Red Hat is modifying the kernel, don't read the headlines? Although it won't effect the userland, try apply many existing patches against the existing Red hat source code. There will be problems.
Exactly.. won't affect userland. Which is what I was talking about. I never said distros don't patch the kernel. On the other hand, you can always plug in a virgin kernel without issue.
It's fragmentation. There's no reason why KDE couldn't have used GTK. But they didn't, and now programs are less portable.
It seems by your definition, the simultaneous existance of C, C++, and Objective-C is an example of fragmentation. If so, that's patently silly. KDE didn't choose GTK because at the time it was vastly inferior to Qt. Fragmentation would be if KDE had split off of the GNOME project.
You admit it's fragmentation, good. You're learning.
I never said that smartass. C/C++ doesn't have the level of standardization that Java does because of its history. Modern languages do and Open Source implementations of those languages are on spec with high precision.
Then you're a fool. There are millions of reasons why Java will fragment if it goes open source. The C standards are solid standards, but GNU made extensions anyway. Open source developers do what they want, standards or not.
The C/C++ standards are *now* becoming solid, but they have an enormous history of variation. Some things that are "traditional C" are not part of the ISO standard and this is why GCC supports them unless you turn off support.
Java already is going Open Source and it's not fragmenting yet. Get your head out of the sand. A forked Java is not in anyone's interests and there's nothing to suggest it will ever happen. It very easily could have by now, but it hasn't. And I'm talking about all the proprietary, non-Sun implementations even.
Open source zealots would choose an open version over Suns clean-room version. It's what you people do. Sooner or later, the Sun version isn't mainstream.
You mean: Open Source advocates will choose their own open clean-room version because its the only one they can legally redistribute and it's just as good if not better than the proprietary versions. Sun's version won't be mainstream too much longer if they don't find a way to participate in the inevitable change.
IBM, for example, could fork the code and really open up to the community. The entire open source community gets behind the IBM version and, after awhile, Sun disappears.
IBM's code is based on Sun's. They can't legally open it. Now, suppose Sun was to GPL their implementation. IBM could participate openly, but they still wouldn't fork because it would break all compatibility and everybody would cry foul.
Uh.. they're everything like Java. They're just behind on implementing the newer specs.
They do not run on every platform supported by Java nor do they have all of the GUI functionality.
Even Sun's JVM doesn't run on every platform supported by Java. (-: As for GUI stuff, it's still in the works, though SWT will probably be possible before AWT or Swing since it's already open.
I'm sorry, but having to install/configure additional compatibility libraries on each platform that is non-linux does not count.
Invalid argument. You're criticizing a young but rapidly growing project for not being what you want today. Who cares. Wait another few years and keep using the official Java implementations until then.
To most people, Kaffe and GNU Classpath do not exist. Sorry to break it to you.
Only because they're not completed yet. I don't use them yet and I never said it was possible to use them in place of the complete Sun/IBM Java runtimes today.
Yes. A certain vendor that did itself try to fork Java, and now has a competing product that is extremely similar to Java.
What this proves is that it doesn't take Sun Open Sourcing their own implementation for a fork to occur! And yet, the existing Open Source implementations of "Java" are not forking from the spec.
If Java went GPL, this company would not hesitate to muck around with it, just because they can, and the company has a history of doing anything it can to 'win'.
How exactly could they muck around with a GPL (or similarly licensed) Sun Java to win? They couldn't release proprietary versions. They couldn't make changes without telling everyone. They couldn't completely fork the code and still call it Java (Sun trademark). And they don't even need Java anymore since they have.NET and CLR. MS is really out of this picture. Oh, one more thing: they won't touch anything GPL with a ten foot pole because that would only further legitimize it to those on the fence.
Sometimes yes, sometimes no. I would not expect Java to fork in the classic sense of the word (except the version that one particular company would make), but look at Linux as a whole; multiple distros out there, by companies, by individuals, by governments. That kind of "forking" alone would make it more difficult for WORA with Java; it'd increase the amount of tweaking neccessary. Every distro tweaking the JVM a little differently than the others... code then running great on some and flakey on others...
Linux distros are not a proper comparison because they are a conglomeration of very diverse software. Here, we're only about talking the JVM and a set of libraries. There's no incentive to fork and any 'tweaking' would only be within the implementation of the specs. With Sun still holding the trademark, they would still be the sole source of endorsement. So "Bob's J*** Derivative" could fork and tweak to heck and back, but typically only experimenters would use it. And if they came up with something really great that SCP approved of.. cool! roll it back into the official codebase that normal people use.
well, maybe that's already true. Why make it worse, though?
Because there's virtually nothing to lose and a lot to potentially gain.
I could see the same easily happening with Java; different bright folks wanting to take it in different directions, without a leader, ending up disagreeing too strongly and going it alone. Would Sun and IBM and Microsoft and Oracle and so on always agree on the next level to take Java to? I doubt that.
Sun would still be a strong leader, even if they Open Sourced their implementations. Weak leadership or poor quality is what encourages forks. As for IBM and Oracle contributing, that's what the SCP is for. It works and is well established. Why would anyone want to diverge? All that would do is make them look bad.
For the record, I'm undecided about the opening-up of Java.
Java is already being opened-up. (http://www.kaffe.org/links.shtml) It's just a matter of how long that takes. If Sun helped out, it would happen a lot sooner and dramatically increase their influence over the whole 'open Java' scene.
The Linux itself is fragmented across distros. Try taking a Red Hat binary to SuSE, there will be trouble.
This is an improper example. Try taking some C source and compile it on both RedHat and SuSE -- same results. Fragmentation would be if RedHat or SuSE modified the Linux kernel or supporting GNU software so much that code would need significantly re-written to work.
There are now 3 projects who will use the X windows standards, and I can tell you there will be fragmentation. Each will try to improve it somehow and bye bye compatibilty.
Not even close. Read this: http://www.xouvert.org/faq.html#2.1 It should be especially noted the strong emphasis on sticking to standards. There are many X11 implementations and they are very much compatible, though some are at different levels of specification advancement.
There may be one GNOME and one KDE, but each use a different GUI toolkit resulting in the same effect as fragmentation.
This is an example of diversity, not fragmentation of an original standard. Would you say that there should only be one programming language? I should hope not. Each has its advantages. It should be noted, of course, that there is a growing movement to harmonize GNOME and KDE, but that's another topic of discussion.
GCC has it's own extensions, as does the GNU C library. Same goes for gmake. You can't compile software that use these extensions on Solaris using native compilers.
It can be easily argued that the only reason C compilers have extensions like this is because there does not exist as active a standards body as say, W3C or SCP, to recommend changes before they are tried experimentally. There are also different ISO C versions. I believe some of the GNU C "extensions" are actually part of the C99 ISO spec. Solaris's native compilers may be using an older spec, but I'm not familiar. Also, you can use the -pedantic option with gcc and it will warn you of any code that uses extensions beyond C89. There is little to make me believe a similar scenario will exist with Open Source Java. The industry has learned since C/C++ the value of creating standards early and then sticking with them. As I said, Open Source developers tend not to deviate when solid standards exist. Java is a solid standard.
No, it would only hurt Sun and Java.
How much have GNU ClassPath and the various open JVM's hurt Sun and Java? And how much have they deviated from Java specifications? The answer to both is: not at all. If Sun relaxed licensing on code for their implementations, it would only serve to promote their own over the clean-room versions.
How does BSD provide less freedom? Are we talking about freedom of the code again? I'll let you on a secret. People can have freedoms and rights. Objects and information do not have rights. People can own information and inanimate objects.
BSD does not insure that what people create and then desire to be free, remains free. There are both philosophical and commercial reasons why this "restriction" can be desireable.
And, by the way, people cannot own information. They can only be granted a temporary monopoly on the use of information within the borders of a soverign state that grants such monopoly. Big difference philosophically and legally speaking. And, in case you're wondering, this viewpoint is fully backed up by the US Constitution and legal system. It's not "GNU/Koolaid" or hippie-speak.
How do you expect Java to be open sourced without Sun's consent? It is the IP of Sun. Careful about calling people idiots when you make statements about Sun being forced to do something with their property.
Sun's Java implementations are copyrighted by Sun. Clean-room Open Source like Kaffe / GNU ClassPath are not. (and legally they're not even Java(tm) but they effectively they are) Nobody ever said anything about Sun being forced to do anything with their own code. It would, however, be mutually beneficial if they released as much as possible under a freer license instead of fighting what they cannot stop.
Open sourcing Java under any license would be a mistake. Numerous forks of the code would occur completely destroying the "write once, run anywhere" nature of java. Maybe you don't understand that not everyone runs either Linux or windows on their desktop (not to mention phones, set top boxes and PDAs).
This is 100% pure recycled FUD. Since when do large, successful Open Source projects fork? Even more so, since when do Open Source projects go against established industry specifications? If Sun GPL'ed their Java code and somebody tried to fork it to be a jerk, nobody else would pay attention. It's entirely against the community ethos to fork without significant cause or out of spite. Practically speaking, large projects have gravity regardless.
As a programmer of in-house proprietary software, I'm painfully aware of forking issues. At times we have two developers working on two future versions and releasing patches to the version in production.:D
That's not forking, it's a management problem. Forking is when somebody purposely takes all the code and runs in a different direction.
Yes, there be ports to other platforms (I'd love to see Java on *BSD). Yes, there'd be performance patches and probably security patches. But with this comes the down side to all open source projects: Fragmentation.
Open Source fragmentation is a myth largely propagated a certain proprietary vendor's FUD campaign. Successful open source projects simply do not fork. Why? Because there's no interest and it's way too difficult with large scale software. There is one GNOME, one KDE, one Apache, one Linux kernel, one GNU Compiler Colection, one OpenOffice, etc. Similarly, an Open Source Java would be a large project with so much momentum that any unlikely attempts to fork would be rapidly ignored and wither away. Not to mention Sun would still hold the Java trademark..
One of Javas goals is to provide compile-once-run-anywhere. At the moment, it does a decent job of doing this. Other platforms do this better (Read: Inferno) but Java is trying. With fragmentation, compile-once-run-anywhere disappears and would be impossible to bring back.
Wrong again. Open Source developers are terrifically anal about sticking to standards. You don't see a dozen different C compilers each with a slightly different dialect do you? You don't see KHTML or Mozilla trying to extend web standards do you? Even more obvious: You don't see the existing clean-room Open Source Java projects deviating from Sun's specifications do you? -- and they're not even under legal obligation!! Fact is, an Open Source Java is the only way WORA can ever hope to fully live up to its promise. (btw, that's Write Once, not Compile.. minor correction) Three reasons: 1.) more ports 2.) better quality control 3.) less Java implementations floating about, whether proprietary or open.
Let Sun has Solaris and Java. Stop trying to force them to open source their software.
Nobody's trying to force Sun to do anything. They're simply asking because it would be mutually beneficial.
I think that there is exactly one person who thinks that Java should be GPLd.
As opposed to what other license? (L)GPL would probably be the best choice for Sun because it ensures that people cannot create proprietary forks (Microsoft anyone?). BSD or similar less freedom-guaranteeing licenses would be an awful choice. And if you think Java shouldn't be open source at all, well.. you're just an idiot. It's going to happen whether Sun wants it to or not. This whole discussion is only a question of whether they remain a player.
Open source is not a career. I don't know who told you it was but you need to disabuse that notion.
Is that why I make my living doing Open Source consulting and paid development? Darn, guess it's not a career.. I'll have to do something else now. Thanks for enlightening me, but my clients will be very disappointed.
So, GIMP developers, clean up the interface and change the product name, and your program has a decent chance of seeing the light of day in the real world.
You forgot one thing: Set up a way that ordinary users (or even companies) can donate to the development of certain features / interface overhauls. Gimp is a hobby project. If people expect it to become more professional (which they clearly do by the tone of this article and the followup posts), the project is going to need one or more full time, paid programmers. It won't happen otherwise -- at least not within a reasonable timeframe. Its really as simple as that.
I would almost guarantee that if every Photoshop user chipped in $5, Gimp would put Photoshop out of the market with 2 years. Why would anyone help pay for something that's free? Well, how much does a Photoshop upgrade cost every year or so? The only problem is how to efficiently advertise a fundraising drive...
Define "isn't moving faster". It's moving a lightning speed. It's being adopted by all major IT companies except one. Even the unmentionable company is flirting with pseudeo OSS. Exactly how fast were you expecting it to move anyway?
I don't share your optimistic appraisal on the "lightning speed" bit. Yes, OSS is moving faster than it ever has before, but it's not enough. It's certainly not a revolution yet. Sure, there are a lot of large, high-profile companies utilizing Linux in one form or another. (and note the distinction.. usually Linux, not OSS) Yes, there are a lot of smaller companies using Linux / OSS in some capacity to meet some of their needs. But the fact is, OSS has the potential to grow exponentially and it's not. It's not for one reason: not enough people are being paid to do full-time OSS development. Why is that? Because too many geeks treat it like a toy and not a career. You can't expect change when you desire OSS but then go work a day job writing proprietary software. For all the brains in the OSS community, there sure is a lack of business intelligence in exploring the many ways to make a career out of supporting or developing Free software.
Open Source will be working when companies worldwide are ditching Windows / Office and other common proprietary desktop software en masse. Open Source will be working when collaboration among consultants to write modular business software is the norm. (think 'Apache') Open Source will be working when most job descriptions drop the acronyms.NET, VB, and MCSE. Open Source will be working when it represents the ethos of almost every IT department.
Open source is _not_ about competing with anybody else. It is _not_ the goal to create a competing technology to win market share or anything. Open source is an exercise in technology, invention and freedom.
That stupid philosophy is precisely the reason why OSS isn't moving faster and providing more geeks with jobs writing Free code rather than proprietary. Yes, OSS is an exercise in technology, invention, and freedom. But, at the same time, that exercise is worthless if it doesn't significantly improve the state of the industry / society. To do that, it must garner market share because that is the only way to diminish the forces that work against freedom. Do you quite realize that if OSS does not dominate the industry in the fairly near future, the powers that be will effectively make it illegal to write software unless you work for a large corporation with patent portfolio cross-licensing? Do you quite realize that an industry dominated by OSS would be a much more pleasurable one to work in? There is far more at stake here than academic toys and utopian musings. Get out of your bubble.
I really hope this isn't serious... how exactly did you plan to get very far in a field you have no formal education in?
Sorry, but that's an extremely closed-minded way of thinking. While I agree that most everyone should complete at least a 2-year program in some area of study, simply for the social side of things, there are plenty of accomplished programmers without a formal CS background. And, likewise, there are plenty of people with formal CS backgrounds that are truly incompetent at their jobs. It all depends on how you personally learn best and what career path you intend to take. No, you won't likely get a big corporate IT job without a CS degree, but not everyone wants such a job. (insert picture of Dilbert)
Here's one possible idea: Become an expert at J2EE and do Open Source based consulting. In a down economy especially, companies are looking to cut their IT costs while still using proven technology. You don't need a degree to hire yourself, and you can certainly do better than $13/hour. If you like, take some classes on the side in something that interests you--business administration maybe? Of course this route is not for everyone. You need dedication and an entrepreneurial mindset to some extent.
How can they possibly say it's a potential good fit for smaller companies?
Small companies are able to adapt quickly, so they are the first candidates for cost-saving tools. Large companies, in comparision, cannot adapt as quickly, but *can* invest in making improvements to Open Source software that is almost but not quite useful to them (such as to save money by removing the need for the proprietary standbys)
- It doesn't work for advanced Excel (read: The Finance Department).
So have everybody else use OpenOffice and let the finance people keep their existing Excel. There's also Gnumeric, but I'm not sure what the comparison is as of late..
And of course, finance data should be kept in a database anyways, not in spreadsheets. C'mon, this is 2004, not the 80's. But that's another discussion..
- Support options are limited (read: DIY in a small company with limited/nonexistent IT resources to begin with).
How often do people need "support resources" for an office suite? And since when does MS Office come with amazing support resources for small customers?
- It takes as much as 10 seconds longer to open big docs sent in Office format (read: anything sent to you most people outside the company).
This is so trivial it's not even worth mentioning.
And, let's overlook Outlook in the comparison. (Evolution, Thunderbird, et. al. do not offer the same functionality)
I assume you refer to Exchange, not just Outlook itself. There are plenty of alternatives to using Exchange/Outlook such as OpenGroupware and Kontact. And beyond that, Web-based groupware solutions are superior anyhow in most cases.
Oh, and feel free to mod me into oblivion for taking a controversial (for/.'ers) stance.
What are you talking about? Controversial stances, regardless how stupid they are, get modded up on/. these days. I call your karma whoring. (-:
While the implications of this treaty are truly frightening, the amazing thing about it is that it originated in Europe.
It's not really so amazing when you consider that the Clinton admin, which also brought us DMCA through the backdoor of a WTO treaty, was largely responsible for drafting / pushing the cybercrime treaty as well.
And here's a fun one: http://www.computerworld.com/governmenttopic s/gove rnment/story/0,10801,55949,00.html Oh how times have changed, eh?
Sadly none of this stuff will be discussed in either party's presidential candidate's 2004 campaign. Why? Because they basically have the same brain-dead stance. So you all know what to do: start writing congress immediately!
In my opinion the more important issue is submarine patents
Even more important is the fact that patents are being granted for trivial things undeserving of a patent. As example: anything software related. Software is an evolutionary technology -- all improvements to the state of the art are, by themselves, merely logical next steps and not true innovation. Implementation is where the software industry innovates, and that is covered by copyright. The reasons why patents should not be granted for software are much the same reasons why patents are not allowed for mathematics or styles of literature, art, and music.
Like a Epson Powerlite S1 or an Infocus X1. The picture will be infinitely better than what you'd get from a homebrew. Don't even consider the homebrew thing. Really. I've seen a few (based on CRT's or overhead projectors and LCD panels) and they are really poor.
I second this opinion. The homebrew thing is not worthwhile today given the cost of the latest round of commercial projectors. Two years ago, when I started tinkering, it was feasible to attempt. Low-end projectors then cost 50% more for significantly worse performance. Today, you will have to spend $600-700 to almost match the performance of a $800-900 entry-level commercial DLP projector like the X1 or S1 mentioned. Don't be fooled by sites that claim it only costs $300-400. That is only for the specialized bulb/ballast, lenses, and cabinetry required. You're looking at another $250+ for a halfway decent LCD screen that you can remove the backlight from and project through. Those old projection panels you can find on eBay are crap. (Proxima, iView, etc.)
Trust me. I'm a hard core DIYer and I've tried the homebrew LCD projector thing. It's not that it doesn't work at all, but this is a case of inadequate technology available to the home experimenter to match commercial products. It's simply not worth your time and efforts. If you want a good project, build a "Sonotube" subwoofer. That will save you lots of money *and* give you outstanding results. Once you get a projector, you can make your own custom screen that fits nicely with your HT room. That's another worthwhile money-saver.
Certainly there's some truth to that. I was simply making the point that assumptions should be avoided. Of course, a counterpoint to your scenario would be that most sane societies, in either case, would have laws prohibiting any sort of dangerously "booby trapped" devices, whether sold commercially or given as a 'gift.' In the end, it all comes down to a balance between encouraging a vibrant economy and ensuring basic rights, freedoms, and safety. Often, those two goals actually go together.
You have some good points, but some things should be clarified..
3.) Most CD's are quite overpriced and the public now realizes this. Allow me to plug Mangatune.com Reasonable price, actually supports the musicians. (:
4.) Copyright duration is way to long and it is having a dramatic effect upon society
5.) Legitimately free music/whatever as advertising is nevertheless a valid business model to gain popularity.
6.) Artists are, in fact, getting ripped off due to the perceived need to cut a record deal to "get known." They would be much better off thinking like entrepeneurs.
7.) "Giving stuff away on the Internet" is not a business model, but it can be part of one if done correctly. Look at Homestar Runner as example: free cartoons that got so popular that the authors now make a living selling plush dolls, t-shirts, and bumper stickers. It would never have succeeded as a pay-for-content site because it has to compete with Cartoon Network, the Simpsons, and the like..
8.) Not everyone is looking for a free ride. The fact that people are more than willing to pay for concert tickets but many now hesistate to buy CDs says more about the market than morals. People are simply putting far less value in recorded music.
9.) In a purely capitalist, laisez faire economic system, there is no such thing as copyright. It's not an assumption or requirement. That's not to say that it's always bad, but rather that there are plenty of natural ways to make money that do not involve artificial government institutions. Open Source has already succeeded in this field; independent music/film is still on its way.
10.) For the majority of human history, it was a right. Copyright is a modern experiment. It may or may not last long term. My guess is that a fairer balance will be struck.
11.) What signifies greed is the motivation, not that they are exercising their legal rights. Numerous studies have shown that P2P and other bootlegging has a minimal effect on profits, while significantly expanding the spread of content. It is more likely that the 'cracking down' is more out of fear that they are losing control of the traditional distribution channels.
One of the unfortunate things that has happened to the OSS movement is that a lot of the loudmouth advocates for it don't understand what it's really about.
Absolutely. In my definition, Open Source is about meeting software needs in the most efficient way possible. That does not always mean a free ride. Open Source is about turning an artificial "manufacturing" market into a labor market, the latter of which allows full, unrestricted motion of the "invisible hand of the market." Capitalism works best with many buyers, many sellers, and minimal cost of entry. That is what Open Source enables.
They view it primarily as a means to get free stuff, and then they turn their eyes from the free stuff to the non-free stuff and think to themselves "maybe I'm entitled to get that one for free too"
While I agree that many mistakenly see OSS as "free lunch," I don't see your secondary point in any true OSS advocates.
Hello... that's the worldwide standard for plain sound cards, one that for years many vendors other than Creative followed.
False. "Sound Blaster" was never a standard, and though it was popular back in the days of DOS games, it is totally irrelevant today. AC97 is the current standard for generic PC soundcards. Regardless, the fact that this editor couldn't get some VM hack to work doesn't mean Linux has poor sound support.
You can do this with XFree86 using any commodity hardware.. two video cards, two USB keyboards and mice. *yawn* And unlike with Windows, it's not a hack and there aren't licensing issues. (-:
I have yet to see the uncrackable DRM scheme, and no reason to assume one can ever exist.
Lets be even more to the point: It is physically impossible for there to exist a non-proprietary DRM scheme. DRM, by very nature, is security through obscurity -- hiding the secret key *somewhere* on the users' own machine. DRM is incompatible with the concept of general purpose computing and most certainly with Open Source software. As an analogy, think of those old pay-TV decoder boxes that were filled with epoxy or tar to prevent tampering.. Is that what you want your computer to become? Just say no to DRM-laden hardware, software, and content! And do support independent media!
Those are fairly rigid requirements and if you are only 99% of the way there then you simply don't have a patent grant..
This is a very important point that seems to have gotten no discussion. What are these supposed Java2-related patents? Would they even stand up in court? And would Sun ever try to use them against Open Source implementations like GNU ClassPath, which are incomplete and/or untested implementations of the specification? Maybe the ClassPath people have already found ways around the patents such that the specification license can't be held over them. Or maybe the patented components can be left out and we can settle for partial compatibility. Regardless, we need answers. (See section 7 of the GPL.) If we cannot conform to the Java specs because of stupid patents, we will need to: a.) fork into our own mostly-compatible language of a different name b.) convince Sun to fully release these patents or c.) forget about Java altogether.
Oh.. that and we need to get software patents ruled unconstitutional in the US (which they are because they do not promote the progress of the sciences and useful arts and also possibly violate freedoms of speech/expression)
I find this highly suspect, though I don't know for sure that it's untrue. If this is the case, why doesn't the FSF regard it as a Free Software license (although, like BSD, obviously not a Copyleft license)? Why hasn't the OSI certified it as an Open Source license? Why isn't it included in Debian?
It is my understanding (IANAL), that even viewing the Sun source code taints you in a NDA sort of way. If you view it, you cannot then work on Open Source clean-room implementations of the JVM and class libraries. This sort of legal restriction is considered dangerous and makes it non-free by Debian (and most other) standards of Free Software. Specifically, this is a restriction of use, which is considered never acceptable. GPL contains no restrictions on use, only redistribution.
Secondarily, you are not free to distribute modified versions of the Sun code without their testing and approval. This violates another component of the widespread definition of Free Software -- the license cannot require permission for redistribution. Instead, any requirements for redistribution are covered in the license itself. With GPL, you do not have to ask permission to redistribute -- you must only agree to the terms of the license, which are quite reasonable.
As for Debian, have a closer look here: http://www.debian.org/doc/manuals/debian-ja va-faq/
In this case, if Java was licensed in GPL, all software written would also have to be licensed in GPL.
Which is why the option of LGPL exists.. or GPL plus some extra exemptions.. or other similar licenses that force freedom of the libraries but allow linking by that which is non-free.
The solution is to provide motivation to write for someone else.
Precisely. What we need is a grassroots commercialization of Open Source software. I say this not to discount the work done by large corporations with a stake in the future of various software, Linux most visibly. However, because Open Source works best within its distributed and community nature, it needs (as a movement) strong roots in smaller, local, highly-efficient, privately-owned businesses. What does this mean in plain English? We need more people who provide real solutions to real companies by making sure that Free Software meets real needs -- full time "consultant-developers" if you will. And what about when these consultant-developers cannot do all the coding work themselves? They need to be able to funnel money from their clients into Open Source projects in exchange for focused and timely development. This way, the larger (non-hobby) projects can afford to have full-time developers that prioritize feature development based on demand. Yes, you heard that correctly: Open Source and Capitalism are more than just compatible. They need each other desperately.
Tired of your IT job? Get out there and innovate. Be entrepreneurs or find others willing to take the risks with you. Stop sitting around and complaining. Go make a difference.:)
Slashdot Mirror
Red Hat is modifying the kernel, don't read the headlines? Although it won't effect the userland, try apply many existing patches against the existing Red hat source code. There will be problems.
Exactly.. won't affect userland. Which is what I was talking about. I never said distros don't patch the kernel. On the other hand, you can always plug in a virgin kernel without issue.
It's fragmentation. There's no reason why KDE couldn't have used GTK. But they didn't, and now programs are less portable.
It seems by your definition, the simultaneous existance of C, C++, and Objective-C is an example of fragmentation. If so, that's patently silly. KDE didn't choose GTK because at the time it was vastly inferior to Qt. Fragmentation would be if KDE had split off of the GNOME project.
You admit it's fragmentation, good. You're learning.
I never said that smartass. C/C++ doesn't have the level of standardization that Java does because of its history. Modern languages do and Open Source implementations of those languages are on spec with high precision.
Then you're a fool. There are millions of reasons why Java will fragment if it goes open source. The C standards are solid standards, but GNU made extensions anyway. Open source developers do what they want, standards or not.
The C/C++ standards are *now* becoming solid, but they have an enormous history of variation. Some things that are "traditional C" are not part of the ISO standard and this is why GCC supports them unless you turn off support.
Java already is going Open Source and it's not fragmenting yet. Get your head out of the sand. A forked Java is not in anyone's interests and there's nothing to suggest it will ever happen. It very easily could have by now, but it hasn't. And I'm talking about all the proprietary, non-Sun implementations even.
Open source zealots would choose an open version over Suns clean-room version. It's what you people do. Sooner or later, the Sun version isn't mainstream.
You mean: Open Source advocates will choose their own open clean-room version because its the only one they can legally redistribute and it's just as good if not better than the proprietary versions. Sun's version won't be mainstream too much longer if they don't find a way to participate in the inevitable change.
IBM, for example, could fork the code and really open up to the community. The entire open source community gets behind the IBM version and, after awhile, Sun disappears.
IBM's code is based on Sun's. They can't legally open it. Now, suppose Sun was to GPL their implementation. IBM could participate openly, but they still wouldn't fork because it would break all compatibility and everybody would cry foul.
Kaffe/GNU ClassPath are nothing like java.
Uh.. they're everything like Java. They're just behind on implementing the newer specs.
They do not run on every platform supported by Java nor do they have all of the GUI functionality.
Even Sun's JVM doesn't run on every platform supported by Java. (-: As for GUI stuff, it's still in the works, though SWT will probably be possible before AWT or Swing since it's already open.
I'm sorry, but having to install/configure additional compatibility libraries on each platform that is non-linux does not count.
Invalid argument. You're criticizing a young but rapidly growing project for not being what you want today. Who cares. Wait another few years and keep using the official Java implementations until then.
To most people, Kaffe and GNU Classpath do not exist. Sorry to break it to you.
Only because they're not completed yet. I don't use them yet and I never said it was possible to use them in place of the complete Sun/IBM Java runtimes today.
Yes. A certain vendor that did itself try to fork Java, and now has a competing product that is extremely similar to Java.
.NET and CLR. MS is really out of this picture. Oh, one more thing: they won't touch anything GPL with a ten foot pole because that would only further legitimize it to those on the fence.
What this proves is that it doesn't take Sun Open Sourcing their own implementation for a fork to occur! And yet, the existing Open Source implementations of "Java" are not forking from the spec.
If Java went GPL, this company would not hesitate to muck around with it, just because they can, and the company has a history of doing anything it can to 'win'.
How exactly could they muck around with a GPL (or similarly licensed) Sun Java to win? They couldn't release proprietary versions. They couldn't make changes without telling everyone. They couldn't completely fork the code and still call it Java (Sun trademark). And they don't even need Java anymore since they have
Sometimes yes, sometimes no. I would not expect Java to fork in the classic sense of the word (except the version that one particular company would make), but look at Linux as a whole; multiple distros out there, by companies, by individuals, by governments. That kind of "forking" alone would make it more difficult for WORA with Java; it'd increase the amount of tweaking neccessary. Every distro tweaking the JVM a little differently than the others... code then running great on some and flakey on others...
Linux distros are not a proper comparison because they are a conglomeration of very diverse software. Here, we're only about talking the JVM and a set of libraries. There's no incentive to fork and any 'tweaking' would only be within the implementation of the specs. With Sun still holding the trademark, they would still be the sole source of endorsement. So "Bob's J*** Derivative" could fork and tweak to heck and back, but typically only experimenters would use it. And if they came up with something really great that SCP approved of.. cool! roll it back into the official codebase that normal people use.
well, maybe that's already true. Why make it worse, though?
Because there's virtually nothing to lose and a lot to potentially gain.
I could see the same easily happening with Java; different bright folks wanting to take it in different directions, without a leader, ending up disagreeing too strongly and going it alone. Would Sun and IBM and Microsoft and Oracle and so on always agree on the next level to take Java to? I doubt that.
Sun would still be a strong leader, even if they Open Sourced their implementations. Weak leadership or poor quality is what encourages forks. As for IBM and Oracle contributing, that's what the SCP is for. It works and is well established. Why would anyone want to diverge? All that would do is make them look bad.
For the record, I'm undecided about the opening-up of Java.
Java is already being opened-up. (http://www.kaffe.org/links.shtml) It's just a matter of how long that takes. If Sun helped out, it would happen a lot sooner and dramatically increase their influence over the whole 'open Java' scene.
The Linux itself is fragmented across distros. Try taking a Red Hat binary to SuSE, there will be trouble.
This is an improper example. Try taking some C source and compile it on both RedHat and SuSE -- same results. Fragmentation would be if RedHat or SuSE modified the Linux kernel or supporting GNU software so much that code would need significantly re-written to work.
There are now 3 projects who will use the X windows standards, and I can tell you there will be fragmentation. Each will try to improve it somehow and bye bye compatibilty.
Not even close. Read this: http://www.xouvert.org/faq.html#2.1 It should be especially noted the strong emphasis on sticking to standards. There are many X11 implementations and they are very much compatible, though some are at different levels of specification advancement.
There may be one GNOME and one KDE, but each use a different GUI toolkit resulting in the same effect as fragmentation.
This is an example of diversity, not fragmentation of an original standard. Would you say that there should only be one programming language? I should hope not. Each has its advantages. It should be noted, of course, that there is a growing movement to harmonize GNOME and KDE, but that's another topic of discussion.
GCC has it's own extensions, as does the GNU C library. Same goes for gmake. You can't compile software that use these extensions on Solaris using native compilers.
It can be easily argued that the only reason C compilers have extensions like this is because there does not exist as active a standards body as say, W3C or SCP, to recommend changes before they are tried experimentally. There are also different ISO C versions. I believe some of the GNU C "extensions" are actually part of the C99 ISO spec. Solaris's native compilers may be using an older spec, but I'm not familiar. Also, you can use the -pedantic option with gcc and it will warn you of any code that uses extensions beyond C89. There is little to make me believe a similar scenario will exist with Open Source Java. The industry has learned since C/C++ the value of creating standards early and then sticking with them. As I said, Open Source developers tend not to deviate when solid standards exist. Java is a solid standard.
No, it would only hurt Sun and Java.
How much have GNU ClassPath and the various open JVM's hurt Sun and Java? And how much have they deviated from Java specifications? The answer to both is: not at all. If Sun relaxed licensing on code for their implementations, it would only serve to promote their own over the clean-room versions.
How does BSD provide less freedom? Are we talking about freedom of the code again? I'll let you on a secret. People can have freedoms and rights. Objects and information do not have rights. People can own information and inanimate objects.
:D
BSD does not insure that what people create and then desire to be free, remains free. There are both philosophical and commercial reasons why this "restriction" can be desireable.
And, by the way, people cannot own information. They can only be granted a temporary monopoly on the use of information within the borders of a soverign state that grants such monopoly. Big difference philosophically and legally speaking. And, in case you're wondering, this viewpoint is fully backed up by the US Constitution and legal system. It's not "GNU/Koolaid" or hippie-speak.
How do you expect Java to be open sourced without Sun's consent? It is the IP of Sun. Careful about calling people idiots when you make statements about Sun being forced to do something with their property.
Sun's Java implementations are copyrighted by Sun. Clean-room Open Source like Kaffe / GNU ClassPath are not. (and legally they're not even Java(tm) but they effectively they are) Nobody ever said anything about Sun being forced to do anything with their own code. It would, however, be mutually beneficial if they released as much as possible under a freer license instead of fighting what they cannot stop.
Open sourcing Java under any license would be a mistake. Numerous forks of the code would occur completely destroying the "write once, run anywhere" nature of java. Maybe you don't understand that not everyone runs either Linux or windows on their desktop (not to mention phones, set top boxes and PDAs).
This is 100% pure recycled FUD. Since when do large, successful Open Source projects fork? Even more so, since when do Open Source projects go against established industry specifications? If Sun GPL'ed their Java code and somebody tried to fork it to be a jerk, nobody else would pay attention. It's entirely against the community ethos to fork without significant cause or out of spite. Practically speaking, large projects have gravity regardless.
As a programmer of in-house proprietary software, I'm painfully aware of forking issues. At times we have two developers working on two future versions and releasing patches to the version in production.
That's not forking, it's a management problem. Forking is when somebody purposely takes all the code and runs in a different direction.
Yes, there be ports to other platforms (I'd love to see Java on *BSD). Yes, there'd be performance patches and probably security patches. But with this comes the down side to all open source projects: Fragmentation.
Open Source fragmentation is a myth largely propagated a certain proprietary vendor's FUD campaign. Successful open source projects simply do not fork. Why? Because there's no interest and it's way too difficult with large scale software. There is one GNOME, one KDE, one Apache, one Linux kernel, one GNU Compiler Colection, one OpenOffice, etc. Similarly, an Open Source Java would be a large project with so much momentum that any unlikely attempts to fork would be rapidly ignored and wither away. Not to mention Sun would still hold the Java trademark..
One of Javas goals is to provide compile-once-run-anywhere. At the moment, it does a decent job of doing this. Other platforms do this better (Read: Inferno) but Java is trying. With fragmentation, compile-once-run-anywhere disappears and would be impossible to bring back.
Wrong again. Open Source developers are terrifically anal about sticking to standards. You don't see a dozen different C compilers each with a slightly different dialect do you? You don't see KHTML or Mozilla trying to extend web standards do you? Even more obvious: You don't see the existing clean-room Open Source Java projects deviating from Sun's specifications do you? -- and they're not even under legal obligation!! Fact is, an Open Source Java is the only way WORA can ever hope to fully live up to its promise. (btw, that's Write Once, not Compile.. minor correction) Three reasons: 1.) more ports 2.) better quality control 3.) less Java implementations floating about, whether proprietary or open.
Let Sun has Solaris and Java. Stop trying to force them to open source their software.
Nobody's trying to force Sun to do anything. They're simply asking because it would be mutually beneficial.
I think that there is exactly one person who thinks that Java should be GPLd.
As opposed to what other license? (L)GPL would probably be the best choice for Sun because it ensures that people cannot create proprietary forks (Microsoft anyone?). BSD or similar less freedom-guaranteeing licenses would be an awful choice. And if you think Java shouldn't be open source at all, well.. you're just an idiot. It's going to happen whether Sun wants it to or not. This whole discussion is only a question of whether they remain a player.
Open source is not a career. I don't know who told you it was but you need to disabuse that notion.
Is that why I make my living doing Open Source consulting and paid development? Darn, guess it's not a career.. I'll have to do something else now. Thanks for enlightening me, but my clients will be very disappointed.
So, GIMP developers, clean up the interface and change the product name, and your program has a decent chance of seeing the light of day in the real world.
You forgot one thing: Set up a way that ordinary users (or even companies) can donate to the development of certain features / interface overhauls. Gimp is a hobby project. If people expect it to become more professional (which they clearly do by the tone of this article and the followup posts), the project is going to need one or more full time, paid programmers. It won't happen otherwise -- at least not within a reasonable timeframe. Its really as simple as that.
I would almost guarantee that if every Photoshop user chipped in $5, Gimp would put Photoshop out of the market with 2 years. Why would anyone help pay for something that's free? Well, how much does a Photoshop upgrade cost every year or so? The only problem is how to efficiently advertise a fundraising drive...
Define "isn't moving faster". It's moving a lightning speed. It's being adopted by all major IT companies except one. Even the unmentionable company is flirting with pseudeo OSS. Exactly how fast were you expecting it to move anyway?
.NET, VB, and MCSE. Open Source will be working when it represents the ethos of almost every IT department.
I don't share your optimistic appraisal on the "lightning speed" bit. Yes, OSS is moving faster than it ever has before, but it's not enough. It's certainly not a revolution yet. Sure, there are a lot of large, high-profile companies utilizing Linux in one form or another. (and note the distinction.. usually Linux, not OSS) Yes, there are a lot of smaller companies using Linux / OSS in some capacity to meet some of their needs. But the fact is, OSS has the potential to grow exponentially and it's not. It's not for one reason: not enough people are being paid to do full-time OSS development. Why is that? Because too many geeks treat it like a toy and not a career. You can't expect change when you desire OSS but then go work a day job writing proprietary software. For all the brains in the OSS community, there sure is a lack of business intelligence in exploring the many ways to make a career out of supporting or developing Free software.
Open Source will be working when companies worldwide are ditching Windows / Office and other common proprietary desktop software en masse. Open Source will be working when collaboration among consultants to write modular business software is the norm. (think 'Apache') Open Source will be working when most job descriptions drop the acronyms
Open source is _not_ about competing with anybody else. It is _not_ the goal to create a competing technology to win market share or anything. Open source is an exercise in technology, invention and freedom.
That stupid philosophy is precisely the reason why OSS isn't moving faster and providing more geeks with jobs writing Free code rather than proprietary. Yes, OSS is an exercise in technology, invention, and freedom. But, at the same time, that exercise is worthless if it doesn't significantly improve the state of the industry / society. To do that, it must garner market share because that is the only way to diminish the forces that work against freedom. Do you quite realize that if OSS does not dominate the industry in the fairly near future, the powers that be will effectively make it illegal to write software unless you work for a large corporation with patent portfolio cross-licensing? Do you quite realize that an industry dominated by OSS would be a much more pleasurable one to work in? There is far more at stake here than academic toys and utopian musings. Get out of your bubble.
I really hope this isn't serious... how exactly did you plan to get very far in a field you have no formal education in?
Sorry, but that's an extremely closed-minded way of thinking. While I agree that most everyone should complete at least a 2-year program in some area of study, simply for the social side of things, there are plenty of accomplished programmers without a formal CS background. And, likewise, there are plenty of people with formal CS backgrounds that are truly incompetent at their jobs. It all depends on how you personally learn best and what career path you intend to take. No, you won't likely get a big corporate IT job without a CS degree, but not everyone wants such a job. (insert picture of Dilbert)
Here's one possible idea: Become an expert at J2EE and do Open Source based consulting. In a down economy especially, companies are looking to cut their IT costs while still using proven technology. You don't need a degree to hire yourself, and you can certainly do better than $13/hour. If you like, take some classes on the side in something that interests you--business administration maybe? Of course this route is not for everyone. You need dedication and an entrepreneurial mindset to some extent.
How can they possibly say it's a potential good fit for smaller companies?
/.'ers) stance.
/. these days. I call your karma whoring. (-:
Small companies are able to adapt quickly, so they are the first candidates for cost-saving tools. Large companies, in comparision, cannot adapt as quickly, but *can* invest in making improvements to Open Source software that is almost but not quite useful to them (such as to save money by removing the need for the proprietary standbys)
- It doesn't work for advanced Excel (read: The Finance Department).
So have everybody else use OpenOffice and let the finance people keep their existing Excel. There's also Gnumeric, but I'm not sure what the comparison is as of late..
And of course, finance data should be kept in a database anyways, not in spreadsheets. C'mon, this is 2004, not the 80's. But that's another discussion..
- Support options are limited (read: DIY in a small company with limited/nonexistent IT resources to begin with).
How often do people need "support resources" for an office suite? And since when does MS Office come with amazing support resources for small customers?
- It takes as much as 10 seconds longer to open big docs sent in Office format (read: anything sent to you most people outside the company).
This is so trivial it's not even worth mentioning.
And, let's overlook Outlook in the comparison. (Evolution, Thunderbird, et. al. do not offer the same functionality)
I assume you refer to Exchange, not just Outlook itself. There are plenty of alternatives to using Exchange/Outlook such as OpenGroupware and Kontact. And beyond that, Web-based groupware solutions are superior anyhow in most cases.
Oh, and feel free to mod me into oblivion for taking a controversial (for
What are you talking about? Controversial stances, regardless how stupid they are, get modded up on
While the implications of this treaty are truly frightening, the amazing thing about it is that it originated in Europe.
, 00 .html
c s/gove rnment/story/0,10801,55949,00.html
It's not really so amazing when you consider that the Clinton admin, which also brought us DMCA through the backdoor of a WTO treaty, was largely responsible for drafting / pushing the cybercrime treaty as well.
http://www.wired.com/news/politics/0,1283,40576
And here's a fun one:
http://www.computerworld.com/governmenttopi
Oh how times have changed, eh?
Sadly none of this stuff will be discussed in either party's presidential candidate's 2004 campaign. Why? Because they basically have the same brain-dead stance. So you all know what to do: start writing congress immediately!
In my opinion the more important issue is submarine patents
Even more important is the fact that patents are being granted for trivial things undeserving of a patent. As example: anything software related. Software is an evolutionary technology -- all improvements to the state of the art are, by themselves, merely logical next steps and not true innovation. Implementation is where the software industry innovates, and that is covered by copyright. The reasons why patents should not be granted for software are much the same reasons why patents are not allowed for mathematics or styles of literature, art, and music.
Like a Epson Powerlite S1 or an Infocus X1. The picture will be infinitely better than what you'd get from a homebrew. Don't even consider the homebrew thing. Really. I've seen a few (based on CRT's or overhead projectors and LCD panels) and they are really poor.
I second this opinion. The homebrew thing is not worthwhile today given the cost of the latest round of commercial projectors. Two years ago, when I started tinkering, it was feasible to attempt. Low-end projectors then cost 50% more for significantly worse performance. Today, you will have to spend $600-700 to almost match the performance of a $800-900 entry-level commercial DLP projector like the X1 or S1 mentioned. Don't be fooled by sites that claim it only costs $300-400. That is only for the specialized bulb/ballast, lenses, and cabinetry required. You're looking at another $250+ for a halfway decent LCD screen that you can remove the backlight from and project through. Those old projection panels you can find on eBay are crap. (Proxima, iView, etc.)
Trust me. I'm a hard core DIYer and I've tried the homebrew LCD projector thing. It's not that it doesn't work at all, but this is a case of inadequate technology available to the home experimenter to match commercial products. It's simply not worth your time and efforts. If you want a good project, build a "Sonotube" subwoofer. That will save you lots of money *and* give you outstanding results. Once you get a projector, you can make your own custom screen that fits nicely with your HT room. That's another worthwhile money-saver.
Certainly there's some truth to that. I was simply making the point that assumptions should be avoided. Of course, a counterpoint to your scenario would be that most sane societies, in either case, would have laws prohibiting any sort of dangerously "booby trapped" devices, whether sold commercially or given as a 'gift.' In the end, it all comes down to a balance between encouraging a vibrant economy and ensuring basic rights, freedoms, and safety. Often, those two goals actually go together.
You have some good points, but some things should be clarified..
3.) Most CD's are quite overpriced and the public now realizes this. Allow me to plug Mangatune.com Reasonable price, actually supports the musicians. (:
4.) Copyright duration is way to long and it is having a dramatic effect upon society
5.) Legitimately free music/whatever as advertising is nevertheless a valid business model to gain popularity.
6.) Artists are, in fact, getting ripped off due to the perceived need to cut a record deal to "get known." They would be much better off thinking like entrepeneurs.
7.) "Giving stuff away on the Internet" is not a business model, but it can be part of one if done correctly. Look at Homestar Runner as example: free cartoons that got so popular that the authors now make a living selling plush dolls, t-shirts, and bumper stickers. It would never have succeeded as a pay-for-content site because it has to compete with Cartoon Network, the Simpsons, and the like..
8.) Not everyone is looking for a free ride. The fact that people are more than willing to pay for concert tickets but many now hesistate to buy CDs says more about the market than morals. People are simply putting far less value in recorded music.
9.) In a purely capitalist, laisez faire economic system, there is no such thing as copyright. It's not an assumption or requirement. That's not to say that it's always bad, but rather that there are plenty of natural ways to make money that do not involve artificial government institutions. Open Source has already succeeded in this field; independent music/film is still on its way.
10.) For the majority of human history, it was a right. Copyright is a modern experiment. It may or may not last long term. My guess is that a fairer balance will be struck.
11.) What signifies greed is the motivation, not that they are exercising their legal rights. Numerous studies have shown that P2P and other bootlegging has a minimal effect on profits, while significantly expanding the spread of content. It is more likely that the 'cracking down' is more out of fear that they are losing control of the traditional distribution channels.
One of the unfortunate things that has happened to the OSS movement is that a lot of the loudmouth advocates for it don't understand what it's really about.
Absolutely. In my definition, Open Source is about meeting software needs in the most efficient way possible. That does not always mean a free ride. Open Source is about turning an artificial "manufacturing" market into a labor market, the latter of which allows full, unrestricted motion of the "invisible hand of the market." Capitalism works best with many buyers, many sellers, and minimal cost of entry. That is what Open Source enables.
They view it primarily as a means to get free stuff, and then they turn their eyes from the free stuff to the non-free stuff and think to themselves "maybe I'm entitled to get that one for free too"
While I agree that many mistakenly see OSS as "free lunch," I don't see your secondary point in any true OSS advocates.
Hello... that's the worldwide standard for plain sound cards, one that for years many vendors other than Creative followed.
False. "Sound Blaster" was never a standard, and though it was popular back in the days of DOS games, it is totally irrelevant today. AC97 is the current standard for generic PC soundcards. Regardless, the fact that this editor couldn't get some VM hack to work doesn't mean Linux has poor sound support.
You can do this with XFree86 using any commodity hardware.. two video cards, two USB keyboards and mice. *yawn* And unlike with Windows, it's not a hack and there aren't licensing issues. (-:
I have yet to see the uncrackable DRM scheme, and no reason to assume one can ever exist.
Lets be even more to the point: It is physically impossible for there to exist a non-proprietary DRM scheme. DRM, by very nature, is security through obscurity -- hiding the secret key *somewhere* on the users' own machine. DRM is incompatible with the concept of general purpose computing and most certainly with Open Source software. As an analogy, think of those old pay-TV decoder boxes that were filled with epoxy or tar to prevent tampering.. Is that what you want your computer to become? Just say no to DRM-laden hardware, software, and content! And do support independent media!
Those are fairly rigid requirements and if you are only 99% of the way there then you simply don't have a patent grant..
This is a very important point that seems to have gotten no discussion. What are these supposed Java2-related patents? Would they even stand up in court? And would Sun ever try to use them against Open Source implementations like GNU ClassPath, which are incomplete and/or untested implementations of the specification? Maybe the ClassPath people have already found ways around the patents such that the specification license can't be held over them. Or maybe the patented components can be left out and we can settle for partial compatibility. Regardless, we need answers. (See section 7 of the GPL.) If we cannot conform to the Java specs because of stupid patents, we will need to: a.) fork into our own mostly-compatible language of a different name b.) convince Sun to fully release these patents or c.) forget about Java altogether.
Oh.. that and we need to get software patents ruled unconstitutional in the US (which they are because they do not promote the progress of the sciences and useful arts and also possibly violate freedoms of speech/expression)
I find this highly suspect, though I don't know for sure that it's untrue. If this is the case, why doesn't the FSF regard it as a Free Software license (although, like BSD, obviously not a Copyleft license)? Why hasn't the OSI certified it as an Open Source license? Why isn't it included in Debian?
a va-faq/
It is my understanding (IANAL), that even viewing the Sun source code taints you in a NDA sort of way. If you view it, you cannot then work on Open Source clean-room implementations of the JVM and class libraries. This sort of legal restriction is considered dangerous and makes it non-free by Debian (and most other) standards of Free Software. Specifically, this is a restriction of use, which is considered never acceptable. GPL contains no restrictions on use, only redistribution.
Secondarily, you are not free to distribute modified versions of the Sun code without their testing and approval. This violates another component of the widespread definition of Free Software -- the license cannot require permission for redistribution. Instead, any requirements for redistribution are covered in the license itself. With GPL, you do not have to ask permission to redistribute -- you must only agree to the terms of the license, which are quite reasonable.
As for Debian, have a closer look here:
http://www.debian.org/doc/manuals/debian-j
In this case, if Java was licensed in GPL, all software written would also have to be licensed in GPL.
Which is why the option of LGPL exists.. or GPL plus some extra exemptions.. or other similar licenses that force freedom of the libraries but allow linking by that which is non-free.
The solution is to provide motivation to write for someone else.
:)
Precisely. What we need is a grassroots commercialization of Open Source software. I say this not to discount the work done by large corporations with a stake in the future of various software, Linux most visibly. However, because Open Source works best within its distributed and community nature, it needs (as a movement) strong roots in smaller, local, highly-efficient, privately-owned businesses. What does this mean in plain English? We need more people who provide real solutions to real companies by making sure that Free Software meets real needs -- full time "consultant-developers" if you will. And what about when these consultant-developers cannot do all the coding work themselves? They need to be able to funnel money from their clients into Open Source projects in exchange for focused and timely development. This way, the larger (non-hobby) projects can afford to have full-time developers that prioritize feature development based on demand. Yes, you heard that correctly: Open Source and Capitalism are more than just compatible. They need each other desperately.
Tired of your IT job? Get out there and innovate. Be entrepreneurs or find others willing to take the risks with you. Stop sitting around and complaining. Go make a difference.