Before you write off pure open source licensing, ask yourself the following questions: How important is it that my company’s modifications to open source code remain private? What does my company gain from the expense of keeping them that way? And finally, what might my company gain from the alternative?
InfoWorld: What’s so bad about the GPL?
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
“How important is it that my company’s modifications to open source code remain private? What does my company gain from the expense of keeping them that way? And finally, what might my company gain from the alternative?”
But, I thought you only have to give the source code to the ones you give the binary code to, so that isn’t everyone.
You could write GPL software yourself, sell or rent it to a person, and that person has the right to the source, but everyone has right to it.
As for derived works also being under GPL, well this is like any derived software to some extent – I mean like this, if you take software code and extend it then you will still have to do some kind of deal with the copyright holder anyhow. The GPL is just an easy hassle-free option, but like was said you still have the option of using extended softare under a second license if the copyright holder permits, and if the copyright holder doesn’t permit, then tough – it is their hard work that wrote it! You’ll have to go and write your own code.
You can still write software under any license you wish that runs on GNU/Linux or other GPL’d libraries. I’ve noticed that there is a lot of scare stories in the press that if you run code on GNU/Linux or that uses GPL’d libraries then suddenly it too becomes GPL. They use FUD and terms like ‘viral’! GPL is nothing of the sort.
Just today I remembered reading Stallman’s interview when he explicitly says he doesn’t use “non-free” software. What I would really like to know is how does he look at something as trivial as Google (just an example, I’m sure you can find better ones). Theoreticly he is using their software (if he does use it) which isn’t “free”. Does he view this as only using their services? If that’s the case, then his ideology is flawed. A lot of free software can be used just as a service on not as traditional software. Many future plans of big software companies (Microsoft beeing one of them) are aiming at making software a service used via the internet. I believe Microsoft is planing the next Office suite to be used via the web – not through a web browser, but through a special client to set up the requiered scripts/GUI,…
I would like to know how does he view this. If Microsoft would choose to open source that Office “client” but kept the actual software (on the server) closed, would that be (in his view) free software?
There are many more examples of this the web becomes more and more oriented at serving software. If I make a web page that relies heavily on Javascript/CSS/PHP on serving that software through the web, but I use an open browser (i.e. GPLed) to access that page, am I still using free software?
I don’t recall him mentioning anything like that, so if you have a link, fill me in.
The GPL can hinder adoption by businesses who, for many reasons, cannot contribute back code. The GPL assimilates other licenses if developers dare try to intermingle code. Followers of The GPL Way don’t see fault in this, but there are many many people who find that other licenses suit their needs better.
PHP and CSS are both open. that example would’ve been more valid had you said “ASP and flash/WMA/MP3/etc”
HTML, XHTML, CSS, and PHP (as well as many others) are all open web formats. where it gets tricky is the pictures… technically, gifs, jpgs, and such require licenses. so if you view a web page that uses those image formats are you not using free software?
I think it’s ok to advocate use of “free software” but in reality it’s near impossible to go through the day without being touched by proprietarity in some form.
The big boys like IBM? They seem totally unconcerned by this kind of issues…
I don’t recall him mentioning anything like that, so if you have a link, fill me in.
I think there were a few proposals for the next GPL revision that tried to deal with this sort of thing. It was all very, very surreal with things that seemed to indicate extension of the license to code connecting via remote connection (The connection being equivalent to linking and therefore making it derivative). I may be remembering the discussion incorrectly, so don’t take my word for it.
Yea a very interesting point about internet and free software. I’ve tought of using another search engine than google.. but is there even a Free Software search engine..?
Ofcourse it’s nearly inevitable to use some kind of closed software if you browse the web. But I hope there comes a strong Free Software searchengine.. Hope, because I can’t code good software :s
grtz
guni
What’s the point exactly? I mean, anyone with a brain would ask themselves the questions at the end of the article when considering using GPL licensed software.
For example, modifying GPL-licensed code doesn’t necessarily mean you have to make your own internal changes public, but once you show your changes to anybody outside your organization, the GPL automatically grants rights to your code to everybody.
<p>No, it doesn’t. You only have to supply source code to the individuals to whom you distributed your changes. It’s pretty bad when an advocate of the GPL can’t even get the requirements right.
Umm, if you are a software company, how are you supposed to make money if you have to give your software to everybody for free? I mean the point of a business is to make money. This author doesn’t quite address this rather fundamental concern.
The article points out that several companies have released source under a dual license. For example, TrollTech ships the Qt development kit under both the GPL and a commercial license; you can download the GPL’d version for free and use it to develop GPL’d code, or you can pay for a commercial license (not cheap, >$1000 per developer last time I checked) and develop proprietary code.
Somewhat off topic, I’ve always thought that “source available” was a good option for providing some of the transparency benefits of open source. This is like Microsoft’s “shared source” program except that the customer would be allowed to build and modify the code provided they don’t redistribute it outside their organization.
I wrote “You could write GPL software yourself, sell or rent it to a person, and that person has the right to the source, but everyone has right to it.”
That should have been “…but NOT everyone has a right to it.“. Ooops, that is a very important point, sorry to have made a mistake, shouldn’t type so quick!
As QuantumG said “You only have to supply source code to the individuals to whom you distributed your changes. It’s pretty bad when an advocate of the GPL can’t even get the requirements right.“.
Its just FUD when they get such facts wrong.
You might like to check out, for example, the business models of Sendmail Inc. (http://www.sendmail.com/) and Sleepycat Software (http://www.sleepycat.com/), amongst others.
Most software is developed by companies who are NOT software companies. They are in other businesses which happen to use software. It makes a great deal of sense for them to lower the development cost by sharing the source.
Very few people make money by selling software. For everyone else, the overwhelming majority, software is an expense, not a profit center. One way to make more money is to lower that expense. Open source lowers the development cost by sharing the cost among users with similar needs.
I think of the software market as three segments: custom, high-end, and mass market. The mass market has huge numbers of users with similar needs. Commercial software companies can spread the development cost over a large pool of customers, lowering the price to commodity levels. However, there are flaws in that model. Distribution in binary form limits the size of that pool to the group with the same hardware and OS platform. It’s still a large pool, but not as large as the pool of users using a standard API on multiple hardware platforms. So providing source enlarges the potential pool of users sharing the development cost, allowing the cost to be spread over an even larger group.
Selling mass market binaries only takes advantage of economies of scale on the selling side. The development side, being closed, is limited to a relatively small number of developers for each company, and each company duplicates the efforts of every other company. Free and Open Source development provides economies of scale to software development. Those economies of scale didn’t come into play until the Internet drastically reduced the cost of collaboration. Programmers can build on each others work, focusing their efforts on extending and improving, rather than re-inventing wheels. Of course, human nature says that some will re-invent the wheel anyway, but it’s not required, so it tends to happen somewhat less than with closed source.
The net effect is that for any market with a large enough pool of users to justify a closed binary solution, that same market will support an open source development solution at a lower developement cost, spread over a larger pool of users, further reducing the cost to the user.
The custom market exists when a user has unique requirements, or thinks he has – he may be fooling himself. Such a user will usually hire employees to write code, or contract out to specialists. In both cases, he generally owns the resulting code. Most programmers work in this market, even though it is largely invisible to the public. It ranges from embedded systems to specialized business software, such airline reservation systems.
The high end exists between the extremes of the mass market and the custom market. It is the domain of complex applications needed by modest numbers of users. Examples would be image processing in the movie industry, or software used to design ICs. The applications tend to be close to the core business of the users, making them reluctant to lower development costs by sharing source. But there are enough users to fund closed development by third parties. Open source is slow to move into these areas, although portions of the application stack tend to drift down from the high-end strategic area to the commodity infrastructure area where open source works best.
The point of this long ramble is that for any market where the sale of closed source binaries works, open source works even better. The best days of commercial software vendors are winding down, as they were a temporary aberration of a rapidly expanding young market. The market is maturing, software is becoming a commodity, and the long term economics favor open source. There won’t be another Microsoft, just as there won’t be another railroad boom. The Internet is doing to closed source what the automobile and airplane did to the railroads. Microsoft will still be around, like the Union Pacific is still around, but new technology has shifted much of the money to new business models.
Ofcourse it’s nearly inevitable to use some kind of closed software if you browse the web.
Nope. 100% Free system here. Plenty of other people with Free systems, too. Did I mention I can watch Flash animations if I want to (GPLFlash)? And that I have 3D acceleration (DRI Project)?
No, it doesn’t. You only have to supply source code to the individuals to whom you distributed your changes. It’s pretty bad when an advocate of the GPL can’t even get the requirements right.
There are two options presented with the GPL: 1) you include the source with the binaries at the original time of distribution, or 2) you include a notice that *anyone* can request the source for (at most) media costs
In the former case, you are correct, but the latter case is more convenient to many companies.
Umm, if you are a software company, how are you supposed to make money if you have to give your software to everybody for free? I mean the point of a business is to make money. This author doesn’t quite address this rather fundamental concern.
1. Nobody claimed the software should be given to everybody or free of charge. The GPL does not require that.
2. Would you consider it acceptable for a company to kill people if their main profit-maker was their assassination department?
So which companies struck it rich in the Internet era?
Amazon, Google, eBay among others. All of them use open source software, yet you bet most of their IP is under lock and key. I can’t download the source code for Google’s search engine, or Amazon’s recommendation system.
Established companies like Wal-Mart and Dell also use homegrown proprietary software as a competitive weapon.
Maybe the best days of shrink wrap software are gone, but the business of proprietary software as service seems to be doing quite well.
Richard Stallman has “nothing against communicating with Google’s network server.”
http://kerneltrap.org/node/4484
Well, i think he really should look at the reverse, what is so good about the GPL. From a business stand point pretty much nothing. They don’t want people working on their software and sending stuff back and such. The especially don’t want others going off and using it.
All the people who want the GPL everywhere and want to see all these companies go to the GPL don’t run big companies, if they did they would for sure look at things totally different.
Furthermore a few companies does not make proof that it works for all. I don’t know of any significantly large and very profitable complete GPL company. Especially one that makes money off software, not just using software as a part of selling their wares.
For Stallman and not using proprietary software, well you can’t go through live without coming in contact with it, everything you interact with through the day is tied to it.
“Maybe the best days of shrink wrap software are gone, but the business of proprietary software as service seems to be doing quite well.”
Yep. You’ll note all of your examples are of companies that make money using software, not by selling software. I tried to make the point that using, not selling, is the key to following the money in software development.
This seems to escaped the author of the article. The GPL makes no restriction on use, the significant issues are regarding distribution. Amazon doesn’t sell the proprietary applications that they’ve developed, so it doesn’t matter if they used libraries licensed under the GPL. No distribution of binaries means no obligation to distribute source. The author appears to assume that the only way to make money from software is by selling it. That’s actually fairly unusual, considering how most development is funded.
I probably should have made a stronger connection between the Internet as the enabling technology behind open source. Sharing source code has been around since computers were invented, but it was hampered by slow and expensive communications. It’s hard to get economies of scale when you are sending tapes around. The Internet makes collaboration much more practical, with near real-time response and trivial cost. So economies of scale show up on the development side.
The Internet has also changed the economics on the distribution side. Lowering distribution costs benefits both open and closed source, but distribution costs used to be a barrier to open source. Distribution costs could be covered as part of the cost of sales for commercial software. Open source didn’t have a good way to allocate distribution cost. With those costs dropping to trivial levels, it’s no longer an issue. Now the other costs of sales for commercial software – marketing, commissions, support, license administration – stand out more.
There will always be closed source software. It’s the best solution for certain situations. But most people don’t grasp that it isn’t best for all situations, because they don’t know any other business models for software development. This isn’t a surprise, as they have no more need to understand the economics of software development than they do the economics of making steel. But they’ve seen one business model, so they think everything must fit into that mold. They’ll eventually learn that all funds for software development come from users, and that passing those funds through third parties with markups and overhead doesn’t make sense most of the time.
Richard Stallman has “nothing against communicating with Google’s network server.”
Big deal. I’m sure he has nothing against using a bank machine to get money either. Assuming of course he actually uses a bank and doesn’t just hide money under the bed in order to avoid supporting an institution or business sector responsible for stupendous 3rd world debt.
If the only food left was dolphin we’d have to fight to stop the eco-warriors hogging the ketchup.
Sooner or later everybody has to compromise a little, it doesn’t mean they’ve given up their ideals. Personally I like the man’s staying power, most comedians don’t have careers this long.
What? You mean my network packets might be travelling through systems using proprietary microcode? … *click*
hac: Distribution in binary form limits the size of that pool to the group with the same hardware and OS platform. It’s still a large pool, but not as large as the pool of users using a standard API on multiple hardware platforms. So providing source enlarges the potential pool of users sharing the development cost, allowing the cost to be spread over an even larger group.
Nonsense. It is quite possible to develop cross-platform software. I’ve known people who have done it. For example, one person I knew of during his spare right wrote a simple library which renders a C program runnable on Windows 3.X, Windows 9X (and NT, 2000, and XP), Windows CE, OS/2, and a number of other platforms. And that was during spare time mind you. (Note: It wasn’t up to QT’s level of quality, but then what do you expect?) It included GUI, threads, and a number of other functions. All you have to do is compile it for each platform. No source modification required.
As if this isn’t enough, with the development of things like Java, you can write closed source programs that run on all platforms that support Java.
In both cases there is no need for the source code to be distributed. None.
hac: Selling mass market binaries only takes advantage of economies of scale on the selling side. The development side, being closed, is limited to a relatively small number of developers for each company, and each company duplicates the efforts of every other company. Free and Open Source development provides economies of scale to software development. Those economies of scale didn’t come into play until the Internet drastically reduced the cost of collaboration. Programmers can build on each others work, focusing their efforts on extending and improving, rather than re-inventing wheels. Of course, human nature says that some will re-invent the wheel anyway, but it’s not required, so it tends to happen somewhat less than with closed source.
Nonsense. A number of companies have offered portions of their work to others at different times. I’ve even licensed some myself (and I’m not very wealthy considering a mere student/teacher and I didn’t get academic discount on some things). Some companies want a lot for their code and that’s to be expected, but others are ridiculously reasonable. Some are even more than happy to share for FREE after a certain period of time has passed. Some at certain times may even be happy to make a group effort on some things.
My point is… With closed-source you do not need to nessecarily reinvent the wheel either.
Also… Despite your proclaimation of how well opens-source works. For a lot of home users, they would rather deal with the closed-source companies that sell software rather than service-based open-source companies. The reason being straight forward…
A lot of service companies for home users these days SUCK. In fact, more often than not, the “evil” closed-source companies often offer a respite from this. You expect these people to enjoy having to depend on the service-based companies even more? If you want that you better come up with a way to help improve service first. (Note: I’m not saying that the closed-source software companies are perfect in this regard. But I’ve found among the regular home users, that the more closed the system is, the more popular it is. The more they have to depend on service the more fustrated they tend to be. To point of simply throwing the computer away and buying a new one whenever something goes wrong, simply so they don’t have to deal with computer repair people. I don’t receive such treatment, but a number of my customers will do that if I’m not available to fix their computer, before they’ll go to another repair person. And it’s not due to my “stories” either but their personal experiences.)
Am I saying that open-source never works? No. I think there’s a number of areas where it does. But I think its best feature for home users, is helping to bring down the price of software and presenting competition for the closed-source companies to ensure that they keep their rear ends moving.
Also… It can be helpful for some code reuse, for whatever happens to be used frequently.
Of course cross platform development is possible. That doesn’t conflict with my point, which is that source is more portable than binary. The binary is created from the source, so the binary CAN’T be more portable than the source. You can reduce the portability of the source to that of the binary as a degenerate case, but the source will never be less portable than the binary. So, to be pedantic, the source has equal or greater portability than the binary.
I agree that the binary form is sufficient for many purposes. The source form is sufficient for the same purposes, and perhaps more. Does it always matter? No, we know that closed source is adequate for some markets. But open source may be better for the same markets. Better doesn’t always win, good enough often succeeds.
Good software can be made using either open or closed source. Bad software can be made with either. Support can suck with either, or it can be good. That’s all beside the point.
I responded to MacManiac’s question about making money by giving away software. You make money with software by using it. That’s where ALL software development funding comes from: the users. The users can fund the development directly, or they can pay others to do it. In the first case, they can save money by collaborating with others, giving the source away. In the second, a middleman hires the developers and sells the software to the users. The middleman has an incentive to keep the source closed as a control measure, because it is his control that allows him to maximize the revenue from sales. Keeping the source closed increases the development cost, but not as much as it increases his revenue.
The user has every reason to give the software away, and the seller has every reason to prevent that. As the money ultimately comes from the user, the user will sooner or later get his way. We are shifting from a seller’s market to a buyer’s market.
Again, my point is not that companies that use closed source are evil, and open source is goodness and light. Companies are pretty much all amoral, regardless of how they develop their software. I just want to explain to MacManiac that giving away source does make economic sense, in some cases more sense than keeping it secret. Open source is nothing more than an application of the academic freedom that created the Industrial Revolution, coupled with the meritocracy that a free market rewards. We shouldn’t wonder why things are open, we should wonder why they are closed.
You only are required to give the source code to the people to whom you provide the binary, that is certainly true. However, you cannot provent them from giving it freely to the other six billion inhabitants of the Earth. So it is unlikely that GPLed code can remain limited to your client base for very long. I suppose if it were in both of your interests to keep it a secret that you could do that. But I think that any contract that required that the client treat the source code as a trade secret would violate the GPL,
hac: Of course cross platform development is possible. That doesn’t conflict with my point, which is that source is more portable than binary. The binary is created from the source, so the binary CAN’T be more portable than the source. You can reduce the portability of the source to that of the binary as a degenerate case, but the source will never be less portable than the binary. So, to be pedantic, the source has equal or greater portability than the binary.
I’ve dealt with some apps that are actually the reverse. The reason being simple, it was unclear how they setup their build environment and what all the requirements were for actually compiling the software. It was actually in a few cases, simpler to acquire and setup an emulator to run an old compiled version of the app, than it was to compile the source. So your argument is wrong compliments of the existance of emulators.
No doubt this case actually often applies to old games like say Ultima Underworld. It is probably easier to get it to work in Virtual PC, VMWare, Bochs or what have you, than to make the source actually compile and run on Windows, Linux, or the Mac.
hac: Support can suck with either, or it can be good.
Agreed. But my point was that from my experience, support for homes generally sucks. As a result, I know quite a few people who would like to avoid it at all costs.
hac: The user has every reason to give the software away, and the seller has every reason to prevent that.
No. The user does not have “every reason to give software away”. One reason is that if I have a certain program and you don’t, I have a strategic advantage. A perfect example of this is a “game cheat program” and a MMORPG. If I have it and you don’t, I can kick your butt all over the place in the MMORPG.
A more real world example, would be if I have a program which somehow improves the customer service for my company. By keeping it out of the hands of my competitors, I have an advantage.
Another reason to not give it away, is that some people believe in sharing not just the fruits, but the cost. If you want it, you’ve got to help pay for it. The reason some people feel this way, is because they don’t want to be “left with the bill all the time”. I know I feel that way myself. I know lots of people who will only contribute if you make them do so. So, I’m only generous with people I actually know are helpful as well.
I’m not telling: You only are required to give the source code to the people to whom you provide the binary, that is certainly true. However, you cannot provent them from giving it freely to the other six billion inhabitants of the Earth.
Indeed. A “competitor” could arrange for your product to be purchased and then either make use of it themselves or if they are not a “direct” competitor, they could have the source posted on the Web.
In my opinion, if you are going to go the open-source route, it is best to just assume that everyone will have access to your source code. You can attempt to keep it to just yourself and your customers and if thats the case then its a bonus, but don’t make that state a requirement for your strategy.
Deletomn: But I’ve found among the regular home users, that the more closed the system is, the more popular it is.
hac: Good software can be made using either open or closed source. Bad software can be made with either. Support can suck with either, or it can be good.
I figured I’d elaborate some more on something else I mentioned (and forgot about in my slightly earlier response. What it is that I forgot is quoted at the top). What I meant by this in particular was things like… The PlayStation 2 vs the “open” PC. Which is easier? From my experience, the PS2 is. You simply plug it in, start it up, and insert a PS2 disc.
I’m big on open standards and what not. But really… Consoles are in my opinion are very close to computers yet a heck of a lot easier to use. In fact, I feel that if Sony made the PS3 capable of wordprocessing and other “computer tasks”, I think people would sooner go with them than with Windows and maybe even prefer it to the Mac.
Since Sony seems to like Linux (they had an add-on kit for the PS2) I think they are actually the most likely path for getting Linux onto the home desktop for the next few years.
PHP and CSS are both open
That was not the point! I can develop a closed programs with it.
Quote: “Umm, if you are a software company, how are you supposed to make money if you have to give your software to everybody for free? I mean the point of a business is to make money. This author doesn’t quite address this rather fundamental concern.”
WTF? I suggest you read the GPL again. You CAN sell software that’s under a GPL. Nothing stopping you. You can also sell your services/support. The GPL works, and works well. Of course, if you’re part of the US corporate style system of screwing over everyone else for a buck, then you won’t like it or understand it. What’s so hard about giving back a little to the community (sharing the src code). It’s long been proven that community style projects that are open are better than closed projects.
Eugenia, obviously you are heading towards the anti Linux/anti GPL crowd by publishing crap like this. I have seen mostly nothing but praising of Macs and Solaris in the past few months of reading your site, and bagging of anything Linux or GPL related. Other posters have seemingly noticed this of late as well, i’m not the only one. Oh hang on, us Linux zealots are all one eyed, two faced, commie idiots. Mac zealots are far worse than Linux zealots by a country mile. Have you got it in for Linux by any chance?
Dave
Thankfully, glibc and the linux kernel have linking exceptions or they would be useless.
TrolTech could be out of business within short time if only open source developers wanted. All you need is an incompatible ABI modification of Qt which conserves source compatibility. You licence it GPL-only, somehow convince major Linux distributions to accept it and done. Proprietary programs would lose ability to run on such system, while the opensource would be unaffected.
To prevent this to happen TrolTech must sponsor healthy amount of money to various people within the community.
If there’s something I hate it’s stallman and his followers. Not only him but all damn advocates of different licenses.
I prefer BSD since I’m not an egocentric paranoid madman… so if others don’t like it… I don’t care. I don’t have any problem with giving, since I’m sure others share my philosophy and it’ll even up in the end. Hell if my code makes it possible for MS or Apple to make better products, go ahead….
The whole point is, stop telling me what to use, I can think for myself, more than I can say of anyone who is a stallman follower
Why do you have a problem with other people saying what they like, why they like it and that they think other people should use it?
Is everyone who reccomends a certain product/style/political standpoint an “egocentric paraniod madman” ?
It is difficult to think for oneself, and come to meaningful conclusions, without information about the subject.
Putting yourself above “Stallman Followers” on the grounds that none of them can think for themselves is just trolling.
People still refuse to see software(and Jobs) for what they are _investments_. When companies hire people or write software(or do just about anything) it is an investment( aka a risk).
Its fairly easy for someone who lives off a trust fund to stand critical of people who won’t have work tommorrow if they make the wrong decisions/investments.
Software is like just about any other investment, you have to pony up the resources FIRST then expect some kind of recoup and hopefully some profit for doing so LATER.
Just because to the end user copying is easy and simple doesn’t mean it hurts no one. Remember that companies have investors to answer to, not just the execs.
Perhaps someone can mail Stallman an economics book…
How important is it that my company’s modifications to open source code remain private?
First ask: Does my company have to distribute binaries based on modifications we made in the source. If that answer is ‘no’, don’t bother the quoted answer and say: ‘Hello FUD!’
Putting yourself above “Stallman Followers” on the grounds that none of them can think for themselves is just trolling.
Yes I know that the general way of discarding someone is saying FUD or trolling or something similar. However, I think you proved the point of not thinking very well. You say “trolling” which is the general “stallman follower” terminology instead of saying something relevant…
Just today I remembered reading Stallman’s interview when he explicitly says he doesn’t use “non-free” software. What I would really like to know is how does he look at something as trivial as Google (just an example, I’m sure you can find better ones). Theoreticly he is using their software (if he does use it) which isn’t “free”. Does he view this as only using their services? If that’s the case, then his ideology is flawed. A lot of free software can be used just as a service on not as traditional software. Many future plans of big software companies (Microsoft beeing one of them) are aiming at making software a service used via the internet. I believe Microsoft is planing the next Office suite to be used via the web – not through a web browser, but through a special client to set up the requiered scripts/GUI,…
This is what I’ve been saying for a while now. Even Stallman isn’t really morally clean of non-free software. Commercial airplanes usually run on HP-UX, so as much as Stallman flies he is using non-free software. Also the public switched telephone network is run on non-free software.
Personally I’ve developed my own idea’s about the issue. I think their is a healthy balance between propreitary and free software. I think it is wrong for the propreitary world to try to kill off free software with patents, but sometimes a patent isn’t a bad idea. How else are authors to be compensated for their unique ideas? I think software patents are better off left undone that done though. Copyright already protects the authors quite enough, and NDAs and licenses allow orginal code to be used quite fine by other people fairly.
I think free software has gotten to the point where it’s great for a desktop replacement, but that isn’t the case for everybody, and isn’t the case for every situation. So the advocates need to stop saying that all software should be and needs to be free, because it just isn’t the case, and they are fooling newcomers to the movement by claiming this. Sometimes software just needs to work, and being able to see the code isn’t really a big deal; as long as it does X job, and does it in a good way that may be better.
And all the people quick to flame Microsoft: Sure I may not agree with some of their business practices, but their OS isn’t all that bad for what it does. Sure Windows has some problems, but so do all OS’s.
In the end I think it’s wrong for the free software people to claim that all software should be free. That’s meaning that they want to take away software that makes our lives easier, which is a dangerous and unfair decision.