get bought out by someone a couple years ago?
--
No matter how hard you work to make something idiotproof, someone will always come along and make a better idiot.
The onus is not on us to prove that patents are bad. The onus is on you to prove that interference in a free people's business by the government is good. You haven't even remotely demonstrated that.
Patents actually *help* big companies to steal from the little guy. First off, your patent is supposed to explain how to do what you're doing. So once they know that, they simply ignore your patent, modify your idea in the slightest way, disguise it (for software that's *really* easy to do) then sell it in their products. If you are a little guy and have the audacity (and somehow afford a lawyer) to sue them, they will counter-sue you claiming infringement of some of the 25,000+ pages of patents that they own. Now, unless you're truely idle i.e. you haven't actually written ANY software, you'll be hit with enormous costs trying to protect yourself in the counter-suite(s) and move forward on your own suite. The discovery process alone would easily bankrupt the little guy. The way out is to give up, cry "uncle" and let the big guy steal your idea. There are a few real-world examples of littles guys that have hung in there and been demolished (even winning their case.)
Patents only work to help large companies. They have the resources to protect themselves.
Since the little guy can't benefit from patents as a practical matter, and they are a huge threat to them, they need to be abolished (for software at least). In areas like pharmacuticals, were ONLY big players can even play (what little guy could ever pay the millions for FDA approval), patents work.
Well, if you feel that a "free" market involves unrestricted trade, and you feel that patents are not "inconsistent" with free markets, do you mean to imply that a legal barrier to anyone competing with me (without my permission) is not a restriction on trade?
The whole point of a patent is to grant the power to restrict access to the market. That's what a patent does, how can the market then be unrestricted?
Patents were invented by governments in order to achieve an effect other than that that occurred when inventors entered a market and traded on equal terms with anyone that chose to copy their inventions. (I admit "artificial" is a sort of vaguely hostile sounding but with no clearly defined meaning in context type word though, and thus not entirely fair).
--
Ross King
There are a lot of complex and ambiguous issues in the field of intellectual property, but this is not one of them. The Oxford Dictionary of Economics defines a free market as: "A market in which people buy and sell voluntarily, without legal compulsion. Neither the quantities traded nor the price at which trade takes place are subject to control by third parties" Note that it says nothing about barriers to competition. The only requirement is the freedom to set price and quantity freely.
I dislike argument by analogy because people end up disputing the accuracy of the analogy rather than the premise. However: If I own an oil well and you don't then you are certainly restricted from competing with me, yet it is still a free market (as long as there is no price/quantity interference from the government). The government restricts your access to the market by enforcing my ownership claim on the well and preventing you from taking oil without my assent. Now you could go and buy another oil well, but even if I owned all the oil wells in the world, it is still a free market, albeit a monopolistic and socially undesirable one, until the government intervenes and tells me that I can not charge more than $X per barrel of oil.
Obviously ideas are not oil wells and it is debatable (and, indeed, debated) whether or not they can be owned. However that is a philosophical and ethical question, not an economic one (at least vis a vis free market theory) Free market theory has nothing to say about ownership except insofar as it is an axiomatic prerequisite to trade. A patent is simply a legal construct signifying ownership, like a deed to an oil well. Holding a patent will certainly restrict competition in products derived from that idea, just as simple ownership of any factor of production (ie oil wells) will restrict competition in any 'downstream' markets. Free market theory doesn't care at all about that.
Bottom line: Free market theory is concerned about freedom of trade, not freedom of competition. Trade can be free under perfect competition, monopolistic maarkets, or any degree in between.
-- James
Presumably you introduced the definition from the Oxford Dictionary of Economics as one that you find acceptable, and as a clarification of your earlier statement "the only characteristic that a market needs to be free is unrestricted trade".
If this is so, then I'm confused as to why you would follow this definition that refers to a lack of price or quantity control by "third parties" by twice stating that a market is free as long as quantity and price are free of "government" control. Do you find the quoted definition acceptable or not? It seems strange to take the Dictionary as authoritative on the reference to freedom from control of price and quantity but to ignore it on the subject of whom may exercise this control.
If I had the equivalent power in the oil drilling industry that a patent would give me in the fields to which patents apply, then if you were not a licensee of mine the quantity of oil that you could extract and sell would be limited to zero. If you wished to sell to anyone but me, I would clearly be a third party, in what way do you feel that these market conditions are consistent with the definition you quoted of a free market?
Furthermore, I could choose to permit you to extract oil but to restrict you by whatever terms I chose, limiting to a certain quantity or price would be available options, and potentially very attractive ones. These factors would be subject to my control, I might or might not choose to exercise control in that way.
--
Ross King
I apologize for the confusion, I used the terms 'government' and 'third party' interchangably as government intervention is far and away the most common type of market intervention. In fact since the government is the only authority with the power to set and enforce laws, all market interventions are ultimately backed by the government. It really doesn't affect the nature of my argument at all, you can substitute 'third party' for government if you like.
As for your example:
A patent is a monopoly. If you had monopoly power in the oil industry then of course the amount of oil I could sell would be limited to zero, not by any third party intervention, but because I don't own any! I don't see how you could consider yourself a third party to the oil market since you are the one who owns all the oil! As a monopolist, you could exercise whatever controls you wanted and it would be my choice to either accept those controls or leave the market, but the transaction is between the two of us, there is no third party interference there. In fact, your last sentence makes my point. You are free to set whatever terms of trade you like and those who transact with you are free to accept, reject or negotiate those terms. Where is the third party intervention?
To simplify:
1) a patent is a monopoly.
2) monopolies are not inconsistent with free markets.
Refute either of those assertions and I will concede the point.
-- James
No, there are two markets here. One is trading in licenses, the other is trading in oil. With respect to the market in oil, it is not me that you wish to trade with, and who I can prevent you from trading with, and in this instance I am clearly a third party. Arguing that the market in licenses is free does not change the fact that a third party has the power to control the quantity, price, or any other element of your dealings in the oil market. (Remembering of course that we are using the oil market as an analogy for a market in which patent type powers exist).
I don't understand how my being able to grant you a license, or not, would make me 2nd party but the government being able to grant you a license, or not, would make them 3rd party. Either way trade is surely equally "free" within the definition you cited.
Incidentally, it is not true that all market interventions must be backed by the government, though you could say that any lawful market interventions aren't being obstructed by the government.
As I said before, argument by analogy tends to devolve away from the specific topic, so unless someone cares to refute the assertions that 1) patents are monopolies and 2) monopolies are consistent with free markets 3) ergo patents are consistent with free markets, then this will be my last post on the subject.
:) I would be interested in hearing an example of a legal market intervention that is not backed by the government.
However, since I don't want to be accused of ducking the issue, I will respond to your example. In my previous post, I assumed that you were talking about a simple monopoly on oil, i.e. you owned all the oil. Now I understand that you are talking about a monopoly on a license to extract oil. Essentially you own a monopoly on a factor of production for oil. A more realistic example would be you being the only company that manufactured oil drilling equipment. Either way, it really doesn't matter, you're just introducing another layer of indirection. I stated before that monopolists exert influence not only on the market in which they hold a monopoly, but also on all downstream or derived markets. That is simply market power and doesn't constitute third party intervention and is certainly in keeping with free market ideology.
You seem to be claiming that since you can choose whether or not (or on what terms) to sell me a license to produce oil, you are a third party intervening in the oil market. That's absurd. You are simply exercising market power in the oil licensing market. If I don't have the license, I don't have any oil so I'm not even in the oil market and the point is moot. If I obtain a license from you then I can produce the oil and sell it at any price and quantity I choose so there is no intervention.
Even if you stipulate a specific price and/or quantity as a condition for selling me the license, I am still free to accept or reject those terms. The difference between that and market intervention is that I freely accept or reject your terms before even entering the market, as the price of gaining access to the goods I wish to sell. By definition, that is not market intervention because until I accept the terms, I am not in the market. If I already had oil and you tried to specify the price/quantity which I could sell, then that would be market intervention.
In regards to your second paragraph, I never claimed that you granting me a license was different from the government granting me a license. Not all government involvement is intervention. A perfect example is the FCC spectrum auctions. In that case the government is acting as a market player, selling a product. Intervention occurs when I already own a product and the government sets conditions on the price/quantity at which I can sell that product.
The whole point of free markets is that as a seller you are free to set whatever terms you want and the market is free to accept or reject those terms. The fact that those terms of sale might have repercussions in other markets is irrelevant. Your power to affect the market is limited to the extent to which the market accepts your terms. Power is shared between the buyer and the seller. Intervention occurs when a third party unilaterally sets terms of trade within a market, giving the market participants no choice.
I don't see much point in more argument by example. As an economist it's obvious to me that the situations you describe do not constitute market intervention and I can't think of many other ways to explain my point. I will be happy to concede the point if you can refute either of the assertions I made at the beginning of this post.
I will concede the point in your last paragraph. You can have illegal interventions such as organized crime syndicates. However, if we're going to accept illegalities, then the whole discussion is somewhat pointless. I'll just secretly pump my oil without your license and you have no influnece on me whatsoever.
-- James
Okay, long message so I'll quote the bits I'm replying to, I'll quote selectively and in the order that is relevant to my thoughts as I write them, the original text is still above so I hope this doesn't come across as distorting anything :
:)
"As I said before, argument by analogy tends to devolve away from the specific topic, so unless someone cares to refute the assertions that 1) patents are monopolies and 2) monopolies are consistent with free markets 3) ergo patents are consistent with free markets, then this will be my last post on the subject."
and also :
"I will be happy to concede the point if you can refute either of the assertions I made at the beginning of this post."
Just the assertions (1 and 2) or that great big "ergo" at the end? Either way, this seems like a mistake as refuting the first assertion or refuting the logic doesn't prove that the contrary of your position is true, nevertheless....
Okay, 1. Meat is food, 2. Eating food is consistent with being a vegetarian, 3. ergo eating meat is consistent with being a vegetarian.
And of course eating food is consistent with being a vegetarian, vegetarians can eat food, just like monopolies can exist in free markets.
If you want your second premise to be interpreted as meaning that ANY monopoly, by virtue of the fact that it is a monopoly regardless of any other characteristics it may have, is consistent with a free market, rather than just that the existence of monopolies in general is consistent with a free market, then you're going to have to support that, not just take it as true unless proven wrong (argumentum ad ignorantiam).
"Even if you stipulate a specific price and/or quantity as a condition for selling me the license, I am still free to accept or reject those terms. The difference between that and market intervention is that I freely accept or reject your terms before even entering the market, as the price of gaining access to the goods I wish to sell. By definition, that is not market intervention because until I accept the terms, I am not in the market. If I already had oil and you tried to specify the price/quantity which I could sell, then that would be market intervention."
This is more interesting, do you mean that in the situation where I am happily developing / producing / selling product x that uses technique y one day, and the next day you announce that you have obtained a patent on technique y and that any subsequent production must only occur on your terms that this WOULD in your view be intervention and incompatible with a free market but that with respect to anyone entering the market after your announcement it would not be? If so, as a matter of interest, is it the granting of the patent that represents the moment of intervention, or the first use of that patent to affect the market?
"In regards to your second paragraph, I never claimed that you granting me a license was different from the government granting me a license. Not all government involvement is intervention. A perfect example is the FCC spectrum auctions. In that case the government is acting as a market player, selling a product. Intervention occurs when I already own a product and the government sets conditions on the price/quantity at which I can sell that product."
Obviously I've misinterpreted you, it seemed to me that you were saying that if I had the power to refuse you permission to drill for oil unless you did so at the price that I set then this would not be inconsistent with a free market, but you also explicitly stated a couple of post back:
"Now you could go and buy another oil well, but even if I owned all the oil wells in the world, it is still a free market, albeit a monopolistic and socially undesirable one, until the government intervenes and tells me that I can not charge more than $X per barrel of oil."
It seems to me that you were distinguishing the government granting a licence from me granting a licence, on the exact same terms. Unless you meant that the government fixing your price would be different unless they called it a licence and then it'd be okay? Or was the "until..." only with reference to the situation being undesirable? If so then I misinterpreted, I honestly don't see that as the obvious meaning of your statement though.
If legislation is passed granting the government absolute ownership of the right to produce, so that nobody may produce goods or services of any kind without a government licence, and then the government proceeds to issue licences exerting considerable control over how much may be produced, to whom it may be sold, what prices may be set, and so on, and perhaps even include a licence condition permitting them to revoke the licences at will, is this in your view consistent with a free market? I take it you MIGHT feel that the initial introduction of this scheme would be problematical, but once the system was up and running there would be no problem since as far as I can see in this example the government would be acting as a monopoly, but that's okay, people have to deal with them to get licences, so they're not a "third party", people would "freely accept or reject" their terms :
"By definition, that is not market intervention because until I accept the terms, I am not in the market."
If this is not consistent with your view of a free market then why not? If it is consistent with a free market then in your view is there any distinguishing feature of a free market compared to anything else other than just the terminology that is used?
"I would be interested in hearing an example of a legal market intervention that is not backed by the government."
I'll think about that once I've clarified whether there's any action at all you'd count as market intervention (maybe failing to call something a property right?)
--
Ross King
Frankly, all of the examples you present are also irrelevant. Having made the assertion, the original poster, or you as his proxy, are under the burden to demonstrate that it is true for all cases. To refute it, my burden is only to a) show logical flaw or b) give one counterexample. I feel that I have done both.
Having said that, I still don't feel that you have given any examples to support the original claim. (Oh, by the way, your vegetarianism analogy is flawed in that vegetarianism by definition excludes certain foods while free market theory doesn't exclude any kind of market (ie monopoly, monopsony, oligopoly, etc), it only excludes certain actions within markets. Substitute 'someone who is picky about the way their food is cooked' for 'vegetarian' and you'll see what I mean)
I will also selectively quote for sake of clarity:
There is no intervention in either case. If you are producing product X using input Y, which you do not own, and I take ownership of Y then I can sell or license Y to you under any terms that you and I agree to. By virtue of owning Y, I am now a market player and not a third party. If you owned Y and I tried to unilaterally place conditions on the terms by which you sold X (or Y for that matter) then that would be intervention. The issue is not timing but ownership. Actually, I wasn't talking about licenses at all, I was talking about simple ownership of the oil. Introducing licensing only adds a layer of indirection that complicates the issue without changing it materially. Would you accept that if I want to sell oil these two situations are equivalent: 1) You own the only oil well or 2) I own the oil well but you own the license necessary to extract oil from that well? In either case you own a required factor of production for the oil that I want to sell. In both cases I must negotiate with you to obtain either outright ownership or licensed use of said factor. That is not intevention, that is the market at work. Now, if the government, who owns neither the oil well nor the license, attempts to mandate the terms at which I can sell my oil, then that is intervention. Can you not see the difference there? If it was the government, not you, that owned the licenses then I would negotiate with the government and that would not be intervention either. I would say that under a strict interpretation, there is nothing in your example that contradicts free market theory. Before you dismiss this as absurd, consider that your description pretty much describes the real world. By virtue of its monopoly on force, legislation, and imprisonment, any government is implicitly 'licensing' the legal actions of its citizens. This leads to all kinds of philosophical/legal discussions about natural rights, etc. To keep things relatively on topic, I will just state that in purely economic terms, there is nothing to preclude viewing a government as a sort of 'ultimate monopolist' in the market of rights from which all other markets are eventually derived (by virtue of, among other things, the right of ownership) Seen this way, then one would have to conclude that there is no such thing a 'third party' intervention. If you subsume all third parties into your concept of 'market' then obviously there can be no third party intervention. I'm not sure why you would want to make that point since if all markets are free, then certainly patents are not inconsistent with free markets!In any case such a viewpoint, although not explicitly excluded, is pretty much found not to be useful in economic discussions. You have to draw a line somewhere and governments have certain unique characteristics that make it logical to consider them 'external forces' in most instances. For example market participants are assumed to be motivated by profit while this is clearly not the case with most governments. Also, their monopoly on force pretty much makes any discussion of trade ultimately irrelevant.
As far as examples of non-government intervention, I would accept any example that satisfied the following (remember we are talking about legal interventions; I concede the illegal ones, i.e. mafia):
Restraint or setting of terms of trade without the use of either:
- Force, because the government has a legal monopoly on this
- Market pressure (this would encompass things like guilds and trade unions) since this is clearly not third party.
-- JamesDo you know of any (or can we compile some) resources to help out with the research for the letters? I'd be perfectly willing (as I'm sure many other people would be as well) to write a hand written letter to my representatives if I knew exactly what I was talking about.
If you really get down to it, we have a case of a serious double-standard, or at least a broad sweep of inconsistency. You see, mathematical algorithms can't be patented, because they represent absolute truth (and you can't patent the truth). So when you talk about a company sinking in millions of dollars into an algorithm, they'll have to be happy with the knowledge that they've helped humanity progress.
I think the problem is alot of people are getting confused between invention and discovery. Inventions can be patented (as well they should be) whereas discoveries shouldn't. (Hehe, can't you just see Columbus trying to patent the Americas?) People *discover* algorithms, they don't invent them.
There seems to be rampant confusion between patents and copyrights here. It seems to me that every person who claims that software patents are a good thing doesn't realize that the protections that they are asking for are already afforded by copyrights. (And copyrights are good...that's how the GPL gets its power).
:)
If you write a piece of code, nobody else can copy it unless you let them, that's a fact that I'm sure we all have learned to live with (by inventing the GPL). However, generally you can only patent the algorithms which are used in programs. Obviously (to me anyway) there are several things wrong with this:
1. Algorithms can often be mathematically proven. A mathematical proof is a universal truth, and you can't patent that. After all, you discover truths, you don't invent them. (An invented truth would be a lie
2. Algorithms are obvious. Because the basis of computer science is in mathematics, alot of the algorithms are obvious to the people who are smart enough, no matter how much you pay somebody else to discover it.
3. User-interface patents are really just plain bad. The whole industry supports unity of interface.
4. That pretty much leaves protocols. And it's usually in the inventor's best interest to *NOT* patent the protocol, otherwise somebody else is likely to develop their own to save from paying royalties. Then all we have is a bunch of incompatible protocols floating about.
If anybody can seriously say that we need software patents after reading this, I'd be interested in hearing.
You're right, there is a thin line. I don't think *all* patents are bad. And I must admit I don't really know as much about patents as I should to argue about them.
I think things that should be patented are things that are: 1) Novel and 2) Arbitrary. If the USPTO didn't grant patents for things that didn't match those categories, I'm sure alot of people would be alot happier, and the people who really deserve patents, the inventors of the world, would still be happy.
I think there is something arbitrary about a plow, even though it can be reduced to a lever. However I would certainly endorse a patent on a plow as opposed to a patent on a lever.
I'm not sure I have this idea completely fleshed out, but I think I can almost make an essay out of it.
This guy is almost a canonical example of 'ways in which the patent system isn't necessarily totally out to lunch'.
I write GPLed software myself and I find it _very_ hard to assume I _must_ be allowed total freedom to use the work of dozens (?) of programmers over a course of years, _without_ their consent. It seems absurd to me that there can even be an argument- if people have a reason to opt out of open source completely then that must be respected...
Contrast this to the pranks of a Microsoft trying to patent 'a URL' or 'a menu item that is context sensitive' or some such nonsense. Can't you _see_ that there are two different things here? It seems absolutely outrageous that anyone would defend just _stealing_ code mindlessly from this admittedly very proprietary company. How can it be right to just rip that, not understand it, not bother to learn it, not be able to maintain it, just _rip_ it and release a knockoff that uses the _engine_ of the proprietary code, unchanged? Doesn't it sound like this is a very complicated set of algorithms to do what's being hinted at, like it's not so much a mathematical truth but a series of design decisions and compromises with years of tailoring to adjust it to where it starts to really click- and _this_ is what's ripped, not some general concept like 'speech recognition', that would be totally fine.
There has _got_ to be a distinction here, the amount of fiddly detail in this company's pricey product must be orders of magnitude greater than the rip-off patent abuses like 'method of depicting an illustrated button to activate program code from clicking with a mouse' and such tripe. Doesn't it seem different, isn't it sort of startling to hear that the brave resourceful copying programmer didn't _understand_ what was being copied, or even _try_ to understand it? What's up with that? Is it really okay to not even try to work and innovate and develop stuff, is it that beneficial to leech off of people who are willing to put in the time and resources?
Only, in my opinion, if the people putting in the resources are consciously choosing to go some open source route in hopes others will follow their lead... otherwise this is itself a horrible abuse. People must _consent_ to have their ideas shared and copied and proliferated... there are _reasons_ to consent to this, social and pragmatic and personal reasons, but the second you start going terrorist with it, well, how is this different from Microsoft ripping all the Linux code and using whatever they want in proprietary stuff, then having their lawyers stomp on whoever the FSF can afford so that all open source becomes a sick joke and just means 'public domain'?
How is this different from stepping all over the boundaries of _proprietary_ coders? You _have_ to respect those boundaries to even hope to get any respect in return.
To the anonymous proprietary poster: I hope, someday, some free software alternative grows to rival or approach what you people sell commercially. I hope someone codes this. I'll tell you right now, _I_ am not coding this, or even trying to. Why? I absolutely cannot spare the time or resources to attempt such an undertaking- I'm not that good and if I was I still wouldn't be able to put in the legwork, all by myself, to get together a competitive product...
...and this is, in fact, your point, isn't it? Well, consider it a point well taken.
"But a piece of code is closely related to an algorythm (or even an heuristic)" ;) 'Argument to resolve the intellectual disputes over patent theory by putting great emphasis on the concept of heuristics as virtuosic implementations of programming style' *hehehe*
Bingo!
A heuristic is not an algorithm or a mathematical truth. Not even slightly... a heuristic is an assumption or method of making decisions or approach to a problem. A heuristic is _style_. A heuristic is _implementation_, not theory. If theory was perfect and everything was mathematics we wouldn't _need_ heuristics, we'd just look up the answer and go on...
And _this_ is the type of thing which it makes some kind of sense to have idea patents. It's not pure theory- it's a method, _style_, an angle of attack on a problem which is NOT A UNIQUE APPROACH but which might be so artistic an approach that all other methods seem klugey. _That_ is what patents are for, not to stop everyone from whole fields of endeavor, but to provide a payoff for people who can come up with these brilliant strokes of virtuosity, and it's all style, it's all about style and implementation. Nowhere is it necessary to patent math- or, worse, to patent types of computer program that might be written- if you can patent particular virtuosic implementations that just plain outperform the regular ways of doing things.
Heuristics are totally the right angle of attack on the patent problem- maybe I should patent this argument
Seriously... heuristics are the key idea here- read up on expert systems and look into what heuristics are (they ain't simply mathematical formulae, they are _value_ judgements and not just a matter of right and wrong algorithm) and think about that. Heuristics are never truth: they are opinion, and a wise opinion is worth anything.
Software patents should be *restricted*, maybe, but not done away with altogether. I agree totally that patents on downloading music and similar are stupid, but that doesn't mean no one may make money off developing software. If a company sinks thousands of dollars and years into developing a certain program, or even an algorithm, assuming it's unique and new enough, they deserve to get something out of it.
Let Evil Co. have its patent. The question is whether anyone else, say dude2, who wants to use the idea is blocked from doing so. Dude2 can always claim "but I just got the idea from this article", or "I just read this here source code".
Unfortunately, that's not how it works. If Evil Co. gets a patent on the software, then they can enforce that patent in court. Do you think dude2 wants to risk getting sued by Evil Co? I sure wouldn't. The thing about software patents is that they aren't like a copyright. With a copyright, you prevent people from stealing your code and using it in their product. With a patent, you just prevent people from implementing your idea (or an idea you found somewhere). They don't have to use the same code to be infringing on the patent. IANAL, so if someone else can correct or clarify this, you're welcome to do so. This is how I understand the situation to be though.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
What good is it to have a 20 year old idea relating to computers and software get into the public domain? If we've lived without the idea for that long, or only bought software from the company that owns that patent, then how has it helped innovation? Software inventors can't keep their ideas secret. If they want to make money from their software, they have to sell the software. When someone learns what the software can do, it can be recreated.
As far as I can tell, there has never been any need for an incentive to develop new software. It will be done with or without patents. The whole idea of patents just doesn't work with software and the government needs to realize this rather than just blindly doing what business wants. They are not helping to promote innovation.
I haven't seen anything truly unique, difficult, or innovative being patented lately. Mostly just vague notions that companies come up with like "using a database to store information for online commerce taking place in a network environment over phone lines" or some such crap. They are being BADLY ABUSED and since the government can't or won't do anything to address the abuses, they should simply stop allowing software patents. All software patents should be declared null and void. Even if this requires refunding patent fees. It will be worth it in the long run. The 20 year term of a patent is ridiculously long for software. That's like a 200 year patent on any other invention. If they are going to allow software patents, they should only last for 2 or 3 years at most.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
We _must_still_ protect the legitimate ideas of inventors and programmers with the ability to patent thier ideas. Let's suppose that I came up with the next super-cool K-rad encryption algorythm. Sure I may be able to get a product out first, but what's to prevent a powerful, industry leading company from *integerating* _MY_ idea into thier next release, muscling me out of the market.
Actually, you would simply copyright your algorithm and then the company that wanted to use it would have to pay you royalties. Patents should not apply to software. An algorithm is not an invention, it's a mathmatical discovery. I agree that you should get the credit for the discovery and should be allowed to reap the benefits. That's why you would simply copyright it. I don't see any reason for a patent in your example.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
Without knowing what your software does, it would be difficult to argue anything about it, let alone that 2 guys working evenings and weekends could do better.
On the other hand, if you want to be abstract about it, I would say that the guy that took your base code and produced a program that worked "a zillion times better" due to his interface ideas was benefiting consumers more than a government enforced patent on something that may or may not be reproduceable in some other way. Btw, are you now implementing any of the interface ideas that the guy used in the program that he made using your ideas? If you can't honestly answer "no" to that question, then I don't think you get any right to complain about other people stealing your ideas. Patents or no patents, you are either for or against taking other people's ideas and using them in your own software. Which is it?
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
I am just now getting a chance to respond to some of the posts in here, but I don't think I could have said it better myself. Thanks.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
Not just reform minor points. Patents are fine for some industries. I think there should be a new form of protection for the software industry. Something like a patent, but that will last for maybe 3 years and probably have a few other restrictions placed on it as well. I'm trying to figure out the details...
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
You say that fifteen programs that do the same thing must use the same algorithm. If that is true, then it's not an invention, it's a law. Like gravity or any mathematical rule you care to name, it isn't created, it is a truth that is discovered. Now I will agree that it takes knowledge and creativity to push the limits in certain areas in order to discover new truths, but if there is only one way of doing something according to the laws of mathematics or physics, the method should not be able to be patented. An invention that is derived from this discovery might patentable, but not the discovery itself. I'm trying to figure out a better way of doing patents anyway, so I may change my mind on some points if I am persuaded by a good argument.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
Okay, I think you argued your point pretty well. I don't think your argument defends patents as they exist today very well, but you do make a point that there has to be some incentive or guarantee that you will have an opportunity to profit from your research if it produces something useful in order for real innovation to occur. That I can agree with.
Now the only issue is whether the current system of patents actually works. I don't believe it does. I'm trying to come up with an alternative that is fair to all involved. Not sure if I'll post it here or somewhere else yet.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
How about Microsoft patenting CSS. The ideas have been around for years and the W3C (which MS is a member of) was working on the standards and Microsoft just decided to go patent it. Even if they do license it for free, it's still wrong for them to own the patent on something that they didn't invent or even buy and for which examples of prior art exist. It doesn't mean that software patents are evil, but that the current system is woefully inadequate to deal with them due to lack of resources and lack of patent researchers/inspectors (whatever the people at the USPTO are called) who are knowledgeable about computers and software engineering. Basically all the good software engineers can get better (and better paying) jobs, so there aren't any knowledgeable people willing to work for the patent office.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
Tell them to cut the lifespan of software patents down drastically as well. As someone else here stated, "20 years is several lifetimes in the computer industry"(paraphrased). They should last less than 5 years. I think 3 would be ideal.
It's not enough to bash in heads, you've got to bash in minds. - Captain Hammer
Software patents, giving psychiatrists a run at "vague"
--
Ben Kosse
Remember Ed Curry!
If you patent something then the algorithms are public, anyone who wants to can read the patent application and discover the algorithms. It is VERY hard to discover algorithms by decompiling source code - much harder than reading the legalese of a patent application.
If the programmer decompiled the program and used that in his product then he could be sued for breach of copyright. Obsfucation products exist, you guys should use them. If an algorithm is genuinely difficult to discover, keep it secret, make sure everybody signs non-disclosure agreements, and you will make plenty of money before anyone can replicate it. Software patents are only beneficial to companies with ideas that CAN be easily rediscovered by someone else.
BTW, I am a hypocrite - I am named as inventor on 5 software patents (taken out by my previous employer) but I'm not proud of it. Some of these ideas were non-obvious but patents are still unnecessary. The protection for a good idea is that it would take a long time for someone else to rediscover it. Many of my ideas that were patented took about 5 minutes to come up with - any competent developer faced with the same problem would come up with these ideas.
You make a good argument that software patents are sometimes necessary, but in their current form, they do much more harm than good.
http://rareformnewmedia.com/
Obfuscation products that operate on object files DO exist - I can't remember the name but a product for reducing the size of .exe files is also very effective for obsufcation purposes.
Your point about if you know the algorithms exist then you can look for them while it takes more guts to do the pure research when you're not even sure a solution is feasible is a very good one.
Although you didn't go into it I also have to admit that NDAs don't really work anyway - what if some Japanese firm hires a couple of programmers - try suing them in Japan - good luck.
I think we're mostly in agreement on this. Software patents in their current form are daft, but they are beneficial in some circumstances. They are SO bad right now though that the industry would be better off without them. If the non-obvious criteria was tightened up HUGELY and length was reduced to 5 years max, I would be more in favor of them.
http://rareformnewmedia.com/
I'd really like to know what you product is that has an algorithm so special that it can only be discovered with millions of dollars in R&D, and why it is that you assume you competitor reverse engineered you product rather than figured it out himself, but putting that aside you have done an excellent job of quashing your own arguments.
You conclude a post about how you were violated by someone selling a product which you admit was better than yours for less than yours with the statement that in 20 years Free Software will still be where commercial software is now. I find this rather odd not only because I think it's already ahead of commercial software in many aspects, but also because you have just argued that somebody was beating the pants off of you until you used patents to step in and keep your inferior, overpriced software in a stagnent, non-competitive environment.
Though I do think your original post was unclear my remarks about you software being overpriced and in a stagnant market were unjustified. Based on several things in you post I assumed that you were one of those people who tries to pass off hypothetical situations as fact and made blanket statements that, though they are probably true for the majority of software patents today (like the patents on delivering customized content to multiple subscribers, style sheets, and a number of other things that are common sense), but don't necessarily apply to your specific instance. I do know how much R&D cost because I work in it, and I should not have made statements when I didn't know whether or not they applied.
I do however think that the only reason free (as in freedom) software would take 20 years to catch up would be a lack of interest (both among developers and corporations). I work in Telecomm, and I know that if copyrights went away tomorrow my company would still employ a whole bunch of programmers to write software to run on their products. My housemate is right now being paid to write free security software. Just because the software is free, that doesn't mean you can't get paid for it.
To clarify my actual position on software patents I think that they should still exist, but they should not be issued except in the instance of a specific algorithm (a specific way to do CSS, and not the concept of CSS) that is non-obvious and they should only last 2-7 years. I also think that a patented algorithm should never be used in an open standard (like MP3). If patents went away completely I think it would hurt the market, but I don't think it would be devastating. (Devastating to some current markets, not to R&D as a whole)
Right on! We should be giving more software patents out! We should be encouraging more monopolies in the industry! That'll get the free market going! Sure as hell would make my job easier if I had to pay exhorbitant licensing fees everytime I write a line of code! Screw people who enjoy programming and appreciate freedom! We have businesses to run!
No they don't.
Fact is, patents (and in particular software patents) are bad for the consumer and usually help large companies more than small ones or individual inventors. Ideas and innovation don't exist in a vaccum; all ideas build on a heritage. It's impossible for an inventor to stand on the shoulders of giants when only other giants can afford it. I can't build an insanely great Quicktime player because I can't afford the patent fees for the codecs. Free software doesn't exist if all technology is patented away.
I'm fairly certain that if patent laws were to be repealed tomorrow, IBM would still be standing and you and I would be much better off.
I'm sorry it took so long to reply to this. I had most of a reply written out last night when Netscape ate itself. I was too tired to redo the whole thing then.
You said it yourself (emphasis mine):
To be more clear: I'm not saying that patents never protect the little guy. I'm just saying that the protection is hardly reliable. Big companies like Microsoft still can and do use strong-arm tactics to get what they want. You brought up the Apple example, and their hardly little guys.
You're one of the lucky few. Most companies claim all IP rights to their employees' ideas.
This is where we have a disconnect. Obviously if ideas aren't treated (and protected) like physical property, you can't buy and sell them. Ideas don't behave like physical property, and therefore I don't consider intellectual property to be a valid concept.
People who have great ideas are valuable. I don't want to buy a great idea, I want to have people who come up with great ideas working for me. Sure, other people may be able to use those ideas too, but I'll get the best access to them, and I can ensure that the idea-generators are working in my interest. Look at law: lawyers can't patent their lines of argument. The difference between a good lawyer and a bad lawyer (you can read that as "highly-paid lawyer and less highly-paid lawyer") isn't that the good lawyer has strict control over the best lines of argument. It's that the good lawyer has the best knowledge of law and the effective use of argument, and has the best chances of being able to use or create effective arguments. Software is not really all that different.
You're approaching this from the perspective of an inventor. I'm approaching it from the perspective of a consumer. All inventors are also consumers, but the reverse is not true. This means that, although you benefit from your own patents, you also suffer somewhat (along with everyone else) because of others' patents.
The issue is one of relative versus absolute benefit. Having a patent may put you in a better position within society, but having no patents puts society itself in a better position. The licence fees you get from your patents may give you a higher standard of living than your less inventive neighbor, but allowing ideas to be used freely will make the general standard of living for everyone (inventors included) higher. To put it into more concrete terms: How much of your income is coming from patent fees and how much of it goes into fees for other patents (and this includes patent fees paid by other companies that are included in the price of products you buy)? It's impossible to calculate, of course, and I'm not positive that it works out to more total cost than benefit, but the question is rarely asked and very important.
Can't argue there :-)
Let me emphasize that it's not always (or even usually) the creator who benefits most from IP laws. Most creators have lawyers, managers, publishers, etc. Often the creators don't have the property rights, having signed them over to their publishers (i.e., large companies). Big business really does have the upper hand here. Do you remember what happened with the Beatles back catalog? Michael Jackson was able to outbid Paul McCartney for the rights to his own songs. Now Jackson collects the royalties.
The flip side of that: "if I want other peoples' ideas to benefit me, I have to admit that my ideas should benefit them." The question is whether we gain more than we lose by allowing restrictions on the use of information.
Well, I'm not sure that I believe that you can't get recognition without a patent. Good ideas tend to stand out on their own merit. Getting money from it is tricky, but let me reiterate my two main points: 1) As a creator of good ideas, you as a person have value greater than the sum of those ideas (and this is speaking of strictly financial value here) and smart people/companies realize this; 2) The benefits of a patent to a patent-holder are obvious because it increases his or her relative wealth in the society, whereas the benefits of a patent-free society are subtle because they increase the general standard of living for everyone. So the inability to get paid for use of your ideas in the absence of patents is a red herring of sorts.
Well I'm the last person to want big companies to benefit at the cost of the individual. But I'm not sure that patents prevent this, or that the lack of patents makes it easier. Using a term like "steal" is loaded and betrays your assumption that ideas are like physical property. If you just say that large companies are reusing ideas created by individuals, then you still have to demonstrate that this harms the individuals more than it does when you restrict everyone's ability to freely use information and ideas.
Hum. This is almost weird enough to be a troll, but I'll run with it. Let's look at this.
One guy was able to reverse engineer your product and build a work-alike. One guy. Assuming that he didn't steal your code wholesale (which would be an entirely different issue), it is safe to assume that the code one person writes is less complex than code written by a passel of semi-coordinated coders. More consistent code, produced faster, at less cost (he was able to sell for 25% of your cost), with the same or better functionality. Nope, no innovation there. I would argue that this guy could counter your algorythm patent with a method patent.
You admitted "his user interface was like a zillion times better than ours was" Nope. No innovation there either. I would posit that an unusuable product (yours), no matter how much genius is embedded under the covers, is basically worthless. This opens the door for innovation in the human-computer interface arena.
Your argument about patents and time basically says that the state of the art for your field is based on your single set of patents. One of the most positive traits of sucessful systems is that there is always more than one way to get a task done. If you limit yourself to the view that your discovery is the only way to skin a cat, you end up with a lot of skinned cats, but not much progress in the cat-skinning technology. Your very declaration (patent) cuts you off from the greater community that could spur still greater discovery. Someone else always find a better way, and you end up one stuck behind the curve, with a patent for cold fusion, or some similar state of fuddy-duddy-ness.
I think not...(*poof*)
Where's the line?
A surgeon can safely ignore a patented method because a life might be at risk. That answers a question of "why", and at the same time preserves the idea that innovators deserve some recognition. (Cleverly sidestepping the issue of ownership.)
But in the case of software, there's the issue of obviousness and fundamental truths. US Patent law says that your patent should be for something that isn't obvious. But a piece of code is closely related to an algorythm (or even an heuristic), and those are closely related to mathematical proofs. A proof is a truth. How can I patent a truth? How can I patent a line of reasoning to a truth (a proof) that by its very nature is designed to make the truth obvious? How can I patent a mathematical method that implements a truth? What happens to software based on this method?
When you follow this line of reasoning back to software, it's obvious imho that there is some point at which there is a transition, and it becomes reasonable to identify a person or group of people with an idea. Whether that connotes ownership is debatable, of course, but as you travel the line between a mathematical truth and a software implementation, there is a point before which no person or group should have control of a basic idea, method, or truth.
Where do you think it is?
I think not...(*poof*)
hmm. Perhaps you're investing a bit more humanity in the notion of a heuristic than I intended. A heuristic is a rule of thumb, or a way of doing something that _generally_ works _better_ than another way. The value embedded in the notion of "better" isn't so central, the way I see it. "Better" can be computed, more or less. And the concept of "generally better" is really an issue of statistics, which is derived from experience and not value. Perhaps that's a little cold, but I don't quite make the leap to "style."
So, I disagree. While it's true that a heuristic is not an algorythm or a truth, it is a fuzzy approximation of same: algorythm vs method, truth vs best choice. I don't think you can distinguish so cleanly between them.
I think not...(*poof*)
Don't be coy; bring it on! Name the product, Coward. You choose the metric (speed, space, ...), and let's see whether free software can do 80%, which ought to be enough to take the sting out. Your boss ought to be real happy when a comparable product is deposited on an ftp server somewhere outside your reach.
:-)
What are we talking, optimization software, layout, simulation?
Should it have a nice GNOME interface?
"Intellectual property" is a misnomer. In the U.S. in particular, patents and copyrights are authorized by the Constitution for the sole purpose of promoting "the Progress of Science and useful Arts" (Article I, Section 8). In other words, copyrights and patents are government grants meant to promote a social good (and certainly not an absolute property right in any meaningful sense).
-----
The problem is that as new technologies are patented, free software is effectively locked out of them for 20 years (it used to be 17, but an international treaty makes it 20 now). The owner of a patent can charge us a license fee that effectively makes free software not free any longer. The patent owner can also refuse to license the patent, and they can sue us for an infringement that has already taken place - so I could be sued for principles that are used in my free software today.
As a volunteer contributor of free software, I can't afford an expensive lawsuit with a deep-pockets corporation as the plaintiff. Something must be done about this if free software is to continue to be viable in the future.
Thanks
Bruce Perens
Bruce Perens.
I certainly don't.
That's the first I've heard of it. It looks like a conventional monolithic Unix kernel to me, and I'm sure Linus would say the same.
Oh, aren't we supposed to use GNU software then? I don't think Linus ever claimed to be responsible for any of that, even if users and distributors have blurred the line.
The glorious revolutionaries behind your-choice-of-BSD are probably much better leaders than Linus, who hasn't achieved any splits so far.
So far, over 2,600 folks have signed. Sorry, the site doesn't allow non-American folk to sign. Not my fault.
Kythe
(Remove "x"'s from
Kythe
I think people in the computing industry need to put things in perspective. The difference between patents stoping a surgical procedure and patents stopping a software program from being written is that, in a surgical procedure, someone may die because of a patent. Unlikely, of course, but "patents could kill people" does make a convincing argument to congress.
The way to show the folly of software patents is to show that software patents usually attempt to "patent" obvious things. Such as using XOR to draw on a screen, or paying to download music.
Then again, our legal system is designed to make put pour as much money as possible in to the courts and lawyers.
- Sam Trenholme
The secret to enjoying Slashdot is to realize that it should not be taken too seriously.
Hope you're still reading this...
You seem to have a good grasp of patents, software and their interactions, and I'd like to pass an idea before your eyes:
What if patents expired when someone improved on the idea? You say you spent millions on your ideas, and they required a good understanding of your field. I agree that your research investment requires protection from people stealing those ideas. But what if someone else comes with a product which adds to yours? Should they be denied access to your thinking? If their addition is minor and obvious then surely you should have thought of it? If their ideas require a deep understanding of the field, and they have out thought you, then shouldn't they get recognition?
I've been thinking of this as a means of allowing patent protection to prevent carbon copy software. It protects your research but forces you to stay at the cutting edge in your field, else the competition takes the prize, and what you have falls into the public domain (although it would still be protected by copyright). It also renders silly patents, like microsoft's style sheets patent, useless since a freeware developer, or W3C can improve trivially on the idea, and freely licence the intellectual property. It would then remain in the public domain, but someone else with a really clever idea which builds on it could still get a patent. It also means that the 20 year time period would self adjust to the pace of development in whatever field the patent was taken out in.
-Jeremy
Patents on software are just a bad idea. They don't help you make money, they help reduce demand for whatever it is that you're selling.
Imagine if Microsoft patented the concept of a "window". There would be evidence for prior work in this area, but just for argument's sake, let's say that's what they did.
Patenting that wouldn't make us OSS people say, "Gee whiz, look at that cool MicroSoft technology. Guess we're just all gonna hafta buy Windows and run it now, 'cause we can only use the console"
It would mean that a lot of software business would go out of the country, that MS would rule the home desktop COMPLETELY, and that everyone would be beholden to them.
Until, of course, the patent expired, and years of seething hatred for the company that had restricted creativity through irresponsable licensing fees and patent law rushed out and swept them from the market completely.
Of course, this has happened before. Binary XOR anyone?
--glyph
Glyph Lefkowitz - Project leader, Twisted Matrix Labs
Writer, Programmer - Not a member of the TSU
an operating system and its attendant utilities?
Monkey see acronym. Monkey repeat acronym.
---------------------------------
"The Internet interprets censorship as damage,
For companies who believe they live or die on their 'trade secrets' and intellectual property, the thought of loosing those things is anathema to them. What the open source community needs to do is have agents (like VA Research, etc) who secure software patents on our behalf. There's nothing stoping a commercial enterprise from recoding some algorithm which happens to be part of an open source project. We do the innovation, and they benefit freely from it.
If the algorithms themselves were patented, then we could set any price we want for licensing those technologies to companies. A small company developing some product might have to make a donation to the FSF, or donate X hours of a workers time to helping the community. Perhaps the biggest win for us would be exchanges with large companies who would really benefit from algorithms developed within the open source community. We can say, we will make this algorithm free to everyone, if you do the same with such and such a patent.
What we would need would be a project which manages the intellectual property of the open source community. This project would aim to ensure that the only one's using the patented algorithms are open source projects or those who have made specific arrangements with the project.
Microsoft in particular is going to use its intellectual property as a weapon against open source software. They will patent every silly thing they can, and use their enormous resources to enforce them. Some of these patents will (and do) be very simple and fundamental to the software we develop. And sometimes we won't be able to conceive of alternatives which are as good.
In short, we can have some real bargaining power, and beat these guys at their own game, or we can moan a lot. Getting rid of software patents is just too much of a big ask. Given that many people in the open source community are university researchers, we would definitely have an edge over the competition in terms of innovation.
You probably also think it's great that Microsoft has a monopoly. Because that's what this is about. If a certain company can grab a couple of strategic patents, it will become a monopoly.
Patents.
Create.
Monopolies.
I mean, it takes 20 years for a patent to expire. 20 years. That's the IBM PC. Hell, that's Microsoft!
And it's not just driving hobbyists out of business, but competitors as well. If a competitor can't touch an idea for 20 years, there _will be_ no competition.
Hey if you happen to work for the competition, you might even lose your job. Then what will you do?
You said it yourself, Big Companies (MS in this case) don't innovate, they buy. So a small software house develops something really great and patents it. Big Company wants to use it, so they want to buy it. Small Guy isn't selling. Big Company buys Small Guy instead. It's how things work...
If it's a useful patent some big guy will want to have it, and pay for it. And once they have it they will defend it with all their power. Patents gravitate towards big companies.
Now I'm not saying that patents should be abolished completely. No. But they do have to change. A 20 year expiration date is just way too long in this field. In 20 years time, the patented idea is horrendously outdated anyway. (Well, most of the time they are.)
Patents provide the basis for innovation. Without them, most will resort to using other people's ideas and inventions. Patents ensure the creators of inventions are rewarded for their effort. IBM prides itself with them number of patents they receive each year. Is this why IBM is still a dominant force in the world? You betcha'
--Ivan, weenie NT4 user, Jon Katz hater: bite me!
--weenie NT4 user: bite me!
"Computers are nothing but a perfect illusion of order" -- Iggy Pop
If you have access to ACM's digital library, then read:
1 998-41-10/p27-hsu/p27-hsu.pdf
Maintaining high living standards through innovation, strong patents by Richard C. Hsu.
http://www.acm.org/pubs/articles/journals/cacm/
--Ivan, weenie NT4 user, Jon Katz hater: bite me!
--weenie NT4 user: bite me!
"Computers are nothing but a perfect illusion of order" -- Iggy Pop
Well, I did say you can read this article if you have access to the ACM Digital Library. I am an ACM member (as any computer scientist should be) and I find their library VERY useful for when I need to research articles. The article itself isn't much of a paper, it is merely a two(?) page column in the CACM.
--Ivan, weenie NT4 user, Jon Katz hater: bite me!
--weenie NT4 user: bite me!
"Computers are nothing but a perfect illusion of order" -- Iggy Pop
Adam Lang Said...
>Tim Moore Said...
>>No they don't.
>If you want me to take this statement right here seriously you'll
>have to present some argument to the contrary. I have come up
>with two patents for the company I work for (and recieved a significant
>amount of recognition (and money) for them.) They were both new ways of
>applying really old mathematical constructs in a very narrow context.
Then it really isn't your idea is it? It's an idea that you stole from
some old mathematicians but lemme guess, they don't see dime one.
>If I'd desired, when I came up with these two patents I could have instead
>quit my job here and filed for both, then offered to sell them to my former
>employer. I considered it at the time, and went to talk to my boss. He told
>me what I could expect if I did that, and I weighed my options, and decided
>not to.
Actually, you couldn't. Your employer owns your ideas. They aren't paying you
because you`re a snappy dresser. You can no more take ideas from your company
and sell them to others than you could take the shoes you made in a factory
and sell them.
[snip]
>Try selling a great idea that's not patented to IBM? They'll ask for details
>and then tell you they'll get back to you. Nobody buys a pig in a poke, and
>once you've told them what it is, why, they don't need to license it any more,
>do they?
This is what contract law is for. Idea's for movie scripts cannot be patented
(yet). However Art Buchwald (or someone with a correctly spelled name) managed
to win a dumptruck full of money from a big movie studio for telling them that
they should make a movie about Eddie Murphy as a prince who must go to New York
and "get down with the brothers".
>>I'm fairly certain that if patent laws were to be repealed tomorrow, IBM
>> would still be standing and you and I would be much better off.
>I'll believe the bit about IBM standing. In fact, I suspect IBM might be one
> of the few that was better off. I don't buy the bit about me being better off.
The people who would be better off are those who not only come up with their
own ideas, but also implement them. In addition the people who like to see good
ideas actually used to improve their lives would be better off. On the other
hand if you feel that people who include video in a video game, or transport
music on the Internet owe you a living then patents are great.
>You make some good points, though. It *is* damned difficult to build something
>great if what you're building requires a lot of patent/license hoopla. And that
>*is* bad.
>However, it is also bad that if a company wants to reprint a picture of a work
>of art, they have to contact the piece's owner (or creator, depending on the
>contracts signed) for permission, and possibly a royalty fee, first.
>Why? Because it limits the proliferation of art. Because it puts limits on
>free trade and the free exchange of information and ideas.
And yet that artist does not have to pay anyone for the right to create an
original work. George Lucas can have sinister music play in his movies
whenever Vader enters the scene, and not pay a royalty to all the people who
have used music to set an emotional tone throughout recorded history. Yet,
if my game does the exact same thing, I owe Lucas Arts some money, or I may
even be forbidden from doing so at all.
>But in this case the limitation on the public's desire to see art is acceptable
>(to me), because the person who created it is deemed to have the right to
>control when and if these pieces are used, INCLUDING the right to sell this
>right to someone else if they so choose.
That's copyright, not patent. Copyright means no-one can use what you create
without your permission. Patent means no-one can create without your permission
>The problem is, if I want my ideas to benefit me, I have to admit that other
>peoples' ideas should benefit them. Now, personally I believe that people
>doing things because they're cool and for no personal benefit is a great thing,
>and I've done a bit of it myself upon occasion. But at the end of the day, if I
>have a staggeringly exciting idea, I want to get some recognition for it, and
>if it's saleable I want to get some money from it. And without patents it's
>just plain unlikely that I would do so. Big companies can already too easily
>steal from brilliant individuals; I don't think we should make it easier than
>it already is.
Patents make it easier to attract investment, as this is a useful asset that
persists even if the company tanks. It is also easier to collect damages if
you find that someone is using your idea, but they got that idea from someone
else you told. In theory publication is improved, although CS has a long and
robust tradition of publication. What you give up is a new way for people to
steal your ideas. Forbid them to think of them in the first place, or deny
required supporting ideas. You also require a large legal investment that will
kill what small software companies might escape Microsoft.
An Object at rest CANNOT BE STOPPED! -The Evil Midnight Bomber What Bombs at Midnight
>Basically, if any two-bit interface artist can rip off software that costs millions to develop, who is going to spend millions?
I think the fact that digital information of _any kind_ can be duplicated exactly and distributed at zero cost, serves to demonstrate how much that information is _actually_ worth.
Companies artificially give their information worth by patenting it, not by first spending to research it. All the money spent on research in the world does not make information worth money.
Only when companies realize how inherently worthless their information is will we begin to see some change in the way things work in this country, and throughout the world.
People will then write software whose primary purpose is to be used, not software whose primary purpose is to be sold.
It's not just software. In a previous life I worked for a company that made pinball machines. You could get a patent on darned near anything (software, mechanical, electrical) by taking an existing invention and placing the words "in a pinball machine" on the patent application. Is it really a new invention to combine an existing technology with an existing product? The patent office apparently thinks so.
Chelloveck
I give up on debugging. From now on, SIGSEGV is a feature.
I think you should post your notes (if you have 'em in postable form) from your Thursday talk -- at least the part concerning patents.
How many of the loudmouth ACs here who don't think that patents matter have seen many GIFs on the web recently, for example. Wanna know why? There is this little matter of the Unisys patent claim on LZW compression, that's why (one of the points Bruce mentioned).
Microsoft's patent claim on cascading style sheets might or might not matter, but the fact is, the gates have been flung open by the PTO and we are already sliding down the slippery slope. This is a sleeper issue and few people seem to be paying enough attention to it.
The other comment Bruce made Thursday is about the looming internationalization of this and other intellectual property issues. Even though Bruce is wrong about no-code licensing for ham radio (sorry, had to get that in), his point about being basically unable to even raise the issue effectively before an international forum is correct. Patent rights are now entangled with, among other things, the ongoing development of GATT and the World Trade Organization and, get this, the biodiversity treaty first launched at the Rio world environmental conference in 1992. The reason for that is Monsanto and other companies paving the way for patenting gene sequences in their plants, fertilizers and insecticides (and you thought Microsoft had a plan for world domination! hah! ever think about where your *dinner* comes from?!?!).
By the way, this would be a very good time for the open source folk to talk with the organic ag folk. We will need to GPL genetic sequences real soon, or else the likes of Monsanto will own most if not all of the genetic material for food production within your lifetime. They've even developed a product called the Terminator which prevents seeds from regerminating, thus forcing farmers to re-buy seeds from them every year.
Don't stop with GNOME, folks. Think genome.
--------
Bill Gates Is My Evil Twin.
First off, let me say that I'm not here to flame. Your case looks to be one in favor of the patent system; of course, there are others that slight it equally well. IMHO, I think a change is needed for a more equitable/sensible system, but the jury's still out on what that would be.
My question is, from what you know, how easy/hard was it for the average Joe with a disassembler and some time to kill to decompile your product, to the point where he could make it into another?
For OSS-goers, binary code is assumed to be a black box, but as your case shows, it isn't quite as black as it might be.
More to the point, would it be possible to build some sort of compiler add-on which would, in a sense, obfuscate the compiled code? i.e. add jumps all over the place, encrypt whole blocks of instructions, etc. etc., in a way that would make binary disassembly more difficult by several orders of magnitude? to the point of sheer impossibility?
(in a way that would not adversely affect performance, of course. I think it can be assumed that a compiler-level tool could generate better spaghetti code than a human programmer).
If binaries can really be made as impenetrable as everyone assumes they are, it would then be sufficient for an enterprising company not to publish their algorithms, nor other details of their implementation. At least then IP laws could regain the "independent re-creation" provision (as similar to that of copyright) without adverse effect, and all the worry of patents stifling innovation would be laid to rest.
In your case, some joe stole your code, which I can agree is wrong. But if some other joe were to perform similar research as your company has, and completely independently develop his own approach which may by chance have a few similarities to yours, current IP law would be very unkind to him. And that, I think, is also wrong.
We need a change.
iSKUNK!
Software patents are a waste of everyone's time. They don't do anything for anyone (except for IP lawyers).
If you have a great idea, publish it or demo it fast. Then no one can patent the idea. The right to patent an idea expires one year after public disclosure.
Let Evil Co. have its patent. The question is whether anyone else, say dude2, who wants to use the idea is blocked from doing so. Dude2 can always claim "but I just got the idea from this article", or "I just read this here source code".
The worrisome case is: what happens if dude2 works for a reasonably large company, that Evil Co. perceives as a threat? Can Evil Co. really sue? Is the public availability of the idea enough to get the case thrown out? I don't know. Any IP lawyers out there?
The patent owner can also refuse to license the patent, and they can sue us for an infringement that has already taken place - so I could be sued for principles that are used in my free software today.
Suing you would be completely ineffectual. You can't compensate the patent's owner, and you can't fix the problem either, since you don't own the source code.
Who exactly would the patent owner sue? The only answer that makes sense to me is: some company that competes with the patent owner, and which uses the open-source software that contains the patented idea. (But who ever said that patent law makes sense?)
http://www.ethepeople.com/affiliates/national/full view.cfm?ETPID=0&PETID=98938
It has about 2600 signatures.
In a world that is Free and Open, who needs Windows and Gates?
I'm afraid I don't see the distinction between "discovering" an efficient algorithm for solving, say, the Travelling Salesman Problem and "inventing" some physical mechanism which does so. Theorems may very well be "discovered", but mathematical proofs are quite definitely invented; I'd argue that algorithms are closer to the latter.
Even for inventions, what is patented is _not_ a physical manifestation, but rather the plans, processes, or ideas involved--- the _algorithm_.
Otherwise the concept is meaningless; somebody building a different widget with the same plan wouldn't fall under the patent.
Further, the idea of algorithms as "discovery" is less convincing when you realize that many published algorithms (and theorems) turn out to be incorrect. If I have really "discovered" something in the Platonic realm, wouldn't it be correct? Proof gives only better certainty, not absolute certainty.
I don't think this is true, just from the nature of software vs. inventions of the types you mentioned. The investment to innovate things at the level of the Cotton Gin or Steamboat is huge in terms of both expertise and raw materials. In software, at the lowest level all you need is an interest in something (anything) related to some aspect of the machine, and with a very cheap investment (hardware and a possibly free OS/software), you can innovate. That's why Free Software has put so much power into the hands of "hobbyists".
Software will probably stabalize somewhat at some point, but it's not about a specific "product"; it's a whole new world of innovation. I think we're a good long ways away from that stabilization. Meanwhile (for the next 20 years at least, most likely), 20 years is far too long.
"Personaly, the guy could have released the source for free and his app and taht would have screwed your company up bigtime, he could have done this anonymously too"
This is an interesting idea. I know it wouldn't go over well with the GNU folks since there couldn't be GPL'd, but supposing someone set up a site for software that violates patents posted anonymously? Can you be arrested for trafficking in software that violates patents? Would you be forced to name names?
Well, I am not an IP lawyer, but I have some knowledge of the field, have worked with enough IP lawyers to have been granted 15 patents of my own (chemical, not software).
:-)], and how good a writer you and your lawyer are, regardless of the quality of your actual invention.
This is NOT a simple issue at all.
As far as the question of prior publication - if the idea has been published, that's an absolute bar to having a valid enforceable patent. If you publish in the US you have 1 year to file a patent IN THE US. You lose all patent rights outside the US immediately if you publish. This can take strange forms - for example a fellow once lost patent rights to a new kind of harvester because he had a picture of it published in a local newspaper a year before the application was filed. If you can point to prior publication you can have a patent re-examined and maybe thrown out. If you can prove the patentee knew about the prior publication and failed to disclose it to the Patent Office you might even get the patenteee brought up on patent fraud charges. (Yes, this does happen).
Personally I think patents are in general a good thing - they allow companies that invest a lot of money in R&D to recoup that investment. The great progress in modern pharmaceuticals we have today would not be possible without patent protection - bringing new drugs to market is a multibillion dollar investment. There are many other fields where patents rights ar an important incentive towards encouraging real progress.
They also encourage the open publication of a lot of technology that would be kept secret otherwise to protect a commercial advantage. This second concept is the real reason for the patent system in the first place - in exchange for open publication of a new technology, rights for exclusive use of the technology are granted. This in fact is giving inventors an incentive to 'Open Source' their technology for the good of all society. Otherwise people would keep everthing as trade secrets (encypted binary) that in fact might get lost on the demise of a company or individual.
The problem is that this is not working in the software industry. Exclusive rights are being given for non-inventions. Patents are supposed to have three qualities - novelty, utility and unobviousness. There needs to be some meat in the publication in exchange for that exclusive right to practice the invention. I think that the vast majority of software patents fail on the first and third points - novelty and unobviousness. At this point I think there are very few truly new ideas in software development - a new data transfer protocol that REALLY is novel is unlikely, and I certainly feel that any process of translating an algorithm is obvious - but certainly new algorithms are being developed BUT according to patent law you can't actually patent a mathematical algorithm. Unfortuantely the patent office doesn't agree with me (or others) on this issue, so legislation might be needed. IMHO it's really a work in progress, so complaints really are needed - the patent office IS subject to whining. Do write your Congressman.
The problem with patent legislation is that there are also international treaties involved that essentially bring the various signers into agreement on what is patentable.
For example there are many countries who don't like pharma patents, but might have signed in exchange for some other issue. If you throw out software patent they might say, OK we are not going to respect your drug patents.
Also - the respondee that mentioned that patent examiners just have a quota to grant is incorrect. Examiners have a quota to inspect, and their grant to reject ratio is also monitored. However patent examiners work in specific technical specialties - yuou may have one in polymer science and another in software technology - both might be expected to examine 5 patents a day, and grant 2. The 2 that get granted though are somewhat of a crap shoot. It may depend on the bitchiness of the examiner that day, whether or not the author is known to the examiner from previous work [let me tell you - after you get 5 patents in a certain field, that 6th one comes a little easier - the patent office has made You the authority now
The patent owner is not going to sue a hobbyist. Patent lawsuits typically cost $1 million to bring. This is not going to be a winning proposition. HOWEVER they might sue Red Hat Labs, or a corporate customer of Red Hat.
It is important to remember that a free market is not necessarily the optimum economic system for maximal societal benefit. A free market (sans regulation) leads to monopolies like Standard Oil and the trashing of the environment by corporations whose bottom line is not affected by the external diseconomies of the effects of their pollution. Many regulations/laws are in place as the result of recognition that a completely free market leads to a lot of societal problems for the simple reason that corporations are too individuals and act in their own self interest. Individual people are subject to certain laws and regulations for the overall good, as so should be corporations.
If you don't believe me, take a look at 19th century industrial history in the United States as an example of what 'free markets' can bring. I don't think that many people would want to return to the days of the Chicago Shirt Waist Factory.
Microsoft has on occasion gotten burnt by ignoring patents - didn't they loose a $100 million judgement to Stacker?
The fact is that if you ignore patents you can get burnt big time. Billions can be at stake. Kodak's instant cameras, and a Pennzoil lawsuit are two examples that ended up costing billions to the loser.