True, but companies dont get to take out the space shuttle for a spin either.
What about Dennis Tito? He pays for part of the russian space budget in exchange for a trip into space. Companies also pay for getting their experiments on the space shuttle. This reduces our taxes.
In the case is this software, there is an inequity in the availability of the software for the taxpayer who paid for it and the corporations who didnt.
Wrong. You never paid for the use of the software. You paid for the stuff they do with space shuttles, rockets, etc. The software is just a byproduct. Companies sponsor Nasa by paying money for this byproduct. Doing so, they lower the taxes you have to pay (or increase the value you get for the taxes).
You pay Nasa for exploring space and to boldy go where no one has gone before. You don't pay them to provide you with the software they use to achieve this goal (software for controlling rockets, shuttles and other stuff that I'm sure you don't have). Selling the software to private parties means that they have more money available to achieve their primary goal. This does benefit the taxpayer.
Or perhaps you would rather pay more taxes because Nasa can't sell their software? In return you'd get all the software to control your own space shuttle for free. Wouldn't that be great?
So in the best case, as an American, I should get full benefit out of the software.
I agree that the software should be used in a way that serves our best interest. But that doesn't mean that open sourcing is necessarily the best way to do that. A lot of government software is specialized enough that few people can use it. The options the gov has in that case:
Keep it to yourself. Has the big advantage that you've got no extra costs. The taxpayer gets the benefit for which this software was written.
Open source the code. This means extra expenses to get it in shape by removing proprietary stuff from other companies, making it less dependant on the specific environment in which the government organisation runs it, making sure there are no references to Roswell and other secret stuff, etc. Those extra costs will bring no benefit to 99.9% of the taxpayers.
Sell the code if there is interest. This pays for new projects and thus benefits the taxpayer.
I'm not convinced that the second option is always the best.
Second, the widespread belief among open source types like myself is that the more eyes that are on the code, the more likely the vulnerabilities will be discovered and ultimately patched.
I'm sure that thousands of developers will eyeball the code (written in Ada or another relatively obscure language) to control the space shuttle's robot arm. Everyone will surely test the code on their own space shuttle. There is a big difference between Linux and a piece of highly specialized software. You can't expect much from these code reviews. Especially since Nasa-software is already extensively reviewed by experts.
It's extremely stupid that Windows tries to read your disks when they are removed. It has irritated me often. MacOS handles that much better by actually removing open windows/drives when you eject a storage device. That isn't possible with a hardware-controlled eject. Learning to reboot with the mouse-button down isn't that hard in those rare cases that a CD won't eject.
PS. What does the key sequence for booting have to do with this (pressing 'c' on my 1997 G3)?
Extreme freedom is the public domain. A bit less is the BSD. A lot less is the GPL. Or do you deny that the GPL limits your ability to distribute software?
They still have to do some sort of mark and sweep algorithim, which is not fast.
A 'complete' mark and sweep is not fast (one that doesn't miss anything). Eden uses a fast, but incomplete algorithm. That destroys all the easy temporary objects (and moves the rest to long term storage which gets the full GC). I don't see why your approach would be so much faster.
The feature that I want to see added to java is that I want to be able for the java vm to create objects on the stack. Then if they get assigned to a member pointer, get promoted to the heap. This way you could keep objects on the stack where they could be easily collected when the function that created them ends scope.
"A generational collector takes advantage of the fact that in most programs the vast majority of objects (often >95%) are very short-lived (i.e. are used as temporary data structures). By segregating newly created objects into an object "nursery", a generational collector can accomplish several things. First, because new objects are allocated contiguously in stack-like fashion in the object nursery, allocation becomes extremely fast, since it involves merely updating a single pointer and performing a single check for nursery overflow. Secondly, by the time the nursery overflows, most of the objects in the nursery are already "dead", allowing the garbage collector to simply move the few surviving objects elsewhere, and avoid doing any reclamation work for dead objects in the nursery."
(* Java is simpler and clearer than C++. Or do you disagree? *)
Roughly 98% of all languages are. That is like bragging that car crashes are less painful than nukes.
Java succeeds in replacing C++ for quite a few tasks, that's the big difference with most languages in that 98%.
(* You can create fast code with a VM! *)
Yes, but it complicates the VM engine. You are lucky Sun is footing the bill. A compiler would be simpler. Write the compiler in C, so that porting it would be relatively easy.
People write compilers for Java. It seems just as complicated as they have a hard enough time just getting it to work on x86. Besides, who cares? It seems to me that you are changing the subject. I remember this being about programming in Java, not writing VM's.
(* Multi-paradigm means chaos then. *)
Okay, then get rid of the OOP. Even many OO fans that I debate with admit that sometimes they mix in procedural solutions as they see fit.
I don't mind having to use another tool for something. I never claimed that Java is perfect for everything. I do mind when someone tries to create a frankenstein language that is everything to anybody. Perl and C++ for instance. Yuck.
I don't think it would "ruin" the language if Java had real functions.
That's not very helpful. Do you want object-independent functions (func(x) instead of x.func() or staticClass.func(x))? Or are you talking about operator overloading (ComplexType + ComplexType)?
(* Do note that we are generally moving away from efficiency towards clarity and simplicity. *)
How do you explain that sh*t Java then?
Java is simpler and clearer than C++. Or do you disagree?
My point was that if it was compiled, it would have an edge over a VM.
You can compile it. But the report (and this one) shows that the compilation vs VM issue is quite irrelevant. You can create fast code with a VM!
Where did I say that Java sucked because that writer was stupid? It is both. Java AND that writer suck.
You put it on a page that is called: "Why Java Sucks". Put it on a seperate page if it's a seperate topic.
(* blaming the language for having flaws that are not explained (what is wrong with the reflection in Java?) *)
It is too convoluted. There is plenty of criticism of it on the web and ng's. Go look.
I know it has limitations. The point is that you should explain your claims. Complaining that X is lousy without giving proof or even telling people why it sucks will make me ignore it. Who are you trying to reach with your critique? Programmers that already know all about Java's reflection limitations (only an issue to experts really)? If they already know the limitations, what good does your page do? Your critique is pointless.
What is wrong with being multi-paradigm? C++ can be multi-paradigm.
Multi-paradigm means chaos then. I prefer clean design, not Perl or C++.
Only zealOOts think one paradigm fits all.
I don't remember saying that Java needs to replace all competing languages. I do think it's a very good one. BTW, you can use various scripting languages with Java. You can also call native code if Java doesn't cut it.
How hard would it be to add functions to Java anyhow?
I read your story: 'There wasn't enough room for comments, so we hacked...'. You then complain that your hack doesn't work.
Further, the current EDI networks are still more reliable and secure than web-based solutions. Many big-name companies don't want to switch until such issues are solved on the web.
Big companies can pay for a subscriber line to their suppliers and/or customers that circumvents the internet. Small companies just have to to take the risk I'm afraid.
Nope! XML takes up more bandwidth than some other solutions because it repeats field/tag-names for each record. I realize that this is generally not a problem, but it does show that you are wrong.
I don't see how this is related to translating. If there is a problem with bandwith, you can simply compress the XML. Do note that we are generally moving away from efficiency towards clarity and simplicity.
And how the heck does the VM contribute to "fast development" and "good features", and "big libraries"?
I wasn't simply talking about the benefits/downsides of a JVM (I never intended to narrow the discussion, the JVM vs P-code was just an example of Java not being incredibly revolutionary). Useful things are usually not revolutionary BTW. They evolve from useless revolutionary technology (which is too complex or slow to be useful). I do claim that all the features of the language (some of which are due to the VM) are good enough to offset the slowless.
I want to refer you to this excellent report on the performance of Java. Some highlights:
"Java has a strong reputation for being a slow language that cannot be seriously considered for real applications. It has not gotten this reputation by coincidence. If you are a C++ programmer then you were probably very tired of all those Java evangelists back in 1995 that claimed that Java was superior in every regard and that 100% pure Java was the best thing there was. They even sometimes claimed that it ran practically as fast as C++, and that any measured differences were insignificant. Chances are that you tried it back then, saw how awfully slow it ran, and then dismissed it as a web development toy and decided that the Java evangelists either was liars or fools. Fortunately, most the hype surrounding Java has since then died out and the compilers and virtual machines has improved significantly in meantime. I heard the same hyped arguments back then and originally dispelled the use of Java as anything but a web applet language, because of the many promises that was clearly not true. I did not return to Java until late in 1998. [...] As we will see in this chapter Java is in fact not as slow as its reputation claims; it all depends on how you use it." [page 32]
Sounds like he had thoughts similar to yours. He changed his mind though:
"One very important factor is the amount of tweaking that has been performed on the Java program. According to the conclusions of chapter 7, a non-tweaked program will be several factors slower than an equivalent C++ program. In my tests it was 2.5 to 4 times slower [on a modern JVM]. [...] However, if the Java program is tweaked (and the C++ program is tweaked too for fairness in comparison) this difference is reduced significantly. Many highly tweaked programs have in fact been shown to be able to run faster in Java than in C++. [...] Judging from the various benchmarks I would say that tweaked Java programs on the average runs 20-50% slower than tweaked C++ program" [page 34&35]
The writer also comments that tweaked code doesn't get the advantage of quick development (about a 30% productivity boost overall), but that will only be true for the 10% of critical code. Furthermore he points out that Swing is too slow. He also shows that JVM's are progressing quickly. A very good point is that Java is plagued by too many poor programmers because the language is easy to learn (the VB-curse). In the end he sees Java as a good option for some games.
Whatta brochure-head you are.
Yep, the kind of brochures with references and benchmarks. And sometimes they are graded at a university with an A+.
Should I put more credit in someone who critiques a language by attacking an article of a random Java supporter, complaining about several style issues, blaming the language for having flaws that are not explained (what is wrong with the reflection in Java?) and using sweeping statements that are unsupported by evidence: "Most heavy OO fans agree". He even attacks the language for being OO ("Java does not make doing procedural programming very easy.").
SOAP and WS is meant to replace non-remote API calls ???????
That might be a good idea (assuming we are talking about calling API's with a few calls only, not basic library-calls). Using GLUE you can reach 300,000 messages/second in the same application. It contains various optimizations for local use.
This is only one implementation of course. But it seems to be possible to get decent performance. I certainly won't dismiss it as a possibility beforehand.
For the "masses"? That is a bit of an exaggeration. EDI could have done the same if somebody busted up the monopoly that made it so expensive. IOW, it is not the technology, but the politics.
AFAIK EDI is complex, rigid and requires a dedicated (=expensive) communications link. Web services seems to improve on this a great deal. Given the fast developments in this area, I expect 50+% of US companies to use it in their business processes in 10 years time.
BTW, once you actually want to use a technology, I see little use in making a distinction between technological or political factors.
The biggest problem with EDI was not the format, but translating subtle differences between companies. Their internal systems had specific restrictions or rules beyond (or below) the EDI standard. XML won't solve this either. More at: http://www.geocities.com/tablizer/softeng.htm
XML is as good in translating as anything. You can use templates (procedural extensions are possible), SAX or DOM. You can use your favorite language for everything but the stylesheets (you aren't forced to use them).
I am not saying that XML does not have some nice features that early formats had a harder time with, but I don't see any evidence that comma-delimited files made headlines as "the great equalizer" when it came out. XML is simply another format to add to the tool box
XML is superior by keeping the metadata with the data for easy processing. There are a great many tools and standards available that makes XML more than 'just another format'.
Besides, since most Java apps run on servers, the VM just slows things down. The "Sandbox" crap mostly makes sense on clients, not servers. Sun hit the wrong target.
Sigh. The slowless is mostly in the GUI, something that server-apps don't have. The actual code is not even that slow on a modern JDK. And a bit of slowness is a small price to pay for the portability, stability, fast development, good features and big libraries. This is especially true for server-applications where network-latencies and database-accesses are usually a far bigger problem than slow code.
The fact that Java has found this niche while Sun hyped the technology for totally different purposes is very positive IMHO. It clearly shows that the market sorts itself out and finds the best tool for the job (eventually).
Is Java really anything revolutionary?
It doesn't have to be revolutionary. Just better. I think it has achieved that goal for at least two niches: server-side apps and mobile phones (Java is a big succes on Japanese phones).
The fact that you blame the technology for not being revolutionary is telling. You yourself have become victim to thinking as a marketeer. Technology doesn't have to be REVOLUTIONARY, NEW AND EXCITING to be useful or a step forward. Because you set impossible standards for new technology to match, you can always dismiss it.
It is cross-language API's that need to be developed, and NOT Yet Another Language (especially one that is not signif diff from existing lanaguages, like OO-Pascal, etc.)
Indeed, Web Services/SOAP is the thing you use for that. Works perfectly with Java.
Why should one have to use language X to access API Y ?????
SOAP is being made transparent. People are trying to do exactly what you want. Are you keeping up?
Is this what you idiots call "progress"? Figures. Maybe if I bash my head with a bat I will finally "get it" and be a blissful retard. That is not very "cross platform" to me. Sun either didn't think or was thinking with their marketing cap on.
Java makes your code cross-platform and SOAP makes your messages/RPC's cross the language barrier. What more do you want?
(* I do care about the large library of Java code that is available and runs on Windows, Macs, Linux and Solaris. *) So does Perl, Python, LISP, and C.
I never said that Java is the only good/possible choice. I was just trying to point out that Java has many a good quality. No other language is superior to it in every respect.
A semi-real-world example? I enjoy a challenge if it tries to reflect real-world situations instead of some lab game.
I gave you the example of XML datamodels. It's real-world and no other technology has succeeded in getting entire industries together to develop them.
Besides, "old" is irrelevant. "Old" simply means that it has not come back into the hypestream under a new name yet. You can take the same wide ties or bell-bottoms out of the closet every 22 years or so (if your butt does not keep growing).
Partly correct, partly wrong. If the old comes back it's usually in a different form. The environment in which the rehashed technology is introduced will also be different from reality as it was 20 years ago. This means that you cannot simply use convential wisdom. Java uses a complex VM which was not new, but was not viable for a long time (mostly because computers were to slow). And some things (hopefully) never come back (goto 10).
I guess my definition of "promising" has changed over the years. There are technologies/standards that I think could exist to solve/reduce real problems, but the industry seems to wallow on stupid stuff instead.
Like Web Services that will bring EDI to the masses? Is it stupid to optimize the information transfer between different companies? Perhaps you can give me a real-world example of what the industry is obsessing over that won't solve real problems?
I really wouldn't care if Java is marginally worse than your favorite obscure language or that XML is warmed over LISP (although you've probably used neither and this is pure hearsay).
I do care about the large library of Java code that is available and runs on Windows, Macs, Linux and Solaris. I do care about almost non-existent leaks, good exception support and all the other niceties. I do care about the help that I can get if I'm stuck with a problem. I do care about the great number of tools that are available for XML-processing. I do care about the fact that XML has been introduced at the right moment and has made everybody rally behind it. I do care that people are starting to develop industry-wide datamodels based on XML (www.hr-xml.org).
I can't get these things with your old, 'superior' technology. I agree that new technologies are usually overhyped and often misused, but a good programmer can seperate the hype from the substance. Your simple condemnation of 'new' things shows that you are indeed out of touch. Why? Because you are unwilling to learn. Smart programmers always jump on promising new technologies to test them out.
I know a university student who's can get very excited about the lottery. But playing in a lottery doesn't have to be irrational. Good reasons can be:
You value a small chance to get a big return much higher than a small guaranteed return. This may be true when those few dollars extra don't really improve your well-being, but a few million would do so greatly. A good example is someone with great debts, those extra dollars won't help him in paying them off. He's better off by taking a great risk.
As an rational person you accept the fact that humans are for a good part motivated/driven by fairly irrational desires (sex when you don't need children, an expensive gadget that you don't need, etc). The lottery can be a fairly cheap way to satisfy your materialistic desires. You basically treat the lottery as entertainment instead of an investment. Entertainment doesn't have to pay off financially.
Some (a lot of?) managers think extreme programming is to team up experienced with in experienced coders. This is wrong, stupid and will piss of the good programmers.
The point of such an exercise is to teach the newbie which is not wrong or stupid. The goal is no longer to write code fast, but to teach the junior programmer about coding style, solving problems, design patters, unit tests, etc. It's a crucial part of making them productive in the future. Of course, you should keep these things in mind:
Don't pair with newbies exclusively. You're not exclusively a teacher. No company lets their programmers study/teach without being very productive all the time. Talk to your manager and explain this. Preferably you should be allowed to choose your own partners and be expected to spend at least x hours pairing with newbies. Experts should also pair with experts and newbies with newbies. The newbies can use the many things you taught them and teach them to each other. I'd expect that 1 hour of intense pairing with you should be matched with 4-5 hours of pairing with another newbie. This gives them enough time to process and use all the knowledge that you shared. It also means that you don't have to program with a newbie that often.
Don't pair with the same guy all day. Especially not with a newbie. 8 hours a day teaching someone is just too long (and not very effective).
Don't tackle the most difficult problems with newbies. You solve those with an expert, the simpler ones are perfect for teaching.
Be patient! You don't have to be very productive when teaching a newbie. Take you time! Before you know it, he's become an expert and you'll be happy to pair with him.
Try to learn from the newbie. They have a fresh perspective on things. Try to understand why they ask 'dumb' questions. Is it a problem with the design? Is your code unclear? Did you comment enough?
This happens a lot. Management is usually very impressed by 'official' products, while programmers are impressed by software that works and doesn't lock them in. Luckily management is often more concerned with getting things working than with their 'official policies' so programmers/administrators can often get away with using open source products that save the company a lot of money. Management just looks the other way (or is kept in the dark). Only very recently has management become aware of the advantages of open source and have sometimes made it their official policy (but almost exclusively in application-development or server administration).
But perhaps you disagree and want to explain to me what open source software is now widely used by office workers. Linux? Nope. OpenOffice? Nope. Mozilla? Nope.
Programmers are consumers... that's an important point that you and many others seem to miss.
I didn't miss it. Programmers are a subset of consumers. If programmers do something, you can't simply generalize it to the superset. It's like claiming that a soccer-team is very good, because some players are individually good. This is not a logical conclusion.
The person in an organization who has to deal with crashed operating systems is going to be the person who cares most about the stability of the OS. In marketing terms that person is known as an "influencer" and those influencers are the people who are making open source happen.
Open source = Linux/BSD??? It's much broader than that. Open source has mostly been the domain of programmers and has only recently become known to less-technical people. I assume that the guys who "deal with crashed operating systems" are server administrators (open source OS's don't fare so well on the desktop in companies), those guys are highly technical. These administrators aren't just 'consumers', they usually hold their own in shell scripting and Perl programming.
I really don't see your point, highly technical people are making this revolution happen. Only very recently have marketeers and CEO's started to contribute. At the same time, I see very little use of open source products by the average consumer. I don't blame them, since open source products are usually not very suited for them.
The right to restrict the distribution of your work. That's what copyright law is about. By using the BSD you renounce these rights almost completely (and thus allow almost unrestricted distribution). The GPL is much, much stricter. Of course, some people like those restrictions.
... by voluntarily giving up the rights that copyright gives you (by using the BSD-license for instance).
This is wrong. If I put code under the BSDL, I am not giving up my copyright on the code. The license (any license,) just says how you can use the code.
I didn't say anything about giving up copyright. I said that you give up the (most important) rights that copyright grants you.
Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
Whoopie doo. Copyright is used to keep a disclaimer with the program/source and make sure that someone else cannot pretend to be the creator. That's all. Does that limit you in a significant way?
If I'm giving money to the government, and they turn around and spend that money to help develop software, then I sure as hell deserve a piece of it!
I agree. Give me access to it with the fewest limitations as possible: a BSD-license.
If the software is licenced under the GPL, then there is no way that a company can take the software that I helped fund and use it to steal me blind.
I see. While you have the option of getting the software that you paid for at no cost, you _must_ have the additions that someone else worked on. That extra work is your property because it based on the work that you funded? Similarly to the way that reviews of movies are the property of the producer, Photoshop plug-ins are the property of Adobe and plane's are the property of the Wright brothers? Oops, they aren't. You don't automatically have the right to someone elses work, because they aren't your slaves. Deal with it.
If I helped fund it, I want something tangible in return, and the GPL provides that.
So does the BSD-license, but that sucks because you might have to pay if you want access to someone elses work. It's tough living in a capitalist society, isn't it?
if you change the code and if you distribute it, then your derived work is still part of the original work.
This is untrue. The derived work becomes available as GPL'ed code but that code can never become part of the original work. KDE 1.0 that was made public in 1998 won't change if someone makes a change to the sources and releases it. If it would, you would have invented a time-machine (code is created on the basis of a product and alters the product as it was released originally).
The point you are really making is that you don't believe in public property. Knowledge that is owned by no-one and whose use cannot be limited. Examples are the invention of the wheel and the books in the Gutenberg project that can be freely used by all. Many BSD-advocates believe very strongly that public property is the basis for lasting progress and innovation. This pool is fed by copyrights that expire and by voluntarily giving up the rights that copyright gives you (by using the BSD-license for instance). Creating a pool of knowledge that can be freely used for whatever means is not welfare, it is progress.
I think the best way to feed the pool is by having short copyrights with code escrow* and having a strong BSD/Apache-community. The first allows people to get paid for their work (and thus stimulating innovation), while contributing to the repository of freely available code and making sure that one cannot live of one succesful product forever (but needs to keep innovating). The second allows us to have free and trustworthy products in certain area's. It's mostly the infrastructure that benefits greatly from being open source as many companies are beginning to learn (like they had to learn about open standards before that). But not all needs can be filled with open source, so why create such a rift between them? Why not let them coexists and feed of each other?
I know this will start a heated debate, but if my tax dollars are paying for something, I want it issused so that some value comes back, not just a welfare-like giveaway.
First of all, many people contribute to BSD-like licensed products such as XFree or Apache. There are some very good reasons to do so. If someone creates a commercial spinoff(which usually would not have been created as a GPL'ed product), that is also valuable. Perhaps you refuse to use it, but many people don't. They will have more choice (see the commercial versions of Apache). This seems like a good thing to me. The fact that you are unwilling to pay for someone else's work or implement the changes they make yourself seems to indicate that you do like welfare.
*I don't think that Bill agrees on that one;)
PS. Open source seems a programmer revolt to me. Most consumers do not even know what source code is, let alone contribute.
Of course not. If you think that something's unreasonable then obviously you think it shouldn't be done.
With basic sales, there are no limitations above those that the law defines. With contracts, you may limit everything the law doesn't prohibit. Of course, that doesn't make every limitation right. Is it so wrong for me to question certain conditions? Or do you not disagree with most EULA's (apart from the questionable legality of a click-through contract)?
I can not give the student my car (you're happy with that as I understand it)
I strongly believe in a responsibility towards those who are less fortunate. But I strongly dislike when people give half-assed donations. I'm not afraid to question these.
Sounds to me like you're way too concerned about a legal distinction that makes no difference to the substance of the transaction.
I think there is a big difference between the transfer of ownership and lending something to someone. In the first case you give someone a good that he can use in a way that he sees fit, in the second case you allow someone to use your property in a certain way. Perhaps you can't see the difference (like the RIAA). In practice the difference is substantial as I've never seen software that was offered as a loan.
But if I say the GPL is like I lend the student my car on the condition that when he's finished with it he passes it on then that sounds good to you? Okay, the GPL is like I lend my car to a student... feel better about it now?
Of course, my analogy breaks down here:) Everyone was taking it far too seriously anyway. Why did everyone attack me when my analogy was certainly better than the one in the post I responded to?
I have to wonder at this point whether you're being deliberately silly. I say to the student, quoting form my earlier post "I have this mustang I no longer need. If you like you can have it but on the condition that when you no longer need it you also pass it on to someone else, including any work you've done on it". Reading this you think there's some pretense that there are no restrictions? How could anyone think that? I even explicitly said there was a condition. Suppose I spray paint "there is a condition attached" across the sides of the car? Would this help? Would you be happier if I used the word "restriction" instead?
It seems that the GPL pretends one is giving things away Freely.
So if I give him the car subject to a certain condition then I've mistrusted him, asserted that I'm always right and deprived him of moral choices but if I use the word "lend" instead of "give" then I haven't?
In that case you are honest about mistrusting him. After which I can attack you on that. Knowing you have a problem is the first step towards a cure;)
I assume you're not asserting that you have in fact considered the infinite possible scenarios and determined that there's a better outcome without the conditions attached?
I've already proven that there are scenario's where the limitations are not preferable if you want to maximize happiness. Under the assumption that human behaviour should be unregulated unless there is a very good reason not to, I prefer the BSD. Of course I have many more arguments, but I'd rather discuss these without the burden of an analogy. Place your arguments on the table if you want to turn this into a basic BSD vs GPL debate.
You feel that he is entitled to be charge for the car once I've given it to him but somehow it's wrong of me to attach conditions?
I think that sales and gifts should not attach conditions on the buyer/reciever unless those restrictions are reasonable. A donating country should not dictate that his donation is spend in the donating country for instance. Doing so makes it far less of a gift. Conditions may even be illegal by law, I may not forbid you to sue me if my product causes damage to you.
Is it wrong of me to let people stay at my house unless they can sublet or even sell the rooms they stay in too?
There is a difference between a tranfer of a good and providing a service. Hiring someone or something for a service does not make you the owner of that person or good. If you don't want the student to take full ownership of the car, you should lend it to him. Don't pretend that you give it away freely (in the no-restriction sense of the word).
Seriously, I can believe that you're so self centred that you'd not only avoid improving the car but actually run it down out of spite at the thought that your profit from my generosity was limited to use of the car while you had it (how terrible, how devastatingly unfair of me) but your belief that everyone else in the world is as mean as you is not, in my opinion, accurate.
I wasn't talking about me (and I didn't imply that the student would intentionally wreck the car). I trust the student to take advantage of the car I give to him in the way that benefits him most. I would be very happy if he decides to fix it up and sell it as he can probably use the money. Hopefully my generosity and the compensation for his hard work will make him a better person. Perhaps he can help someone else one day (when he has a good job).
You, on the other hand, do not trust the student to use his own judgement. You want to force him to give away the car after he's done with it, instead of trusting him to do the right thing. How does it reflect on your view of the world that you believe that the student shouldn't be allowed to make a moral decision? Are you always right? Have you considered the infinite possible scenario's and concluded that the student always does the best thing by giving away the car (instead of fixing it up, selling it to a rich guy and giving the money to a good cause or using the money to help himself and helping others when he's succesful in society).
As a server-side app, it doesn't have to be distributed to others. So the GPL won't do any good. I like the interchange of info between databases, but that can be very hard to implement correctly. You'll run into all kinds of mismatches where you need to do the right thing (when do records denote the same book?). You can fix a lot of these problems by focusing on the ISBN, but you'll still run into inconsistent data and the like. There will also be security issues, you have to make sure that the guy who you synchronize with hasn't polluted his database with wrong data. Lastly, copying an entire database takes a lot of horsepower. You don't want people doing this to your database during busy hours.
Apart from these technical considerations, it's very likely that one database will become the standard that everyone uses, even when there is more than one choice (everyone will use the most extensive, most correct database). I doubt that there will be many syncronizations into this database, most people will be interested in making a copy of it.
All in all I think the best bet is to incorporate a feature that makes it easy to automatically dump the database to a mirror at a certain time (midnight or so). There's no good way to make sure that this actually get's done though, but you can always write a screenscrape application to punish the organization that doesn't supply you with a dump (scraping the entire database will hurt them bad). When the organization goes nasty, you can switch over to a copy and try to beat them with an open alternative.
Slashdot Mirror
True, but companies dont get to take out the space shuttle for a spin either.
What about Dennis Tito? He pays for part of the russian space budget in exchange for a trip into space. Companies also pay for getting their experiments on the space shuttle. This reduces our taxes.
In the case is this software, there is an inequity in the availability of the software for the taxpayer who paid for it and the corporations who didnt.
Wrong. You never paid for the use of the software. You paid for the stuff they do with space shuttles, rockets, etc. The software is just a byproduct. Companies sponsor Nasa by paying money for this byproduct. Doing so, they lower the taxes you have to pay (or increase the value you get for the taxes).
You pay Nasa for exploring space and to boldy go where no one has gone before. You don't pay them to provide you with the software they use to achieve this goal (software for controlling rockets, shuttles and other stuff that I'm sure you don't have). Selling the software to private parties means that they have more money available to achieve their primary goal. This does benefit the taxpayer.
Or perhaps you would rather pay more taxes because Nasa can't sell their software? In return you'd get all the software to control your own space shuttle for free. Wouldn't that be great?
I agree that the software should be used in a way that serves our best interest. But that doesn't mean that open sourcing is necessarily the best way to do that. A lot of government software is specialized enough that few people can use it. The options the gov has in that case:
- Keep it to yourself. Has the big advantage that you've got no extra costs. The taxpayer gets the benefit for which this software was written.
- Open source the code. This means extra expenses to get it in shape by removing proprietary stuff from other companies, making it less dependant on the specific environment in which the government organisation runs it, making sure there are no references to Roswell and other secret stuff, etc. Those extra costs will bring no benefit to 99.9% of the taxpayers.
- Sell the code if there is interest. This pays for new projects and thus benefits the taxpayer.
I'm not convinced that the second option is always the best.Second, the widespread belief among open source types like myself is that the more eyes that are on the code, the more likely the vulnerabilities will be discovered and ultimately patched.
I'm sure that thousands of developers will eyeball the code (written in Ada or another relatively obscure language) to control the space shuttle's robot arm. Everyone will surely test the code on their own space shuttle. There is a big difference between Linux and a piece of highly specialized software. You can't expect much from these code reviews. Especially since Nasa-software is already extensively reviewed by experts.
It's extremely stupid that Windows tries to read your disks when they are removed. It has irritated me often. MacOS handles that much better by actually removing open windows/drives when you eject a storage device. That isn't possible with a hardware-controlled eject. Learning to reboot with the mouse-button down isn't that hard in those rare cases that a CD won't eject.
PS. What does the key sequence for booting have to do with this (pressing 'c' on my 1997 G3)?
Extreme freedom is the public domain. A bit less is the BSD. A lot less is the GPL. Or do you deny that the GPL limits your ability to distribute software?
The German tragedy suggests gun control is ineffective
Europe with it's strict gun control laws has far less shootings than the US. How are the laws ineffective?
They still have to do some sort of mark and sweep algorithim, which is not fast.
A 'complete' mark and sweep is not fast (one that doesn't miss anything). Eden uses a fast, but incomplete algorithm. That destroys all the easy temporary objects (and moves the rest to long term storage which gets the full GC). I don't see why your approach would be so much faster.
The feature that I want to see added to java is that I want to be able for the java vm to create objects on the stack. Then if they get assigned to a member pointer, get promoted to the heap. This way you could keep objects on the stack where they could be easily collected when the function that created them ends scope.
That's already part of Hotspot:
"A generational collector takes advantage of the fact that in most programs the vast majority of objects (often >95%) are very short-lived (i.e. are used as temporary data structures). By segregating newly created objects into an object "nursery", a generational collector can accomplish several things. First, because new objects are allocated contiguously in stack-like fashion in the object nursery, allocation becomes extremely fast, since it involves merely updating a single pointer and performing a single check for nursery overflow. Secondly, by the time the nursery overflows, most of the objects in the nursery are already "dead", allowing the garbage collector to simply move the few surviving objects elsewhere, and avoid doing any reclamation work for dead objects in the nursery."
Here's an in-depth explanation with pictures.
(* Java is simpler and clearer than C++. Or do you disagree? *)
Roughly 98% of all languages are. That is like bragging that car crashes are less painful than nukes.
Java succeeds in replacing C++ for quite a few tasks, that's the big difference with most languages in that 98%.
(* You can create fast code with a VM! *)
Yes, but it complicates the VM engine. You are lucky Sun is footing the bill. A compiler would be simpler. Write the compiler in C, so that porting it would be relatively easy.
People write compilers for Java. It seems just as complicated as they have a hard enough time just getting it to work on x86. Besides, who cares? It seems to me that you are changing the subject. I remember this being about programming in Java, not writing VM's.
(* Multi-paradigm means chaos then. *)
Okay, then get rid of the OOP. Even many OO fans that I debate with admit that sometimes they mix in procedural solutions as they see fit.
I don't mind having to use another tool for something. I never claimed that Java is perfect for everything. I do mind when someone tries to create a frankenstein language that is everything to anybody. Perl and C++ for instance. Yuck.
I don't think it would "ruin" the language if Java had real functions.
That's not very helpful. Do you want object-independent functions (func(x) instead of x.func() or staticClass.func(x))? Or are you talking about operator overloading (ComplexType + ComplexType)?
(* Do note that we are generally moving away from efficiency towards clarity and simplicity. *)
How do you explain that sh*t Java then?
Java is simpler and clearer than C++. Or do you disagree?
My point was that if it was compiled, it would have an edge over a VM.
You can compile it. But the report (and this one) shows that the compilation vs VM issue is quite irrelevant. You can create fast code with a VM!
Where did I say that Java sucked because that writer was stupid? It is both. Java AND that writer suck.
You put it on a page that is called: "Why Java Sucks". Put it on a seperate page if it's a seperate topic.
(* blaming the language for having flaws that are not explained (what is wrong with the reflection in Java?) *)
It is too convoluted. There is plenty of criticism of it on the web and ng's. Go look.
I know it has limitations. The point is that you should explain your claims. Complaining that X is lousy without giving proof or even telling people why it sucks will make me ignore it. Who are you trying to reach with your critique? Programmers that already know all about Java's reflection limitations (only an issue to experts really)? If they already know the limitations, what good does your page do? Your critique is pointless.
What is wrong with being multi-paradigm? C++ can be multi-paradigm.
Multi-paradigm means chaos then. I prefer clean design, not Perl or C++.
Only zealOOts think one paradigm fits all.
I don't remember saying that Java needs to replace all competing languages. I do think it's a very good one. BTW, you can use various scripting languages with Java. You can also call native code if Java doesn't cut it.
How hard would it be to add functions to Java anyhow?
Excuse me? Could you elaborate?
How is it "rigid"?
...'. You then complain that your hack doesn't work.
I read your story: 'There wasn't enough room for comments, so we hacked
Further, the current EDI networks are still more reliable and secure than web-based solutions. Many big-name companies don't want to switch until such issues are solved on the web.
Big companies can pay for a subscriber line to their suppliers and/or customers that circumvents the internet. Small companies just have to to take the risk I'm afraid.
Nope! XML takes up more bandwidth than some other solutions because it repeats field/tag-names for each record. I realize that this is generally not a problem, but it does show that you are wrong.
I don't see how this is related to translating. If there is a problem with bandwith, you can simply compress the XML. Do note that we are generally moving away from efficiency towards clarity and simplicity.
And how the heck does the VM contribute to "fast development" and "good features", and "big libraries"?
I wasn't simply talking about the benefits/downsides of a JVM (I never intended to narrow the discussion, the JVM vs P-code was just an example of Java not being incredibly revolutionary). Useful things are usually not revolutionary BTW. They evolve from useless revolutionary technology (which is too complex or slow to be useful). I do claim that all the features of the language (some of which are due to the VM) are good enough to offset the slowless.
I want to refer you to this excellent report on the performance of Java. Some highlights:
"Java has a strong reputation for being a slow language that cannot be seriously considered for real applications. It has not gotten this reputation by coincidence. If you are a C++ programmer then you were probably very tired of all those Java evangelists back in 1995 that claimed that Java was superior in every regard and that 100% pure Java was the best thing there was. They even sometimes claimed that it ran practically as fast as C++, and that any measured differences were insignificant. Chances are that you tried it back then, saw how awfully slow it ran, and then dismissed it as a web development toy and decided that the Java evangelists either was liars or fools. Fortunately, most the hype surrounding Java has since then died out and the compilers and virtual machines has improved significantly in meantime. I heard the same hyped arguments back then and originally dispelled the use of Java as anything but a web applet language, because of the many promises that was clearly not true. I did not return to Java until late in 1998. [...] As we will see in this chapter Java is in fact not as slow as its reputation claims; it all depends on how you use it." [page 32]
Sounds like he had thoughts similar to yours. He changed his mind though:
"One very important factor is the amount of tweaking that has been performed on the Java program. According to the conclusions of chapter 7, a non-tweaked program will be several factors slower than an equivalent C++ program. In my tests it was 2.5 to 4 times slower [on a modern JVM]. [...] However, if the Java program is tweaked (and the C++ program is tweaked too for fairness in comparison) this difference is reduced significantly. Many highly tweaked programs have in fact been shown to be able to run faster in Java than in C++. [...] Judging from the various benchmarks I would say that tweaked Java programs on the average runs 20-50% slower than tweaked C++ program" [page 34&35]
The writer also comments that tweaked code doesn't get the advantage of quick development (about a 30% productivity boost overall), but that will only be true for the 10% of critical code. Furthermore he points out that Swing is too slow. He also shows that JVM's are progressing quickly. A very good point is that Java is plagued by too many poor programmers because the language is easy to learn (the VB-curse). In the end he sees Java as a good option for some games.
Whatta brochure-head you are.
Yep, the kind of brochures with references and benchmarks. And sometimes they are graded at a university with an A+.
Should I put more credit in someone who critiques a language by attacking an article of a random Java supporter, complaining about several style issues, blaming the language for having flaws that are not explained (what is wrong with the reflection in Java?) and using sweeping statements that are unsupported by evidence: "Most heavy OO fans agree". He even attacks the language for being OO ("Java does not make doing procedural programming very easy.").
SOAP and WS is meant to replace non-remote API calls ???????
That might be a good idea (assuming we are talking about calling API's with a few calls only, not basic library-calls). Using GLUE you can reach 300,000 messages/second in the same application. It contains various optimizations for local use.
This is only one implementation of course. But it seems to be possible to get decent performance. I certainly won't dismiss it as a possibility beforehand.
For the "masses"? That is a bit of an exaggeration. EDI could have done the same if somebody busted up the monopoly that made it so expensive. IOW, it is not the technology, but the politics.
m
AFAIK EDI is complex, rigid and requires a dedicated (=expensive) communications link. Web services seems to improve on this a great deal. Given the fast developments in this area, I expect 50+% of US companies to use it in their business processes in 10 years time.
BTW, once you actually want to use a technology, I see little use in making a distinction between technological or political factors.
The biggest problem with EDI was not the format, but translating subtle differences between companies. Their internal systems had specific restrictions or rules beyond (or below) the EDI standard. XML won't solve this either. More at:
http://www.geocities.com/tablizer/softeng.ht
XML is as good in translating as anything. You can use templates (procedural extensions are possible), SAX or DOM. You can use your favorite language for everything but the stylesheets (you aren't forced to use them).
I am not saying that XML does not have some nice features that early formats had a harder time with, but I don't see any evidence that comma-delimited files made headlines as "the great equalizer" when it came out. XML is simply another format to add to the tool box
XML is superior by keeping the metadata with the data for easy processing. There are a great many tools and standards available that makes XML more than 'just another format'.
Besides, since most Java apps run on servers, the VM just slows things down. The "Sandbox" crap mostly makes sense on clients, not servers. Sun hit the wrong target.
Sigh. The slowless is mostly in the GUI, something that server-apps don't have. The actual code is not even that slow on a modern JDK. And a bit of slowness is a small price to pay for the portability, stability, fast development, good features and big libraries. This is especially true for server-applications where network-latencies and database-accesses are usually a far bigger problem than slow code.
The fact that Java has found this niche while Sun hyped the technology for totally different purposes is very positive IMHO. It clearly shows that the market sorts itself out and finds the best tool for the job (eventually).
Is Java really anything revolutionary?
It doesn't have to be revolutionary. Just better. I think it has achieved that goal for at least two niches: server-side apps and mobile phones (Java is a big succes on Japanese phones).
The fact that you blame the technology for not being revolutionary is telling. You yourself have become victim to thinking as a marketeer. Technology doesn't have to be REVOLUTIONARY, NEW AND EXCITING to be useful or a step forward. Because you set impossible standards for new technology to match, you can always dismiss it.
It is cross-language API's that need to be developed, and NOT Yet Another Language (especially one that is not signif diff from existing lanaguages, like OO-Pascal, etc.)
Indeed, Web Services/SOAP is the thing you use for that. Works perfectly with Java.
Why should one have to use language X to access API Y ?????
SOAP is being made transparent. People are trying to do exactly what you want. Are you keeping up?
Is this what you idiots call "progress"? Figures. Maybe if I bash my head with a bat I will finally "get it" and be a blissful retard.
That is not very "cross platform" to me. Sun either didn't think or was thinking with their marketing cap on.
Java makes your code cross-platform and SOAP makes your messages/RPC's cross the language barrier. What more do you want?
(* I do care about the large library of Java code that is available and runs on Windows, Macs, Linux and Solaris. *)
So does Perl, Python, LISP, and C.
I never said that Java is the only good/possible choice. I was just trying to point out that Java has many a good quality. No other language is superior to it in every respect.
A semi-real-world example? I enjoy a challenge if it tries to reflect real-world situations instead of some lab game.
I gave you the example of XML datamodels. It's real-world and no other technology has succeeded in getting entire industries together to develop them.
Besides, "old" is irrelevant. "Old" simply means that it has not come back into the hypestream under a new name yet. You can take the same wide ties or bell-bottoms out of the closet every 22 years or so (if your butt does not keep growing).
Partly correct, partly wrong. If the old comes back it's usually in a different form. The environment in which the rehashed technology is introduced will also be different from reality as it was 20 years ago. This means that you cannot simply use convential wisdom. Java uses a complex VM which was not new, but was not viable for a long time (mostly because computers were to slow).
And some things (hopefully) never come back (goto 10).
I guess my definition of "promising" has changed over the years. There are technologies/standards that I think could exist to solve/reduce real problems, but the industry seems to wallow on stupid stuff instead.
Like Web Services that will bring EDI to the masses? Is it stupid to optimize the information transfer between different companies? Perhaps you can give me a real-world example of what the industry is obsessing over that won't solve real problems?
I really wouldn't care if Java is marginally worse than your favorite obscure language or that XML is warmed over LISP (although you've probably used neither and this is pure hearsay).
I do care about the large library of Java code that is available and runs on Windows, Macs, Linux and Solaris. I do care about almost non-existent leaks, good exception support and all the other niceties. I do care about the help that I can get if I'm stuck with a problem. I do care about the great number of tools that are available for XML-processing. I do care about the fact that XML has been introduced at the right moment and has made everybody rally behind it. I do care that people are starting to develop industry-wide datamodels based on XML (www.hr-xml.org).
I can't get these things with your old, 'superior' technology. I agree that new technologies are usually overhyped and often misused, but a good programmer can seperate the hype from the substance. Your simple condemnation of 'new' things shows that you are indeed out of touch. Why? Because you are unwilling to learn. Smart programmers always jump on promising new technologies to test them out.
The point of such an exercise is to teach the newbie which is not wrong or stupid. The goal is no longer to write code fast, but to teach the junior programmer about coding style, solving problems, design patters, unit tests, etc. It's a crucial part of making them productive in the future. Of course, you should keep these things in mind:
Developers using open-source software behind bosses' backs
This happens a lot. Management is usually very impressed by 'official' products, while programmers are impressed by software that works and doesn't lock them in. Luckily management is often more concerned with getting things working than with their 'official policies' so programmers/administrators can often get away with using open source products that save the company a lot of money. Management just looks the other way (or is kept in the dark). Only very recently has management become aware of the advantages of open source and have sometimes made it their official policy (but almost exclusively in application-development or server administration).
But perhaps you disagree and want to explain to me what open source software is now widely used by office workers. Linux? Nope. OpenOffice? Nope. Mozilla? Nope.
Programmers are consumers... that's an important point that you and many others seem to miss.
I didn't miss it. Programmers are a subset of consumers. If programmers do something, you can't simply generalize it to the superset. It's like claiming that a soccer-team is very good, because some players are individually good. This is not a logical conclusion.
The person in an organization who has to deal with crashed operating systems is going to be the person who cares most about the stability of the OS. In marketing terms that person is known as an "influencer" and those influencers are the people who are making open source happen.
Open source = Linux/BSD??? It's much broader than that. Open source has mostly been the domain of programmers and has only recently become known to less-technical people. I assume that the guys who "deal with crashed operating systems" are server administrators (open source OS's don't fare so well on the desktop in companies), those guys are highly technical. These administrators aren't just 'consumers', they usually hold their own in shell scripting and Perl programming.
I really don't see your point, highly technical people are making this revolution happen. Only very recently have marketeers and CEO's started to contribute. At the same time, I see very little use of open source products by the average consumer. I don't blame them, since open source products are usually not very suited for them.
The right to restrict the distribution of your work. That's what copyright law is about. By using the BSD you renounce these rights almost completely (and thus allow almost unrestricted distribution). The GPL is much, much stricter. Of course, some people like those restrictions.
... by voluntarily giving up the rights that copyright gives you (by using the BSD-license for instance).
This is wrong. If I put code under the BSDL, I am not giving up my copyright on the code. The license (any license,) just says how you can use the code.
I didn't say anything about giving up copyright. I said that you give up the (most important) rights that copyright grants you.
Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
Whoopie doo. Copyright is used to keep a disclaimer with the program/source and make sure that someone else cannot pretend to be the creator. That's all. Does that limit you in a significant way?
If I'm giving money to the government, and they turn around and spend that money to help develop software, then I sure as hell deserve a piece of it!
I agree. Give me access to it with the fewest limitations as possible: a BSD-license.
If the software is licenced under the GPL, then there is no way that a company can take the software that I helped fund and use it to steal me blind.
I see. While you have the option of getting the software that you paid for at no cost, you _must_ have the additions that someone else worked on. That extra work is your property because it based on the work that you funded? Similarly to the way that reviews of movies are the property of the producer, Photoshop plug-ins are the property of Adobe and plane's are the property of the Wright brothers? Oops, they aren't. You don't automatically have the right to someone elses work, because they aren't your slaves. Deal with it.
If I helped fund it, I want something tangible in return, and the GPL provides that.
So does the BSD-license, but that sucks because you might have to pay if you want access to someone elses work. It's tough living in a capitalist society, isn't it?
if you change the code and if you distribute it, then your derived work is still part of the original work.
;)
This is untrue. The derived work becomes available as GPL'ed code but that code can never become part of the original work. KDE 1.0 that was made public in 1998 won't change if someone makes a change to the sources and releases it. If it would, you would have invented a time-machine (code is created on the basis of a product and alters the product as it was released originally).
The point you are really making is that you don't believe in public property. Knowledge that is owned by no-one and whose use cannot be limited. Examples are the invention of the wheel and the books in the Gutenberg project that can be freely used by all. Many BSD-advocates believe very strongly that public property is the basis for lasting progress and innovation. This pool is fed by copyrights that expire and by voluntarily giving up the rights that copyright gives you (by using the BSD-license for instance). Creating a pool of knowledge that can be freely used for whatever means is not welfare, it is progress.
I think the best way to feed the pool is by having short copyrights with code escrow* and having a strong BSD/Apache-community. The first allows people to get paid for their work (and thus stimulating innovation), while contributing to the repository of freely available code and making sure that one cannot live of one succesful product forever (but needs to keep innovating). The second allows us to have free and trustworthy products in certain area's. It's mostly the infrastructure that benefits greatly from being open source as many companies are beginning to learn (like they had to learn about open standards before that). But not all needs can be filled with open source, so why create such a rift between them? Why not let them coexists and feed of each other?
I know this will start a heated debate, but if my tax dollars are paying for something, I want it issused so that some value comes back, not just a welfare-like giveaway.
First of all, many people contribute to BSD-like licensed products such as XFree or Apache. There are some very good reasons to do so. If someone creates a commercial spinoff(which usually would not have been created as a GPL'ed product), that is also valuable. Perhaps you refuse to use it, but many people don't. They will have more choice (see the commercial versions of Apache). This seems like a good thing to me. The fact that you are unwilling to pay for someone else's work or implement the changes they make yourself seems to indicate that you do like welfare.
*I don't think that Bill agrees on that one
PS. Open source seems a programmer revolt to me. Most consumers do not even know what source code is, let alone contribute.
Of course not. If you think that something's unreasonable then obviously you think it shouldn't be done.
:) Everyone was taking it far too seriously anyway. Why did everyone attack me when my analogy was certainly better than the one in the post I responded to?
;)
With basic sales, there are no limitations above those that the law defines. With contracts, you may limit everything the law doesn't prohibit. Of course, that doesn't make every limitation right. Is it so wrong for me to question certain conditions? Or do you not disagree with most EULA's (apart from the questionable legality of a click-through contract)?
I can not give the student my car (you're happy with that as I understand it)
I strongly believe in a responsibility towards those who are less fortunate. But I strongly dislike when people give half-assed donations. I'm not afraid to question these.
Sounds to me like you're way too concerned about a legal distinction that makes no difference to the substance of the transaction.
I think there is a big difference between the transfer of ownership and lending something to someone. In the first case you give someone a good that he can use in a way that he sees fit, in the second case you allow someone to use your property in a certain way. Perhaps you can't see the difference (like the RIAA). In practice the difference is substantial as I've never seen software that was offered as a loan.
But if I say the GPL is like I lend the student my car on the condition that when he's finished with it he passes it on then that sounds good to you? Okay, the GPL is like I lend my car to a student... feel better about it now?
Of course, my analogy breaks down here
I have to wonder at this point whether you're being deliberately silly. I say to the student, quoting form my earlier post "I have this mustang I no longer need. If you like you can have it but on the condition that when you no longer need it you also pass it on to someone else, including any work you've done on it". Reading this you think there's some pretense that there are no restrictions? How could anyone think that? I even explicitly said there was a condition. Suppose I spray paint "there is a condition attached" across the sides of the car? Would this help? Would you be happier if I used the word "restriction" instead?
It seems that the GPL pretends one is giving things away Freely.
So if I give him the car subject to a certain condition then I've mistrusted him, asserted that I'm always right and deprived him of moral choices but if I use the word "lend" instead of "give" then I haven't?
In that case you are honest about mistrusting him. After which I can attack you on that. Knowing you have a problem is the first step towards a cure
I assume you're not asserting that you have in fact considered the infinite possible scenarios and determined that there's a better outcome without the conditions attached?
I've already proven that there are scenario's where the limitations are not preferable if you want to maximize happiness. Under the assumption that human behaviour should be unregulated unless there is a very good reason not to, I prefer the BSD. Of course I have many more arguments, but I'd rather discuss these without the burden of an analogy. Place your arguments on the table if you want to turn this into a basic BSD vs GPL debate.
You feel that he is entitled to be charge for the car once I've given it to him but somehow it's wrong of me to attach conditions?
I think that sales and gifts should not attach conditions on the buyer/reciever unless those restrictions are reasonable. A donating country should not dictate that his donation is spend in the donating country for instance. Doing so makes it far less of a gift. Conditions may even be illegal by law, I may not forbid you to sue me if my product causes damage to you.
Is it wrong of me to let people stay at my house unless they can sublet or even sell the rooms they stay in too?
There is a difference between a tranfer of a good and providing a service. Hiring someone or something for a service does not make you the owner of that person or good. If you don't want the student to take full ownership of the car, you should lend it to him. Don't pretend that you give it away freely (in the no-restriction sense of the word).
Seriously, I can believe that you're so self centred that you'd not only avoid improving the car but actually run it down out of spite at the thought that your profit from my generosity was limited to use of the car while you had it (how terrible, how devastatingly unfair of me) but your belief that everyone else in the world is as mean as you is not, in my opinion, accurate.
I wasn't talking about me (and I didn't imply that the student would intentionally wreck the car). I trust the student to take advantage of the car I give to him in the way that benefits him most. I would be very happy if he decides to fix it up and sell it as he can probably use the money. Hopefully my generosity and the compensation for his hard work will make him a better person. Perhaps he can help someone else one day (when he has a good job).
You, on the other hand, do not trust the student to use his own judgement. You want to force him to give away the car after he's done with it, instead of trusting him to do the right thing. How does it reflect on your view of the world that you believe that the student shouldn't be allowed to make a moral decision? Are you always right? Have you considered the infinite possible scenario's and concluded that the student always does the best thing by giving away the car (instead of fixing it up, selling it to a rich guy and giving the money to a good cause or using the money to help himself and helping others when he's succesful in society).
As a server-side app, it doesn't have to be distributed to others. So the GPL won't do any good. I like the interchange of info between databases, but that can be very hard to implement correctly. You'll run into all kinds of mismatches where you need to do the right thing (when do records denote the same book?). You can fix a lot of these problems by focusing on the ISBN, but you'll still run into inconsistent data and the like. There will also be security issues, you have to make sure that the guy who you synchronize with hasn't polluted his database with wrong data. Lastly, copying an entire database takes a lot of horsepower. You don't want people doing this to your database during busy hours.
Apart from these technical considerations, it's very likely that one database will become the standard that everyone uses, even when there is more than one choice (everyone will use the most extensive, most correct database). I doubt that there will be many syncronizations into this database, most people will be interested in making a copy of it.
All in all I think the best bet is to incorporate a feature that makes it easy to automatically dump the database to a mirror at a certain time (midnight or so). There's no good way to make sure that this actually get's done though, but you can always write a screenscrape application to punish the organization that doesn't supply you with a dump (scraping the entire database will hurt them bad). When the organization goes nasty, you can switch over to a copy and try to beat them with an open alternative.