IBM and your division is doing consultancy on the use of smart card technology? And you give away the software, but people need your help to use it? Or do you use the software in the consultancy process?
BTW, it seems to me that your company does actually sell some hardware and software. You make it seem like you make 40 billion dollar without selling hardware or software.
Sony's own press-release offers much more information than the article on smartmoney.com. Just like it's predecessor (the SDR-3X), the SDR-4X offers a MemoryStick-slot to supply additional control programs.
Actually, an employer can sign you to any damn contract they like. The only thing that prevents them from doing so is that we have laws that we say supercede contract law - minimum wage being one of them.
Quite irrelevant, the part of the contract that goes against the law is null and void. It's just like that part of the contract was never signed. My point still stands.
So long as the software creators terms do not violate any of those type of laws, then he can declare the terms of the contract, and your only option is to accept it or not.
Usually that is true, although the government may regulate monopolies by altering contracts. There might be other limitations as well.
To argue otherwise is to suggest that because everybody has a "right" to not starve, I should be able to walk into a restaurant and demand they feed me for nothing.
No, the government should make sure you can feed yourself:
Everyone, as a member of society, has the right to social security and is entitled to realization, through national effort and international co-operation and in accordance with the organization and resources of each State, of the economic, social and cultural rights indispensable for his dignity and the free development of his personality. Universal Human Rights, Article 22.
(1) Everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control. Universal Human Rights, Article 25.
...so would you argue that the end user is under no moral obligation whatsoever to respect the terms with which a developer distributes her software?
I've reread my post several times to figure out why you would derive this from my post. I haven't got a clue. The fact that I believe that developers may not simply dictate the terms of the contract does not mean that nothing can be agreed upon. It just means that certain things cannot and should not be negotiable. This is certainly not limited to IP. I'm not intimitely familiar with the US law, so I'll give an example from dutch law:
A car wash put up a sign at the entrance that said: 'no liability'. After a car was damaged he tried to use the sign as a defense. The court ruled that the sign could not supercede the law and thus was null and void.
BTW, IMO the BSD license is merely a notice of the fact that the piece of software you get is free and you may thus not expect much from it. In dutch law there is the concept of 'Friendship-Service' where someone helps another for free at his request. There can be no liability in that case unless the helper does damage willfully.
If not, then you've just agreed with what I've been saying all along, that the end user is morally obligated to either accept the terms of a developer's license or not use that developer's software at all.
"Do you think that self-defense is wrong? If not, then you've just agreed with my claim that murder is perfectly ok." Not agreeing with one extreme standpoint doesn't mean that I agree with the opposite extreme. Your reasoning is a basic debating fallacy.
Oh, and the only thing that keeps a person in the US from being able to accept a contract to work for fifty cents an hour is the law. There is no moral reason why two parties could not enter into such a contract if they both wanted to. Is there?
Yes, I do think that is usually immoral (and that happens to be the reason why the law is as it is). People should be able to earn a decent income or be supported by the state. It seems that many feel this way as this is part of the universal human rights. Of course, I suspect that you are a libertarian and thus believe that a contract is always just. And when children in third world countries work 16 hours a day for a few cents, you'll probably blame it on the government, no?
Everything of value is helpless. Lucebert, Dutch poet and draughtsman
Perhaps I was a bit unclear, but the basic point still stands:
Protection of IP is an artificial monopoly. It is a right granted for the common good. Again, quid pro quo. You get the right to disallow everyone else to sell a copy of your creation, as an incentive to create. The cost to you is that one day your creation becomes public property.
Why would the law grant you extra rights, without you giving something back? Why should the judges+police do work just to make you rich? Doesn't that seem selfish to you?
My code has value to me - the value of the hundreds of hours of my finite life that I put into creating it. That is not value that can, or should, belong to anyone else - even mankind as a whole. The ideas used to create the code (even if they are my own ideas), however, can belong to mankind as whole (after the patent incentive) - but that does not give anyone any rights (moral or legal) to my code.
You can always keep it to yourself. If you do, none except you can profit from your creation (that seems to appeal to you very much which saddens me). But as soon as you sell it and use the law to make sure you will be repaid for your hard work (by getting money or more code through the GPL), you (automatically) accept the consequences of that. Your creation becomes part of our culture and the law makes sure that it will one day become free. This is very fortunate or we would still be paying the heirs of Shakespeare. And it is extremely fortunate that we thus may preserve works that the IP owner doesn't care to sell anymore (see Project Gutenberg). Doesn't this make you even a bit happy?
Well, Lawrence Lessig is much better at explaining these things. I fully agree with him. I don't want to become a slave of IP-owners. But I not only disagree with the RIAA/MPAA's quest for control over our lives, but also with the programmers that want to do the same (like the GPL programmers that want to disallow certain uses of software).
No, American AC in Paris claimed that the creator is omnipotent. He can declare the terms of the contract and your only option is to accept it or not.
I argue that the consumer has certain rights that cannot be taken away, regardless of the existence of a contract. This is comparable to the fact that an employer may not have you sign a contract that pays you less than the minimum wage.
Developers deserve the same moral consideration as anyone else. If they make something and give it to the world with certain conditions, it follows that the world should honor their conditions. And I'm sure you really don't mean to say that developers can go to hell, seeing as you're not attacking them.
I strongly disagree. Our societies (not the world*) grant certain rights to people (the right to call the police when someone takes something that you 'own' for instance). Those rights are not a given, in the absence of law+police they would be (at best) meaningless. Another society can give their citizens radically different rights (the indians didn't have land property and communist communes lacked private property).
The reason to grant certain rights is (hopefully) to better the entire society. Our societies have extended exclusive rights (an artificial monopoly) to IP-owners for a limited time based on the following assumptions:
Pro - Many people need a monetary incentive to create something/make it available.
Con - People need to be able to build on the work of others, teach, criticize and transform. If this is prohibited, less IP will be created (note that most if not all IP is not 100% original). - A monopoly will usually inflate prices. - If all IP becomes private property, power will gravitate to the few who can control/buy the rights to the IP. They will be able to prevent new innovations or push certain creations onto us (Britney Spears, N'Sync and the other crappy music we are supposed to like). - If people are able to live on one creation forever, they will probably less inclined to create.
The difference between 'real' property and IP is extremely important in this regard. IP can be infinitely duplicated and thus become a big asset to society. A good example is the difference between a single machine and the blueprints for it. The blueprints can be infinitely duplicated and can educate and enrich many, the machine only has value for the owner and will some day break down. The ultimate goal of the IP laws (and patents) in the US constitution is to extend the publicly available IP by coercing people/companies into creating it and making it available, using a monopoly period which is as short as possible. There is absolutely no right for IP owners to extend this period by themselves (using EULA's) or to limit the rights you 'buy' in other unlawful ways (disallowing you to write a negative review for instance).
So in short: No, those that create are not free to define the conditions under which they are willing to share. If they want to profit from the protection of their IP by law+police, they should also be willing to accept the limitations set forth by law. They can't expect the protection of law to come for free. Quid pro quo: One thing for another.
*Please don't use these meaningless words. Everytime I see a victim of violence complaining that the 'world' doesn't do anything, I wait for a crevice to open and consume the offender. I can't come up with any credible definition of the world were it is an entity that can make decisions and/or act. We haven't got a world government you know.
BSD advertising clause
on
Abusing the GPL?
·
· Score: 2, Flamebait
The advertising clause was removed from the BSD license three years ago. If you want to use the license you should use this template.
Besides, if you happen on a file with the old license that is copyrighted to the University of California, the advertising clause is null and void.
What good is an 'individual' capable only of ripping off someone else's work and expanding on it?
I suppose you wrote your own compiler, used no libraries and built your own hardware (not just assembled it), while happily reinventing all computer science, physics and other scientific discoveries that you would need.
We all built on the work of others, that is called progress. The idea behind copyright and patents is to give creators a _temporary_ solitary grant to 'own' the IP they come up with as an incentive to create things, while the inventions are released into the open after a certain period, leading to a large base of truly free stuff. If the grants would be eternal, we would become encumbered by them and ultimately be unable to progress futher. Big IP owners would just stifle all innovation and keep making us pay for the same recycled 'innovations'. This is already happening unfortunately.
It is clear that the current laws stifle innovation because the 'temporary' grant lasts far too long. But the GPL is no solution to this, it just prevents people from making a profit when they build on the work of others. The only problem it solves is to force the open sourcing of software, while ignoring the true battle: to fight for progress. Something that will not happen when people can't earn a paycheck with their inventions.
The BSD-license (and public domain) releases the source into the open immediatly, making it a stepstone for everyone, including commercial and BSD programmers. I've argued that this will probably mean more contributions to your code as well. I believe that open source software will never be able to fill every void (I can give plenty of examples), BSD is perfect for allowing open source and commercial software to coexist.
Time advances: facts accumulate; doubts arise. Faint glimpses of truth begin to appear, and shine more and more unto the perfect day. The highest intellects, like the tops of mountains, are the first to catch and to reflect the dawn. They are bright, while the level below is still in darkness. But soon the light, which at first illuminated only the loftiest eminences, descends on the plain, and penetrates to the deepest valley. First come hints, then fragments of systems, then defective systems, then complete and harmonious systems. The sound opinion, held for a time by one bold speculator, becomes the opinion of a small minority, of a strong minority, of a majority of mankind. Thus, the great progress goes on. Thomas Babington Macaulay (1800-1859), English historian.
But the layers of indirection added through the JVM will always make it slower, and never a language that will replace C++. Just as C++ will never replace C
[sarcasm]And C will never replace assembly. The abstractions are just to slow to write a kernel with. You really need all the speed you can get, that's why all the kernels around today are written in assembly.[/sarcasm]
Why don't you just replace 'never' with 'it will take longer than people think'. This is almost universally the case with hypes/new technologies, in the short term they fail to achieve the desired traction (and critics put them down), but in the long term they have far greater impact than expected. I have no doubt that garbage collection (gc) will make it into (serious) kernels one day, although no-one will believe me. Ten years ago nobody believed you if you told them that gc would be in a mainstream programming language. As computers get faster and we learn more about how to effectively use technologies like garbage collection, they will be used more and more for low-level programming.
And when speed is paramount, what is a Java programmer going to do?
He will follow this list until the problem is fixed:
1. Find the problem and try to fix the basic algorithm 2. Buy "Java Performance Tuning" and optimize 3. Buy a profiler and optimize 4. Find someone to write the routines in C and call them using JNI
It seems to me that C programmers have to go through steps 1-3 as well (unless they are born as expert programmers). It appears that Java programmers are worse off since they might get an application with both Java and C code, but I think that this actually an advantage. 95+% of their code will be fairly safe against buffer overflows, memory leaks and can be written quickly. The mean, lean code is neatly seperated and gets a dedicated bad-ass programmer. A coder that knows how to prevent leaks and other problems.
Compare this to a 100% C program of which 25-50% is written by less experienced programmers. Programmers who leak, let their buffers overflow and write slow C code. In fact, I work at a company that has very good programmers, but their programs still leak a lot of memory. This is a big problem for web applications, it's a bitch to retire instances every few hours. I just hope they migrate to Java soon;)
Ok, so I pay you before I know if the software is any good or will even be produced. I'm getting confused here, why would I want this instead of buying an off the shelf solution?
Heck, would I buy a book like this? How would I know if it's any good? Stephen King has such a good reputation that he could barely pull it off (and he is probably the best known writer). How could other writers do this? I didn't know I wanted the book 'Java Performance Tuning' before I read a review. I wouldn't have bought it before it was written, cause that would be very risky. Plus, I buy the book when I have a problem, it should be available to me quickly. And given the choice between paying the author to create a new 'vaporware' book in a year and getting a free good-enough version now, what option are most people going to take? The weaker book, so the better one is never written. Your plan just doesn't make sense, it would lead to mediocrity and a lack of choice.
Your plan could be applied to regular products just as well, but history has proven that payment before production only works for services (geared to a specific situation). And people/companies are always trying to get away from them to regular products (barbers gave away to philishaves, performing artists gave away to LP's/CD's/videoclips, specialized applications gave away to spreadsheets, ERP systems, CRM systems, etc). The ideal situation is where the consumer has all the information to make a choice. He knows what he gets at what price and there are few risks. That is best achieved by having a number of products that are already available. A pre-paid service will always be a second choice, people/companies only go for it if there is no acceptable choice available.
What kind of quality assurance is that It isn't. Read the licence. It's called the GNU GPL. I'm surprised you haven't come across it before. There is no "quality assurance" If you need someone to hold your hand, or you are using Linux for production purposes, go and talk to Redhat or another distributor who will provide the quality assurance you seem to want. They test all their kernels with test suites and simulated production workloads.
Open source software is no excuse for a lack of QA. There are people who want to use Linux for serious work, there should be a -stable that is truly stable. That's the point of concurrently having a 2.4 and a 2.5 release. Red Hat's job is to create a distro from various components and test them together. They can't be expected to transform an unstable kernel into a stable one. Their job is hard enough as it is (keeping track of hunderds of packages is not easy).
Of course there is always a compromise between stability and (fast) progress, the *BSD's are far more focussed on stability. IMHO Linux is focussing far too much on progress, creating great instability. The newest developements belong in the unstable releases for the adventurous to use and test (new VM's for instance). Kernels in the 2.4.x tree should be well-tested before they are released into the open.
What I would really like is a topic where all these Linux 2.2.0.2.3.4.1.3 and MacOS X 10.1.3 updates are posted so I can skip them. I like to read about a substantially new version of an OS, but why should we be bothered with these minor bug-fixes. The people who use the respective OS's will learn about the update from their own sources (I learned about this update from 5 sites before Slashdot told me about it) and the others couldn't care less about minor fixes that don't affect them.
Besides, the most important reason to read slashdot is for the comments and these topics get the worst comments of all. It's mostly trolling (Mac/Linux/BSD rulez/sucks) and uninteresting trivia (did you know that this update fixes DVD on external monitors?). Get me some serious technology news that illicits comments about the problems of EPIC or an ethical story about censorship.
These are precendents that can make you win the case if you can keep the DMCA out of this:
[...] Over the years, the use of reverse engineering of software has been extremely controversial. In fact, there have been two major lawsuits concerning products that were written with the help of decompilation.
The first major lawsuit was Atari vs. Nintendo. In this case, Atari Corporation reverse engineered Nintendo's game cartridges in order to create alternative games that could run on Nintendo's system. The judge in this case found that decompilation was acceptable to figure out unprotected elements of the software. However, Atari lost the case because they used fraud to obtain Nintendo's source code from the Copyright Office [7].
The next major case, Sega vs. Accolade, created a much stronger precedent because there were no fraud issues involved. In a similar manner to Atari, Accolade reverse engineered Sega's Genesis technology to discover how to make games for their system. Accolade then created a book with the relevant (and non-protectable) elements of Sega's technology, and passed the book on to their developers. These developers created a new game, Ishido, to compete with Sega's games.
In deciding the case, the court looked at many factors (including public policy concerns). In the end, the judge decided that reverse engineering software for the sole purpose of creating a compatible package is an acceptable use (under the "fair use" doctrine). In addition, the appeals court stated: "[i]f disassembly of copyrighted object code is per se an unfair use, the owner of the copyright gains a de facto monopoly over the functional aspects of his work - aspects that were expressly denied copyright protection by congress" [1]. Thus, the court decided to adopt the policy encouraging competition (as opposed to IP protection) in the software industry.
This is blatantly stolen from Stanford. The guy should be able to argue that he is reverse engineering for fair use, especially since he's advertising programming the system and backing up savegames & games. Of course, the DMCA is created to take away your rights to fair use by disallowing any decrypting.
I'd call upon Nintendo to prove that the device doesn't merely reverse-engineer for fair use, but it truly circumvents an encryption device. This means that they have to prove that such an encryption device exists. But of course, IANAL. You should really get a lawyer to look at this and write a letter back. And you're in big sh*t if you find out that Nintendo uses an XOR-encryption scheme for it's code;)
I don't know it it really is better on a Mac. I can't see how it would be.
MacOS 8/9: Applications install 99% of their stuff in their application's folder. Sometimes they install an extension in a special folder, I use an utility to manage these files. This allows me to identify them, group the files and delete them. This works great for a power-user. A reinstall of the OS can be done without many problems, most software needn't be reinstalled (no registry, copy the needed extensions with the utility or by hand). If you copy the preference files from the preference folder, you don't have to set up applications again (I do and sometimes go through it and delete the obvious cruft). I do this at most once a year and it works extremely well, bloat is very small (~50mb per year, mostly because I'm lazy and don't clean up everything perfectly). But the OS allows you to do so almost perfectly, Windows hides everything so managing installed software is extremely hard.
OS X: Some applications are shipped as self-contained packages, only preference files are left in the system (no (de-)installer needed). Others do install cruft, I'm still figuring out how to manage it (it's still a bit of a chaos, developers are still learning about the workings of the OS). I suspect this mess will improve in time.
Windows: My experiences with Win9x have been horrible, reinstalls were regular (but I do install a lot of software) and very painful (having to reinstall almost all my software). Win2000 is doing well for me, it's fairly stable and doesn't start to lose features (Win98->where did my OpenGL go?). So I needn't reinstall, which would probably be just as painful as reinstalling Win9x.
I've never used assertions, but they seem a great complement to unit testing. Unit testing allows you to write code to test your functions and easily see if something breaks, the major problem is that they lack an easy way to look inside objects to keep an eye on internal consistency. Assertions can be great to catch those silly little boundary mistakes.
A good unit testing framework for Java is JUnit, they are available for other languages as well.
BTW, you can create your own assertions with Log4J, so even JDK 1.1/1.2/1.3 users can use them:
if(logger.isDebugEnabled())
if (bla>10)
logger.warn("bla>10, bla=" + bla);
This uses almost no CPU-time if debugging is disabled. Log4J is a very good logging package, it surely beats System.out.println, check it out!
The free Quicktime Streaming Server 4 has been released today and its open source companion the Darwin Streaming Server (has the same features, but it does run on Linux, BSD and Windows). It already supports Mpeg-4:
"MPEG-4 Support: now you can serve ISO-compliant hinted MPEG-4 files to any ISO-compliant MPEG-4 client, including any MPEG-4 enabled device that supports playback of MPEG-4 streams over IP. You can serve on-demand or live MPEG-4 streams, and reflect playlists of MPEG-4 files."
I'll bet they tried to mention MPEG-4 as many times as possible.
You can now also stream MP3's with it, set up your own radio station! The streaming uses the standard Icecast streaming format so any MP3-player that supports streaming should work.
The worst thing that can happen is that some freeloading company doesn't use my code.
No, the worst thing that can happen is that companies do not choose to use and enchance your product, because of which it doesn't gain momentum and fades away. This means that your work is wasted.
The best thing that can happen with BSD-licensed software is that everyone gives up their own commercial products and decides to built upon the open source code. A good example is Apache, companies have made great contributions to it (I think that most programmers who work on it are paid by corporations) and most competitors have given up. This is extremely unlikely with GPL'ed software.
You can imagine that companies prefer the BSD license all you like, it doesn't affect the fact that most of the worthwhile software is currently being developed under the GPL.
Apache, XFree, Xerces, Xalan, Tomcat, Python, PHP and Perl are all licensed under a non-copyleft license.
But I guess that this software isn't worthwhile.
It's also realistic. Had a look at the headlines lately? *cough* Enron *cough*
One example of a company that went crazy and got punished for it (the story could turn into a Disney movie, except for the fact that the employees got punished and those responsible got a lot of money). I counter your one example and up you one with Apple and Sun, who both put a great deal of effort into open source.
But it there a point to this? We could talk all day about the morality of corporations. In the end what counts is whether GPL is necessary to protect software and/or to get enough contributions. I can give many examples of BSD-like licensed software that get great support by companies and have never been abused. I'm quite sure that you'll only be able to find a few examples where the open version floundered, while a hijacked closed version flourished or examples of abuse.
Why do you get so hung up on the few bad examples? Suppose that Newton would have kept his discoveries secret because his calculations might be used for evil (like aiming cannons) or that the automobile would not have been produced out of fear that it would hurt people or flying would have been banned because planes might be used for fighting. We would have still been in the middle ages if everyone thought like you.
Engineers and scientists are and should not feel responsible for all the uses of their inventions. They should try to do good and focus on promoting and working on the uses that benefit mankind. But IMO they are not responsible for building systems that never can be used for evil, for that automatically means that many good uses become impossible.
Hmm. Ever wonder why they didn't choose to get behind BSD? Maybe because they don't want their hard work going to finance Microsoft any more than I do. I could be wrong, of course. Maybe its just because linux is a better system.
Do you mean the BSD-license or *BSD (the OS's)? I'll assume the latter.
Linux has become a hype and has gained some serious momentum, these companies want to profit from that. Other companies (Yahoo, Apple, many ISP's) have chosen *BSD because it suits them better and they don't have to sell it or it is a minor feature of their system (in marketing that is).
Maybe because they don't want their hard work going to finance Microsoft any more than I do.
AFAIK just about the only BSD-software that went into Windows is an early TCP/IP-stack (now fully replaced) and some related utilities (ftp.exe, etc). Without it, the Internet may not have grown that fast and Windows-users may have been shut out for a while. So I don't see anything wrong with it.
But of course, I don't see anything wrong with commercial companies using freely available code as long as they behave morally and lawfully. We should encourage the use of standard, well-written software, just like we should encourage open standards like DDR DRAM. This will benefit the entire market, small and large companies alike (both small and large companies are making money of Apache for instance, it creates an open playing field). If companies do not behave, they should be punished by law or by the power of money, we should not buy their products.
MS is an example of the failure of American law in allowing a monopoly to abuse it's powers, not the failure of open source under a non-copyleft license.
I don't see that happening though, so I will guarantee that at least my contribution will always lead to more openness.
GPL cannot guarantee it, your code can still fade away. A strong, supported BSD-product will also lead to more and more openness as no-one will have the ability to force it into closedness (and thus taking away many advantages). An example is Apache, do you believe it will be closed up again ever?
My crystal ball tells me that eventually, the only role for commercial software will be specialized applications. Everything else will be open source.
Your crystal ball seems to be manipulated by Sauron. Tell me how open source can ever prevail for games that have a 1-year life-span and must be created in a short period by a full-time team as not to fall behind in graphics?
The open source community already has a hard time creating a decent browser and word processor, software that interests millions. How will open source ever be able to cater to fairly small, non-programmer communities? An example that was given by somebody else on/. was a fishing simulator.
BTW, I assume that specialized apps mean applications adapted or designed for one entity (aka custom-made products). You might also define it as everything outside the OS, so it's a bit unclear what you really mean.
Re:A visionary's gutfeel regarding 64-bit widespre
on
Inside the Itanium
·
· Score: 1
I know that he said that, but perhaps he didn't see the the PPC roadmap on Motorolas website recently. The G5 is 64-bit and is 32-bit backwards compatable.
That's not true. There will be a 64-bit version with 32-bit backwards compatibility and a 32-bit version (just like with the PowerPC 620 vs 601). That is what it says on the roadmap. Since programmers are still transitioning to OS X, I wouldn't expect the switch soon, at least not until the high-end users start complaining about a lack of RAM. Don't forget, 64-bit will hardly make a difference in speed for most uses, especially since the G4 already has a 128-bit processing unit (Altivec). Any application that would benefit greatly from 64-bit instructions already supports Altivec. On the other hand, 64-bit instructions/data means moving twice as much data, even for 32-bit instructions. Given that modern CPU's are for a good part limited by bandwith issues, this may make a 64-bit CPU slower than the 32-bit version.
The larger registers and caches will increase the size of the die and thus the cost of the chip. So I'm not looking forward to it yet, my next computer will probably not (need to) move beyond 2GB of RAM, so I can wait.
Umm, I don't have to depend on a company's goodwill
You do if you want them to contribute to open source. You start out at the proposition that people use your open source product, I take the whole picture into account:
total contribution by companies = use of open source*contribution per project
BSD maximizes the first, which will probably maximize the total figure (as I have pointed out that it has great advantages to contribute). You just focus on the second figure, but 0.0001*50 is still not very much.
Companies exist to make money, not to be altruistic. If they that think they can get away with something, they generally will. I have no qualms with taking the decision to use my code in a way I don't like out of their hands.
That's a very cynical view. Most companies do actually worry about more than pure profits. Did you check out the list of companies that double their employees donations for the EFF? It's long and heartwarming, snif.
I really don't see that many 'abuses' of BSD. Most companies who use BSD (or similarly) licensed products do contribute (or have even created the product in the first place), so why would you need to force them?
With both IBM and SUN recently getting behind linux in a big way, it looks lik the GPL is, in fact, becoming more pervasive. These are not small time players - they know what they are getting into.
They like it as a server-OS and certainly did not pick it because of the GPL, just about all the other products they support or open sourced themselves are non-copyleft.
What you are claiming as a legitimate use, I see as an abuse.
Sure, the examples I gave would truly have hurt the community (not). But of course, usage of IP is an abuse if it doesn't follow your rules. That's the same logic in EULA's, DVD regions and copy-protection, protected CD's, protected ebooks, the DMCA, etc. It's the thinking (and laws that support it) that I want to abolish, the abuse of power because you can create/control and others cannot. You focus so heavily on enforcing your rules, why are you unwilling to discuss the validity of the rules themselves?
It seems to me that you know you have lost this discussion and are trying to weazel out of the question at hand: Is copyleft truly necessary to force companies to contribute? I claim this is not so and thus it is best to try and pick a license that maximizes use and reduces adversity and legal issues. The BSD-license seems to do this well.
Of course, GPL has it's place for certain uses, but BSD is my first choice. Why? Because I want open source to achieve it's true potential, a strong force that goes hand in hand with commercial software. Open source will be able to supply some of the software we need, while commercial software will fill the niches that cannot be filled by open source.
*I* don't get any extra choice. Nobody is ever going to use my own code as a starting point to enforce a software patent, possibly against me.
You do get an extra choice, even if you aren't willing to consider it. Others may. I don't have a problem with sufficiently advanced patents. They do have their place (but should be strictly limited).
But your crusade against patents should not be fought over open source code. Patent issues are not specific to open source, they can/do harm commercial software just as much. The proper way to fight them is through politics (write your congressman) or through law (donate to the EFF).
I don't care what any company chooses to do or not to do. If they want to use my code, they *are* going to contribute back - no maybes about it.
Then you shouldn't use the GPL but a more restrictive license ("If you use this software, all your base belong to us"). GPL'ed code can be used without giving back as well (by not distributing the code). Companies are not forced to make changes and thus may leech. And I wonder why you don't mind when good code isn't used because the license doesn't allow perfectly valid uses. This a big shame. If more companies will use open source they will lower their costs, thus learn about the advantages of sharing their IP and start contributing.
Your simple-minded argument doesn't hold up very well. If you do have to depend on companies' goodwill and commitment for them to contribute to open source, why not go the whole way and actually threat them like adults. Give them freedom to make their own decisions. If you truly believe in the merits of open source, you must believe that it will be beneficial to companies and they will thus support it.
I trust that companies will give code back. It is often in their best interest (as I already pointed out). GPL will just ensure that your code isn't used, and thus you won't get anything back from them.
The more pervasive the GPL becomes, the less likely this will be a problem.
I believe that GPL will (hopefully and probably) not become that pervasive among companies (except for Linux). Instead it may give open source a bad name when companies get bitten by the GPL license. Is it not better to let money do the talking, instead of lawyers?
1. I make changes to the software that are specific to my business. It can't be open because that would expose my secrets.
You don't have to release the source unless you intend on distributing your program.
Suppose that Shell wants to distribute software to gas stations that link into it's own systems for greater efficiency (with JNI for instance). This specific code may not be open sourced because provisions for future products are put into the code before they are available. Those gas stations are independent and the software is thus 'distributed'.
2. I have created my super-duper commercial package and want to link it to an open source web-server. I'm willing to open source every addition to the web-server I make to suit my needs, but the rest of the software is my property.
If the web server is LGPL, no problem. If it's GPL, find another web server.
This is not a denial of my point. Argueing that GPL software cannot be used when I claim that some legitimate and important uses are not possible with GPL merely proves my point. The LGPL is far different from the GPL of course, let's keep focus.
3. I want to add video codec X, it's not mine. My clients are willing to pay for it though. What's the problem with making a closed version that includes the codec?
I think this one depends. IANAL, but I don't think that your licensing a codec and including it in some GPL software for someone would be considered distribution as long as they weren't otherwise distributing it. For instance, if I worked for a company and we had standardized on some GPL video editing software, what would be the problem with my doing anything I wanted with that software as long as it didn't leave the company?
Every company would have to license the codec and integrate it seperately??? Suppose we are talking about 20 codecs by 20 different companies, many of which need to be adapted to work with the software. A central entity can make this happen and ask a fee for their hard work (and to cover the license costs). Single companies may not have the resources to expend this much for just themselves and they may not have the clout to get the codec-providers to adapt their software (small companies are especially hurt by this, making it harder for them to compete with large corporations). So either this will be impossible or it will cost far more to accomplice (a bulk license for the codecs will usually be cheaper).
This points out a big problem with the GPL, BTW. It discriminates against a network of small companies in favor of large corporations. The latter can share an altered GPL-program without opening the code (for which there may be good reasons), while the network cannot. They are at a disadvantage. So choosing the GPL unwillingly promotes mega-corporations.
Unless, of course, they tie it in with some patented or otherwise encumbered code that is not legal to duplicate.
Which couldn't be used in the GPL-version anyway. So you still get an extra choice that you would not have had with the GPL-software. Nowhere is anything taken away from the free version by this!
BTW, code can't be patented, only an idea or algorithm. Of course the real answer to the inability to use an trivially simple idea or algorithm in open source (or closed source by another company) is to limit patents.
Human nature is what's wrong. I don't trust my fellow man enough to assume that he will be as forthcoming with his additions to my code as I was with the original code. Therefore, if he wants to have the advantages of my code, he is going to contribute his advantages back to me.
I'm convinced that a very large percentage of (usable) open source code has been written by cooperations or in the boss' time. Sun wrote a lot of code for the Jakarta project. IBM and Lotus wrote a lot of code for the Apache XML Project. A lot of companies contributed to Apache itself. Apple is paying a FreeBSD-developer and makes some nice open source software. The list goes on and on.
Why are you so paranoid when many companies are already proving that they are willing to contribute? Of course this usually requires that they can actually use their own contributions for their uses and a non-copyleft license is thus very important.
I don't care what they use my code for, as long as they give back.
I trust that companies will give code back. It is often in their best interest (as I already pointed out). GPL will just ensure that your code isn't used, and thus you won't get anything back from them.
I guess I'm just pragmatic.
What 'legitimate and important uses' require that you can't give back your source? I can't think of one.
1. I make changes to the software that are specific to my business. It can't be open because that would expose my secrets.
2. I have created my super-duper commercial package and want to link it to an open source web-server. I'm willing to open source every addition to the web-server I make to suit my needs, but the rest of the software is my property.
3. I want to add video codec X, it's not mine. My clients are willing to pay for it though. What's the problem with making a closed version that includes the codec? I'm willing to contribute every change that is 100% mine to give away. A good example of this kind of thing is Darwin, Apple can't give away the driver code for ATI or NVidia cards, so that code is only distributed as binary.
These uses sound perfectly legitimate and important to me. They make or break the ability to provide useful features for clients with open source software. These uses do not take anything away from the original code, not do they cause a company to withhold code that they would otherwise open up. So why are they not allowed under the GPL?
Do people write code in order to write good code and improve the state of computing, or do they do it in order to coerce other programmers into helping along?
The GPL guarantees both, while BSD only guarantees one.
Not necessarily. Many companies are afraid to use GPL-products, they want to have the option to ship a version that is linked to their own software or add a feature that is licensed from a third party or whatever. These are perfectly valid uses that are disallowed by the GPL (the second use may clash with LGPL).
These things can be done with BSD, so companies can actually use it. Any changes they make that are useful to others will usually be put in the main tree so:
1. There are fewer diffs between the open version and their product->easier merges
2. Their code is tested and audited and expanded and debugged. This is a big reason for using open source.
3. Why not?
4. It will be good PR.
5. It may actually make them feel better (especially the programmers).
BSD-licensed code will thus probably get you more good code. If someone does create a closed version with features you like, open source programmers can just copy the features. They haven't taken away anything from you, but may have filled a need that open source developers did not fill. What's wrong with more options?
I don't really understand why people get so upset when there might be a chance that someone uses the code he gives away for free for things he doesn't like. This is the same paranoia that the RIAA/MPAA exhibit: "if we don't put some überprotection on our stuff we will get screwed". They don't care if many legitimate and important uses are no longer possible, like making back-ups or listening to music on my computer. Strangely enough the Slashdot crowd is enraged over this, we tell them to be more trusting and to find ways to make money without creating a policestate. On the other hand it is okay with many of us if programmers make legitimate and important uses impossible with the license they choose for their IP.
Options are sometimes considered to be a good thing.
You are clearly not a minimalist. Many see the advantages of having one clear way to do something, instead of many options, all with their advantages, disadvantages and different abilities. A single, consistent design gives you clean and logical interfaces, which are fairly easy to use. Especially for authentication it may be better to have a slighty less capable interface, instead of an interface that people can't comprehend and thus misuse.
Of course you may sometimes need something more potent, this should be achieved by extending the basic design ('advanced options') or by adding a separate more complex interface. This should be a superset of the simpler interface. Users can thus be pointed to the simple interface by default and can migrate to the more complex one if their application becomes more complex.
I can't really see why they have the need to come out with ever faster graphics boards. I can play games like Deus Ex and CounterStrike fine already with my GeForce3 (in fact, it is already a little overkill). You really have to start asking the question "Are they ever going to make a game detailed enough to take advantage of this capacity?"
No, of course they are never going to make a game that is as pretty as Final Fantasy or Shrek.
This is a typical case of duh. Remember your post when Doom 3 or Unreal 2 comes out and the Geforce 3 can't reach the highest resolutions at decent speed.
Slashdot Mirror
IBM and your division is doing consultancy on the use of smart card technology? And you give away the software, but people need your help to use it? Or do you use the software in the consultancy process?
BTW, it seems to me that your company does actually sell some hardware and software. You make it seem like you make 40 billion dollar without selling hardware or software.
PS, what bonus points?
Sony's own press-release offers much more information than the article on smartmoney.com. Just like it's predecessor (the SDR-3X), the SDR-4X offers a MemoryStick-slot to supply additional control programs.
Press-release
Actually, an employer can sign you to any damn contract they like. The only thing that prevents them from doing so is that we have laws that we say supercede contract law - minimum wage being one of them.
Quite irrelevant, the part of the contract that goes against the law is null and void. It's just like that part of the contract was never signed. My point still stands.
So long as the software creators terms do not violate any of those type of laws, then he can declare the terms of the contract, and your only option is to accept it or not.
Usually that is true, although the government may regulate monopolies by altering contracts. There might be other limitations as well.
To argue otherwise is to suggest that because everybody has a "right" to not starve, I should be able to walk into a restaurant and demand they feed me for nothing.
No, the government should make sure you can feed yourself:
Everyone, as a member of society, has the right to social security and is entitled to realization, through national effort and international co-operation and in accordance with the organization and resources of each State, of the economic, social and cultural rights indispensable for his dignity and the free development of his personality.
Universal Human Rights, Article 22.
(1) Everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control.
Universal Human Rights, Article 25.
I've reread my post several times to figure out why you would derive this from my post. I haven't got a clue. The fact that I believe that developers may not simply dictate the terms of the contract does not mean that nothing can be agreed upon. It just means that certain things cannot and should not be negotiable. This is certainly not limited to IP. I'm not intimitely familiar with the US law, so I'll give an example from dutch law:
A car wash put up a sign at the entrance that said: 'no liability'. After a car was damaged he tried to use the sign as a defense. The court ruled that the sign could not supercede the law and thus was null and void.
BTW, IMO the BSD license is merely a notice of the fact that the piece of software you get is free and you may thus not expect much from it. In dutch law there is the concept of 'Friendship-Service' where someone helps another for free at his request. There can be no liability in that case unless the helper does damage willfully.
If not, then you've just agreed with what I've been saying all along, that the end user is morally obligated to either accept the terms of a developer's license or not use that developer's software at all.
"Do you think that self-defense is wrong? If not, then you've just agreed with my claim that murder is perfectly ok." Not agreeing with one extreme standpoint doesn't mean that I agree with the opposite extreme. Your reasoning is a basic debating fallacy.
Oh, and the only thing that keeps a person in the US from being able to accept a contract to work for fifty cents an hour is the law. There is no moral reason why two parties could not enter into such a contract if they both wanted to. Is there?
Yes, I do think that is usually immoral (and that happens to be the reason why the law is as it is). People should be able to earn a decent income or be supported by the state. It seems that many feel this way as this is part of the universal human rights. Of course, I suspect that you are a libertarian and thus believe that a contract is always just. And when children in third world countries work 16 hours a day for a few cents, you'll probably blame it on the government, no?
Everything of value is helpless.
Lucebert, Dutch poet and draughtsman
Perhaps I was a bit unclear, but the basic point still stands:
Protection of IP is an artificial monopoly. It is a right granted for the common good. Again, quid pro quo. You get the right to disallow everyone else to sell a copy of your creation, as an incentive to create. The cost to you is that one day your creation becomes public property.
Why would the law grant you extra rights, without you giving something back? Why should the judges+police do work just to make you rich? Doesn't that seem selfish to you?
My code has value to me - the value of the hundreds of hours of my finite life that I put into creating it. That is not value that can, or should, belong to anyone else - even mankind as a whole. The ideas used to create the code (even if they are my own ideas), however, can belong to mankind as whole (after the patent incentive) - but that does not give anyone any rights (moral or legal) to my code.
You can always keep it to yourself. If you do, none except you can profit from your creation (that seems to appeal to you very much which saddens me). But as soon as you sell it and use the law to make sure you will be repaid for your hard work (by getting money or more code through the GPL), you (automatically) accept the consequences of that. Your creation becomes part of our culture and the law makes sure that it will one day become free. This is very fortunate or we would still be paying the heirs of Shakespeare. And it is extremely fortunate that we thus may preserve works that the IP owner doesn't care to sell anymore (see Project Gutenberg). Doesn't this make you even a bit happy?
Well, Lawrence Lessig is much better at explaining these things. I fully agree with him. I don't want to become a slave of IP-owners. But I not only disagree with the RIAA/MPAA's quest for control over our lives, but also with the programmers that want to do the same (like the GPL programmers that want to disallow certain uses of software).
No, American AC in Paris claimed that the creator is omnipotent. He can declare the terms of the contract and your only option is to accept it or not.
I argue that the consumer has certain rights that cannot be taken away, regardless of the existence of a contract. This is comparable to the fact that an employer may not have you sign a contract that pays you less than the minimum wage.
Developers deserve the same moral consideration as anyone else. If they make something and give it to the world with certain conditions, it follows that the world should honor their conditions. And I'm sure you really don't mean to say that developers can go to hell, seeing as you're not attacking them.
I strongly disagree. Our societies (not the world*) grant certain rights to people (the right to call the police when someone takes something that you 'own' for instance). Those rights are not a given, in the absence of law+police they would be (at best) meaningless. Another society can give their citizens radically different rights (the indians didn't have land property and communist communes lacked private property).
The reason to grant certain rights is (hopefully) to better the entire society. Our societies have extended exclusive rights (an artificial monopoly) to IP-owners for a limited time based on the following assumptions:
Pro
- Many people need a monetary incentive to create something/make it available.
Con
- People need to be able to build on the work of others, teach, criticize and transform. If this is prohibited, less IP will be created (note that most if not all IP is not 100% original).
- A monopoly will usually inflate prices.
- If all IP becomes private property, power will gravitate to the few who can control/buy the rights to the IP. They will be able to prevent new innovations or push certain creations onto us (Britney Spears, N'Sync and the other crappy music we are supposed to like).
- If people are able to live on one creation forever, they will probably less inclined to create.
The difference between 'real' property and IP is extremely important in this regard. IP can be infinitely duplicated and thus become a big asset to society. A good example is the difference between a single machine and the blueprints for it. The blueprints can be infinitely duplicated and can educate and enrich many, the machine only has value for the owner and will some day break down. The ultimate goal of the IP laws (and patents) in the US constitution is to extend the publicly available IP by coercing people/companies into creating it and making it available, using a monopoly period which is as short as possible. There is absolutely no right for IP owners to extend this period by themselves (using EULA's) or to limit the rights you 'buy' in other unlawful ways (disallowing you to write a negative review for instance).
So in short: No, those that create are not free to define the conditions under which they are willing to share. If they want to profit from the protection of their IP by law+police, they should also be willing to accept the limitations set forth by law. They can't expect the protection of law to come for free. Quid pro quo: One thing for another.
*Please don't use these meaningless words. Everytime I see a victim of violence complaining that the 'world' doesn't do anything, I wait for a crevice to open and consume the offender. I can't come up with any credible definition of the world were it is an entity that can make decisions and/or act. We haven't got a world government you know.
The advertising clause was removed from the BSD license three years ago. If you want to use the license you should use this template.
Besides, if you happen on a file with the old license that is copyrighted to the University of California, the advertising clause is null and void.
What good is an 'individual' capable only of ripping off someone else's work and expanding on it?
I suppose you wrote your own compiler, used no libraries and built your own hardware (not just assembled it), while happily reinventing all computer science, physics and other scientific discoveries that you would need.
We all built on the work of others, that is called progress. The idea behind copyright and patents is to give creators a _temporary_ solitary grant to 'own' the IP they come up with as an incentive to create things, while the inventions are released into the open after a certain period, leading to a large base of truly free stuff. If the grants would be eternal, we would become encumbered by them and ultimately be unable to progress futher. Big IP owners would just stifle all innovation and keep making us pay for the same recycled 'innovations'. This is already happening unfortunately.
It is clear that the current laws stifle innovation because the 'temporary' grant lasts far too long. But the GPL is no solution to this, it just prevents people from making a profit when they build on the work of others. The only problem it solves is to force the open sourcing of software, while ignoring the true battle: to fight for progress. Something that will not happen when people can't earn a paycheck with their inventions.
The BSD-license (and public domain) releases the source into the open immediatly, making it a stepstone for everyone, including commercial and BSD programmers. I've argued that this will probably mean more contributions to your code as well. I believe that open source software will never be able to fill every void (I can give plenty of examples), BSD is perfect for allowing open source and commercial software to coexist.
Time advances: facts accumulate; doubts arise. Faint glimpses of truth begin to appear, and shine more and more unto the perfect day. The highest intellects, like the tops of mountains, are the first to catch and to reflect the dawn. They are bright, while the level below is still in darkness. But soon the light, which at first illuminated only the loftiest eminences, descends on the plain, and penetrates to the deepest valley. First come hints, then fragments of systems, then defective systems, then complete and harmonious systems. The sound opinion, held for a time by one bold speculator, becomes the opinion of a small minority, of a strong minority, of a majority of mankind. Thus, the great progress goes on.
Thomas Babington Macaulay (1800-1859), English historian.
But the layers of indirection added through the JVM will always make it slower, and never a language that will replace C++. Just as C++ will never replace C
;)
[sarcasm]And C will never replace assembly. The abstractions are just to slow to write a kernel with. You really need all the speed you can get, that's why all the kernels around today are written in assembly.[/sarcasm]
Why don't you just replace 'never' with 'it will take longer than people think'. This is almost universally the case with hypes/new technologies, in the short term they fail to achieve the desired traction (and critics put them down), but in the long term they have far greater impact than expected. I have no doubt that garbage collection (gc) will make it into (serious) kernels one day, although no-one will believe me. Ten years ago nobody believed you if you told them that gc would be in a mainstream programming language. As computers get faster and we learn more about how to effectively use technologies like garbage collection, they will be used more and more for low-level programming.
And when speed is paramount, what is a Java programmer going to do?
He will follow this list until the problem is fixed:
1. Find the problem and try to fix the basic algorithm
2. Buy "Java Performance Tuning" and optimize
3. Buy a profiler and optimize
4. Find someone to write the routines in C and call them using JNI
It seems to me that C programmers have to go through steps 1-3 as well (unless they are born as expert programmers). It appears that Java programmers are worse off since they might get an application with both Java and C code, but I think that this actually an advantage. 95+% of their code will be fairly safe against buffer overflows, memory leaks and can be written quickly. The mean, lean code is neatly seperated and gets a dedicated bad-ass programmer. A coder that knows how to prevent leaks and other problems.
Compare this to a 100% C program of which 25-50% is written by less experienced programmers. Programmers who leak, let their buffers overflow and write slow C code. In fact, I work at a company that has very good programmers, but their programs still leak a lot of memory. This is a big problem for web applications, it's a bitch to retire instances every few hours. I just hope they migrate to Java soon
Ok, so I pay you before I know if the software is any good or will even be produced. I'm getting confused here, why would I want this instead of buying an off the shelf solution?
Heck, would I buy a book like this? How would I know if it's any good? Stephen King has such a good reputation that he could barely pull it off (and he is probably the best known writer). How could other writers do this? I didn't know I wanted the book 'Java Performance Tuning' before I read a review. I wouldn't have bought it before it was written, cause that would be very risky. Plus, I buy the book when I have a problem, it should be available to me quickly. And given the choice between paying the author to create a new 'vaporware' book in a year and getting a free good-enough version now, what option are most people going to take? The weaker book, so the better one is never written. Your plan just doesn't make sense, it would lead to mediocrity and a lack of choice.
Your plan could be applied to regular products just as well, but history has proven that payment before production only works for services (geared to a specific situation). And people/companies are always trying to get away from them to regular products (barbers gave away to philishaves, performing artists gave away to LP's/CD's/videoclips, specialized applications gave away to spreadsheets, ERP systems, CRM systems, etc). The ideal situation is where the consumer has all the information to make a choice. He knows what he gets at what price and there are few risks. That is best achieved by having a number of products that are already available. A pre-paid service will always be a second choice, people/companies only go for it if there is no acceptable choice available.
What kind of quality assurance is that
It isn't. Read the licence. It's called the GNU GPL. I'm surprised you haven't come across it before. There is no "quality assurance" If you need someone to hold your hand, or you are using Linux for production purposes, go and talk to Redhat or another distributor who will provide the quality assurance you seem to want. They test all their kernels with test suites and simulated production workloads.
Open source software is no excuse for a lack of QA. There are people who want to use Linux for serious work, there should be a -stable that is truly stable. That's the point of concurrently having a 2.4 and a 2.5 release. Red Hat's job is to create a distro from various components and test them together. They can't be expected to transform an unstable kernel into a stable one. Their job is hard enough as it is (keeping track of hunderds of packages is not easy).
Of course there is always a compromise between stability and (fast) progress, the *BSD's are far more focussed on stability. IMHO Linux is focussing far too much on progress, creating great instability. The newest developements belong in the unstable releases for the adventurous to use and test (new VM's for instance). Kernels in the 2.4.x tree should be well-tested before they are released into the open.
What I would really like is a topic where all these Linux 2.2.0.2.3.4.1.3 and MacOS X 10.1.3 updates are posted so I can skip them. I like to read about a substantially new version of an OS, but why should we be bothered with these minor bug-fixes. The people who use the respective OS's will learn about the update from their own sources (I learned about this update from 5 sites before Slashdot told me about it) and the others couldn't care less about minor fixes that don't affect them.
Besides, the most important reason to read slashdot is for the comments and these topics get the worst comments of all. It's mostly trolling (Mac/Linux/BSD rulez/sucks) and uninteresting trivia (did you know that this update fixes DVD on external monitors?). Get me some serious technology news that illicits comments about the problems of EPIC or an ethical story about censorship.
These are precendents that can make you win the case if you can keep the DMCA out of this:
;)
[...] Over the years, the use of reverse engineering of software has been extremely controversial. In fact, there have been two major lawsuits concerning products that were written with the help of decompilation.
The first major lawsuit was Atari vs. Nintendo. In this case, Atari Corporation reverse engineered Nintendo's game cartridges in order to create alternative games that could run on Nintendo's system. The judge in this case found that decompilation was acceptable to figure out unprotected elements of the software. However, Atari lost the case because they used fraud to obtain Nintendo's source code from the Copyright Office [7].
The next major case, Sega vs. Accolade, created a much stronger precedent because there were no fraud issues involved. In a similar manner to Atari, Accolade reverse engineered Sega's Genesis technology to discover how to make games for their system. Accolade then created a book with the relevant (and non-protectable) elements of Sega's technology, and passed the book on to their developers. These developers created a new game, Ishido, to compete with Sega's games.
In deciding the case, the court looked at many factors (including public policy concerns). In the end, the judge decided that reverse engineering software for the sole purpose of creating a compatible package is an acceptable use (under the "fair use" doctrine). In addition, the appeals court stated: "[i]f disassembly of copyrighted object code is per se an unfair use, the owner of the copyright gains a de facto monopoly over the functional aspects of his work - aspects that were expressly denied copyright protection by congress" [1]. Thus, the court decided to adopt the policy encouraging competition (as opposed to IP protection) in the software industry.
This is blatantly stolen from Stanford. The guy should be able to argue that he is reverse engineering for fair use, especially since he's advertising programming the system and backing up savegames & games. Of course, the DMCA is created to take away your rights to fair use by disallowing any decrypting.
I'd call upon Nintendo to prove that the device doesn't merely reverse-engineer for fair use, but it truly circumvents an encryption device. This means that they have to prove that such an encryption device exists. But of course, IANAL. You should really get a lawyer to look at this and write a letter back. And you're in big sh*t if you find out that Nintendo uses an XOR-encryption scheme for it's code
I don't know it it really is better on a Mac. I can't see how it would be.
MacOS 8/9: Applications install 99% of their stuff in their application's folder. Sometimes they install an extension in a special folder, I use an utility to manage these files. This allows me to identify them, group the files and delete them. This works great for a power-user. A reinstall of the OS can be done without many problems, most software needn't be reinstalled (no registry, copy the needed extensions with the utility or by hand). If you copy the preference files from the preference folder, you don't have to set up applications again (I do and sometimes go through it and delete the obvious cruft). I do this at most once a year and it works extremely well, bloat is very small (~50mb per year, mostly because I'm lazy and don't clean up everything perfectly). But the OS allows you to do so almost perfectly, Windows hides everything so managing installed software is extremely hard.
OS X: Some applications are shipped as self-contained packages, only preference files are left in the system (no (de-)installer needed). Others do install cruft, I'm still figuring out how to manage it (it's still a bit of a chaos, developers are still learning about the workings of the OS). I suspect this mess will improve in time.
Windows: My experiences with Win9x have been horrible, reinstalls were regular (but I do install a lot of software) and very painful (having to reinstall almost all my software). Win2000 is doing well for me, it's fairly stable and doesn't start to lose features (Win98->where did my OpenGL go?). So I needn't reinstall, which would probably be just as painful as reinstalling Win9x.
I've never used assertions, but they seem a great complement to unit testing. Unit testing allows you to write code to test your functions and easily see if something breaks, the major problem is that they lack an easy way to look inside objects to keep an eye on internal consistency. Assertions can be great to catch those silly little boundary mistakes.
A good unit testing framework for Java is JUnit, they are available for other languages as well.
BTW, you can create your own assertions with Log4J, so even JDK 1.1/1.2/1.3 users can use them:
if(logger.isDebugEnabled())
if (bla>10)
logger.warn("bla>10, bla=" + bla);
This uses almost no CPU-time if debugging is disabled. Log4J is a very good logging package, it surely beats System.out.println, check it out!
The free Quicktime Streaming Server 4 has been released today and its open source companion the Darwin Streaming Server (has the same features, but it does run on Linux, BSD and Windows). It already supports Mpeg-4:
"MPEG-4 Support: now you can serve ISO-compliant hinted MPEG-4 files to any ISO-compliant MPEG-4 client, including any MPEG-4 enabled device that supports playback of MPEG-4 streams over IP. You can serve on-demand or live MPEG-4 streams, and reflect playlists of MPEG-4 files."
I'll bet they tried to mention MPEG-4 as many times as possible.
You can now also stream MP3's with it, set up your own radio station! The streaming uses the standard Icecast streaming format so any MP3-player that supports streaming should work.
The worst thing that can happen is that some freeloading company doesn't use my code.
/. was a fishing simulator.
No, the worst thing that can happen is that companies do not choose to use and enchance your product, because of which it doesn't gain momentum and fades away. This means that your work is wasted.
The best thing that can happen with BSD-licensed software is that everyone gives up their own commercial products and decides to built upon the open source code. A good example is Apache, companies have made great contributions to it (I think that most programmers who work on it are paid by corporations) and most competitors have given up. This is extremely unlikely with GPL'ed software.
You can imagine that companies prefer the BSD license all you like, it doesn't affect the fact that most of the worthwhile software is currently being developed under the GPL.
Apache, XFree, Xerces, Xalan, Tomcat, Python, PHP and Perl are all licensed under a non-copyleft license.
But I guess that this software isn't worthwhile.
It's also realistic. Had a look at the headlines lately? *cough* Enron *cough*
One example of a company that went crazy and got punished for it (the story could turn into a Disney movie, except for the fact that the employees got punished and those responsible got a lot of money). I counter your one example and up you one with Apple and Sun, who both put a great deal of effort into open source.
But it there a point to this? We could talk all day about the morality of corporations. In the end what counts is whether GPL is necessary to protect software and/or to get enough contributions. I can give many examples of BSD-like licensed software that get great support by companies and have never been abused. I'm quite sure that you'll only be able to find a few examples where the open version floundered, while a hijacked closed version flourished or examples of abuse.
Why do you get so hung up on the few bad examples? Suppose that Newton would have kept his discoveries secret because his calculations might be used for evil (like aiming cannons) or that the automobile would not have been produced out of fear that it would hurt people or flying would have been banned because planes might be used for fighting. We would have still been in the middle ages if everyone thought like you.
Engineers and scientists are and should not feel responsible for all the uses of their inventions. They should try to do good and focus on promoting and working on the uses that benefit mankind. But IMO they are not responsible for building systems that never can be used for evil, for that automatically means that many good uses become impossible.
Hmm. Ever wonder why they didn't choose to get behind BSD? Maybe because they don't want their hard work going to finance Microsoft any more than I do. I could be wrong, of course. Maybe its just because linux is a better system.
Do you mean the BSD-license or *BSD (the OS's)? I'll assume the latter.
Linux has become a hype and has gained some serious momentum, these companies want to profit from that. Other companies (Yahoo, Apple, many ISP's) have chosen *BSD because it suits them better and they don't have to sell it or it is a minor feature of their system (in marketing that is).
Maybe because they don't want their hard work going to finance Microsoft any more than I do.
AFAIK just about the only BSD-software that went into Windows is an early TCP/IP-stack (now fully replaced) and some related utilities (ftp.exe, etc). Without it, the Internet may not have grown that fast and Windows-users may have been shut out for a while. So I don't see anything wrong with it.
But of course, I don't see anything wrong with commercial companies using freely available code as long as they behave morally and lawfully. We should encourage the use of standard, well-written software, just like we should encourage open standards like DDR DRAM. This will benefit the entire market, small and large companies alike (both small and large companies are making money of Apache for instance, it creates an open playing field). If companies do not behave, they should be punished by law or by the power of money, we should not buy their products.
MS is an example of the failure of American law in allowing a monopoly to abuse it's powers, not the failure of open source under a non-copyleft license.
I don't see that happening though, so I will guarantee that at least my contribution will always lead to more openness.
GPL cannot guarantee it, your code can still fade away. A strong, supported BSD-product will also lead to more and more openness as no-one will have the ability to force it into closedness (and thus taking away many advantages). An example is Apache, do you believe it will be closed up again ever?
My crystal ball tells me that eventually, the only role for commercial software will be specialized applications. Everything else will be open source.
Your crystal ball seems to be manipulated by Sauron. Tell me how open source can ever prevail for games that have a 1-year life-span and must be created in a short period by a full-time team as not to fall behind in graphics?
The open source community already has a hard time creating a decent browser and word processor, software that interests millions. How will open source ever be able to cater to fairly small, non-programmer communities? An example that was given by somebody else on
BTW, I assume that specialized apps mean applications adapted or designed for one entity (aka custom-made products). You might also define it as everything outside the OS, so it's a bit unclear what you really mean.
I know that he said that, but perhaps he didn't see the the PPC roadmap on Motorolas website recently. The G5 is 64-bit and is 32-bit backwards compatable.
That's not true. There will be a 64-bit version with 32-bit backwards compatibility and a 32-bit version (just like with the PowerPC 620 vs 601). That is what it says on the roadmap. Since programmers are still transitioning to OS X, I wouldn't expect the switch soon, at least not until the high-end users start complaining about a lack of RAM. Don't forget, 64-bit will hardly make a difference in speed for most uses, especially since the G4 already has a 128-bit processing unit (Altivec). Any application that would benefit greatly from 64-bit instructions already supports Altivec. On the other hand, 64-bit instructions/data means moving twice as much data, even for 32-bit instructions. Given that modern CPU's are for a good part limited by bandwith issues, this may make a 64-bit CPU slower than the 32-bit version.
The larger registers and caches will increase the size of the die and thus the cost of the chip. So I'm not looking forward to it yet, my next computer will probably not (need to) move beyond 2GB of RAM, so I can wait.
Umm, I don't have to depend on a company's goodwill
You do if you want them to contribute to open source. You start out at the proposition that people use your open source product, I take the whole picture into account:
total contribution by companies = use of open source*contribution per project
BSD maximizes the first, which will probably maximize the total figure (as I have pointed out that it has great advantages to contribute). You just focus on the second figure, but 0.0001*50 is still not very much.
Companies exist to make money, not to be altruistic. If they that think they can get away with something, they generally will. I have no qualms with taking the decision to use my code in a way I don't like out of their hands.
That's a very cynical view. Most companies do actually worry about more than pure profits. Did you check out the list of companies that double their employees donations for the EFF? It's long and heartwarming, snif.
I really don't see that many 'abuses' of BSD. Most companies who use BSD (or similarly) licensed products do contribute (or have even created the product in the first place), so why would you need to force them?
With both IBM and SUN recently getting behind linux in a big way, it looks lik the GPL is, in fact, becoming more pervasive. These are not small time players - they know what they are getting into.
They like it as a server-OS and certainly did not pick it because of the GPL, just about all the other products they support or open sourced themselves are non-copyleft.
What you are claiming as a legitimate use, I see as an abuse.
Sure, the examples I gave would truly have hurt the community (not). But of course, usage of IP is an abuse if it doesn't follow your rules. That's the same logic in EULA's, DVD regions and copy-protection, protected CD's, protected ebooks, the DMCA, etc. It's the thinking (and laws that support it) that I want to abolish, the abuse of power because you can create/control and others cannot. You focus so heavily on enforcing your rules, why are you unwilling to discuss the validity of the rules themselves?
It seems to me that you know you have lost this discussion and are trying to weazel out of the question at hand: Is copyleft truly necessary to force companies to contribute? I claim this is not so and thus it is best to try and pick a license that maximizes use and reduces adversity and legal issues. The BSD-license seems to do this well.
Of course, GPL has it's place for certain uses, but BSD is my first choice. Why? Because I want open source to achieve it's true potential, a strong force that goes hand in hand with commercial software. Open source will be able to supply some of the software we need, while commercial software will fill the niches that cannot be filled by open source.
*I* don't get any extra choice. Nobody is ever going to use my own code as a starting point to enforce a software patent, possibly against me.
You do get an extra choice, even if you aren't willing to consider it. Others may. I don't have a problem with sufficiently advanced patents. They do have their place (but should be strictly limited).
But your crusade against patents should not be fought over open source code. Patent issues are not specific to open source, they can/do harm commercial software just as much. The proper way to fight them is through politics (write your congressman) or through law (donate to the EFF).
I don't care what any company chooses to do or not to do. If they want to use my code, they *are* going to contribute back - no maybes about it.
Then you shouldn't use the GPL but a more restrictive license ("If you use this software, all your base belong to us"). GPL'ed code can be used without giving back as well (by not distributing the code). Companies are not forced to make changes and thus may leech. And I wonder why you don't mind when good code isn't used because the license doesn't allow perfectly valid uses. This a big shame. If more companies will use open source they will lower their costs, thus learn about the advantages of sharing their IP and start contributing.
Your simple-minded argument doesn't hold up very well. If you do have to depend on companies' goodwill and commitment for them to contribute to open source, why not go the whole way and actually threat them like adults. Give them freedom to make their own decisions. If you truly believe in the merits of open source, you must believe that it will be beneficial to companies and they will thus support it.
I trust that companies will give code back. It is often in their best interest (as I already pointed out). GPL will just ensure that your code isn't used, and thus you won't get anything back from them.
The more pervasive the GPL becomes, the less likely this will be a problem.
I believe that GPL will (hopefully and probably) not become that pervasive among companies (except for Linux). Instead it may give open source a bad name when companies get bitten by the GPL license. Is it not better to let money do the talking, instead of lawyers?
1. I make changes to the software that are specific to my business. It can't be open because that would expose my secrets.
You don't have to release the source unless you intend on distributing your program.
Suppose that Shell wants to distribute software to gas stations that link into it's own systems for greater efficiency (with JNI for instance). This specific code may not be open sourced because provisions for future products are put into the code before they are available. Those gas stations are independent and the software is thus 'distributed'.
2. I have created my super-duper commercial package and want to link it to an open source web-server. I'm willing to open source every addition to the web-server I make to suit my needs, but the rest of the software is my property.
If the web server is LGPL, no problem. If it's GPL, find another web server.
This is not a denial of my point. Argueing that GPL software cannot be used when I claim that some legitimate and important uses are not possible with GPL merely proves my point. The LGPL is far different from the GPL of course, let's keep focus.
3. I want to add video codec X, it's not mine. My clients are willing to pay for it though. What's the problem with making a closed version that includes the codec?
I think this one depends. IANAL, but I don't think that your licensing a codec and including it in some GPL software for someone would be considered distribution as long as they weren't otherwise distributing it. For instance, if I worked for a company and we had standardized on some GPL video editing software, what would be the problem with my doing anything I wanted with that software as long as it didn't leave the company?
Every company would have to license the codec and integrate it seperately??? Suppose we are talking about 20 codecs by 20 different companies, many of which need to be adapted to work with the software. A central entity can make this happen and ask a fee for their hard work (and to cover the license costs). Single companies may not have the resources to expend this much for just themselves and they may not have the clout to get the codec-providers to adapt their software (small companies are especially hurt by this, making it harder for them to compete with large corporations). So either this will be impossible or it will cost far more to accomplice (a bulk license for the codecs will usually be cheaper).
This points out a big problem with the GPL, BTW. It discriminates against a network of small companies in favor of large corporations. The latter can share an altered GPL-program without opening the code (for which there may be good reasons), while the network cannot. They are at a disadvantage. So choosing the GPL unwillingly promotes mega-corporations.
Unless, of course, they tie it in with some patented or otherwise encumbered code that is not legal to duplicate.
Which couldn't be used in the GPL-version anyway. So you still get an extra choice that you would not have had with the GPL-software. Nowhere is anything taken away from the free version by this!
BTW, code can't be patented, only an idea or algorithm. Of course the real answer to the inability to use an trivially simple idea or algorithm in open source (or closed source by another company) is to limit patents.
Human nature is what's wrong. I don't trust my fellow man enough to assume that he will be as forthcoming with his additions to my code as I was with the original code. Therefore, if he wants to have the advantages of my code, he is going to contribute his advantages back to me.
I'm convinced that a very large percentage of (usable) open source code has been written by cooperations or in the boss' time. Sun wrote a lot of code for the Jakarta project. IBM and Lotus wrote a lot of code for the Apache XML Project. A lot of companies contributed to Apache itself. Apple is paying a FreeBSD-developer and makes some nice open source software. The list goes on and on.
Why are you so paranoid when many companies are already proving that they are willing to contribute? Of course this usually requires that they can actually use their own contributions for their uses and a non-copyleft license is thus very important.
I don't care what they use my code for, as long as they give back.
I trust that companies will give code back. It is often in their best interest (as I already pointed out). GPL will just ensure that your code isn't used, and thus you won't get anything back from them.
I guess I'm just pragmatic.
What 'legitimate and important uses' require that you can't give back your source? I can't think of one.
1. I make changes to the software that are specific to my business. It can't be open because that would expose my secrets.
2. I have created my super-duper commercial package and want to link it to an open source web-server. I'm willing to open source every addition to the web-server I make to suit my needs, but the rest of the software is my property.
3. I want to add video codec X, it's not mine. My clients are willing to pay for it though. What's the problem with making a closed version that includes the codec? I'm willing to contribute every change that is 100% mine to give away. A good example of this kind of thing is Darwin, Apple can't give away the driver code for ATI or NVidia cards, so that code is only distributed as binary.
These uses sound perfectly legitimate and important to me. They make or break the ability to provide useful features for clients with open source software. These uses do not take anything away from the original code, not do they cause a company to withhold code that they would otherwise open up. So why are they not allowed under the GPL?
Do people write code in order to write good code and improve the state of computing, or do they do it in order to coerce other programmers into helping along?
The GPL guarantees both, while BSD only guarantees one.
Not necessarily. Many companies are afraid to use GPL-products, they want to have the option to ship a version that is linked to their own software or add a feature that is licensed from a third party or whatever. These are perfectly valid uses that are disallowed by the GPL (the second use may clash with LGPL).
These things can be done with BSD, so companies can actually use it. Any changes they make that are useful to others will usually be put in the main tree so:
1. There are fewer diffs between the open version and their product->easier merges
2. Their code is tested and audited and expanded and debugged. This is a big reason for using open source.
3. Why not?
4. It will be good PR.
5. It may actually make them feel better (especially the programmers).
BSD-licensed code will thus probably get you more good code. If someone does create a closed version with features you like, open source programmers can just copy the features. They haven't taken away anything from you, but may have filled a need that open source developers did not fill. What's wrong with more options?
I don't really understand why people get so upset when there might be a chance that someone uses the code he gives away for free for things he doesn't like. This is the same paranoia that the RIAA/MPAA exhibit: "if we don't put some überprotection on our stuff we will get screwed". They don't care if many legitimate and important uses are no longer possible, like making back-ups or listening to music on my computer. Strangely enough the Slashdot crowd is enraged over this, we tell them to be more trusting and to find ways to make money without creating a policestate. On the other hand it is okay with many of us if programmers make legitimate and important uses impossible with the license they choose for their IP.
Options are sometimes considered to be a good thing.
You are clearly not a minimalist. Many see the advantages of having one clear way to do something, instead of many options, all with their advantages, disadvantages and different abilities. A single, consistent design gives you clean and logical interfaces, which are fairly easy to use. Especially for authentication it may be better to have a slighty less capable interface, instead of an interface that people can't comprehend and thus misuse.
Of course you may sometimes need something more potent, this should be achieved by extending the basic design ('advanced options') or by adding a separate more complex interface. This should be a superset of the simpler interface. Users can thus be pointed to the simple interface by default and can migrate to the more complex one if their application becomes more complex.
I can't really see why they have the need to come out with ever faster graphics boards. I can play games like Deus Ex and CounterStrike fine already with my GeForce3 (in fact, it is already a little overkill). You really have to start asking the question "Are they ever going to make a game detailed enough to take advantage of this capacity?"
No, of course they are never going to make a game that is as pretty as Final Fantasy or Shrek.
This is a typical case of duh. Remember your post when Doom 3 or Unreal 2 comes out and the Geforce 3 can't reach the highest resolutions at decent speed.