If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
No announcement yet.
Towards A Real Business Model For Open-Source Software
Wyatt, sorry to hear anything above a fifth-grade level hurts your mind. If you actually read the thread, I already gave proof of two hybrid models that have done great in the past, Ghostscript and Mac OS X. Oh that's right, you're the high-handed asshat who can't be bothered to read the thread, then demands exacting data that has already been linked to. Anyone who would call my behaviour trolling clearly doesn't know what trolling is, but you're probably right that most people who use that term, like you and kraftman, don't know what it actually refers to and use it as a meaningless epithet, just like kraftman will randomly throw out the words "straw man." I am not really defending my idea here nor spending much effort, I'm destroying the idiotic justifications that people give for the alternative, purist open source model. What did you think would happen, that people would actually have worthwhile objections to my hybrid model?
Hilarious how when I point out how my hybrid model can be tailored for each segment, you then say it only applies to one of those segments. Are web browsers part of the niche segment of medical software too?
F/OSS web browsers seem to be doing fine, considering market share.
I don't care what's the way forward for F/OSS (as I don't care about the F for "Freedom") plus pure F/OSS already is the poor man's option, nothing changes that.
As it is often chosen over proprietary software in many markets, this is false. I say this as someone who works in a field which almost exclusively uses F/OSS and in the scheme of human development, this is not a niche market.
However, my hybrid model will result in ten times more open source code being written, since it will be funded by the closed sections. What do you find indeterminate about a 18-month or 5-year time limit? You really need to think about your arguments before writing them: one post you say 5 years is too long, the next post you say it's indeterminate, which is it?
It is you that has not understood the point. It has to remain closed long enough to make a healthy profit. This could be anything from 1 hour to forever. It is indeterminate, it has to be to be a viable business model. This means the availability of the code is "who knows" and easily forgotten. If you legally bind yourself to a set period of time, nothing stops you, shareholders or future CEOs from revoking this and going full proprietary. This is basically a proprietary model in disguise. You also lose the technical benefits of outside contributions as the project is closed and the open parts are incomplete. Certainly this does not appeal to many OSS people, the very people you are marketing this towards.
Not sure I understand your Windows Vista/7 analogy, but there will always be people who will pay more for the latest OS.
This was a "poor-man's choice" comparison, as an example.
It is true that open source devs usually clone popular software like h.264 or Flash but because there's not much money coming in, what they can do is limited. A hybrid model will give them more resources because of the money coming in from a product model
Whether they would fair any better is up for debate. There are examples that both support and contradict this.
and hybrid code will actually be able to compete with proprietary code like Windows, not merely being happy to have 1/80th the share like the desktop linux fanboys.
What planet are you on? There is no money behind desktop Linux, there is no demand. A small rag-tag of people are trying to "generate" demand. The focus is enterprise, where the money is. Developing a desktop OS *now* is not viable in a proprietary model either. There have been many studies showing the cost to benefit ratio of becoming even comparable to OS X or Windows, from scratch, to be astronomically high at this point. It's amazing it has gotten this far since 1991.
You're thinking too much in terms of the current closed/open source dichotomy so you're happy that there are a few purely open source projects like linux or KDE, even though most people only use closed-source software and as a result there are orders of magnitude more closed lines of code.
Depends how you look at it. On the client side it isn't supreme but many people use Firefox, Chrome, Android, Webkit. On the server side people use it constantly. Either as admins or just as users. I think you'll find a lot of people use Amazon, Ebay, Google, Youtube and many others. All of them use F/OSS.
You seem to be fixated on this desktop paradigm. No one cares about the desktop anymore. Windows 95 won that and XP solidified the victory a long time ago. This is an old battleground, the web is the future.
What I look at is the percentage of lines of code across all software that is open, which is what this hybrid model will grow a lot. When all software is hybrid-source and 70% of it is open, that's a lot more open source code than today, even though very few individual programs will be purely open source then.
The percentage is meaningless. I couldn't care less about whether a cat diary program made by now defunct company in 2001, is open or closed. Also, you seem to be assuming that all proprietary code would suddenly become "mostly open" without an incentive. This is simply fantasy.
There just doesn't seem to be an incentive to giving it away. Sure it may start out like this while your products are insignificant but once you truly hit the big time and shareholders get involved everything becomes incredibly proprietary. No incentive if you aren't benefiting much from the technical benefits of F/OSS.
So they usually spend a ton of money on the hardware and usually just grab the cheapest OS to put on it, which is Linux.
Really? they spend vast quantities of money building the fastest possible system only to stick Linux on there because it's cheap? You don't seem to be in touch with reality here. I think you'll find it's because it's the best thing for the job. It should be, all the vendors spend so much time optimising it for this purpose.
I never said there is no commercial demand for linux: I said carving out 1-20% market share in a few isolated markets
*cough* 40-60% for "isolated" markets like web servers and growing faster than any other platform.
or dominating a fairly useless niche like supercomputers is not a success.
Tell NASA and CERN that.
Funny how you keep saying that open source only doesn't do well in niches, when the truth is that pure open source has only done well in a few niches.
The internet != a niche. Science, the very thing pushing the human civilization forward for millennia, != a niche.
Nobody said open or closed source has infinite resources. Hilarious how you quote me saying resources are always limited, then say that means I think closed source has infinite resources.
You were the one suggesting that, suddenly, by magic, taking up this model would solve all resource problems. You're suggesting this model on those grounds, remember? I'm only commenting.
You are the one who implied something like that when you said that pure open source would magically clone all hybrid code, despite very little money coming in through its weaker support model.
No, you didn't understand my point. There are two groups of people
1) People that care about F/OSS
2) People that don't
The people that do would hate a half-open / half-closed model and would rather create an F/OSS alternative. The people that don't would be your (if successful) shareholders, CEOs who don't see the point of your actions as they generate no appreciable benefit from releasing your IP without getting technical benefits from doing so.
This model hinders you against proprietary competitors while gaining no meaningful support from the F/OSS community. The worst of both worlds.
If F/OSS were to become a second-class citizen, that would be a step up from it's fourth-class citizen status now, not backwards.
There are plenty of non-desktop fields where this is simply false. Not everything has to be on the desktop to be successful.
What part of having an explicit 18-month time limit means releasing "when we can" to you? Funny how you continue to just make up shit.
Yes, I make up things like a need for profit or (if successful) shareholders and CEOs who see little technical benefit to your approach. These are just total fantasy.
Apparently you seem to have some funny notion that you can just release the code on time while losing money, this is unrealistic. Your schedule has to be dictated by market performance. eg "who knows?"
F/OSS will always be the poor man's choice because it's free, that's all poor men can afford to pay for. However, my goal is to lower the cost of software for those paying for it also, by having more sharing and innovation come from the open source parts of a hybrid codebase, and of course the poor can use the much greater volume of quality open source code generated from such a hybrid model also.
A noble goal and I think it may work in areas with no OSS involvement or BSD / PD projects, where contribution is lacking, but I really don't think it will become significant like the traditional closed model or the GPL model for the reasons already stated.
Hoodlum, it would be a waste of time to debunk every fallacy in your post, so I'll just hit the ones that stood out. If a hybrid source vendor has a contract with a customer saying they have to deliver the source in 2 years, nothing future shareholders or CEOs decide can change that, unless they want to get sued, just as companies who use BSD or GP-licensed code and then don't follow the terms of the license can be sued. You actually make a relevant point about how the time limit has to be long enough for companies to recoup their investment, but that's not hard to do. All you'd have to do is set a higher limit like 3 years and then let competition from other hybrid-source vendors drive the time limit down. I am not marketing this idea to purist OSS zealots, I'm writing about it for the pragmatists who will use whatever model works best. Glad to read that you know there's no money behind desktop linux, but I disagree that Windows can't be beat. Mac OS X made a dent with their cruder hybrid model, my time-limited hybrid model can go the rest of the way. All the people using Chrome to access Amazon, Google, etc on their Windows computer are using much more closed-source lines of code in Windows and in Google's proprietary search algorithms every time they do something on the web, when compared to the open source lines of code in Chrome or in the linux kernel on Google's server. So if you do the math, closed source still dominates on the web, as I alluded to in my original article. I never said proprietary code would become open, I said hybrid source code would put closed-source companies out of business. As for incentives, I laid them out already, perhaps you don't understand enough economics to follow? Hmm, so you think proprietary code always wins ("everything becomes incredibly proprietary")? Interesing position, maybe true if not for hybrid models.
You're not in touch with reality if you don't know that most of what drives supercomputers is hardware, not software. Science, specifically supercomputing, is a very tiny niche, please get in touch with reality. The internet is actually composed of mostly closed-source software, from Cisco's IOS on their routers to Ebay and Amazon's closed-source webapps. I never said this model would solve all resource problems, I said it would be more efficient. You were the one magically suggesting that open source would somehow compete with no money. I've already made the case for the open part of a hybrid model, ie sharing and innovation, if you can't understand those benefits, that's your problem. You're probably right that the F/OSS purists wouldn't contribute to a hybrid codebase, but I think the OSS pragmatists probably outnumber them by at least 4:1. Those are the people who will contribute and who don't waste their time yammering about "freedom" like Stallman and you, while Linus was actually pushing a working kernel out the door. I never said linux only failed on the desktop. Yes, you are making up shareholders and CEOs who don't see the benefit, as Apple is wildly profitable with a mixed model and keeps their shareholders very happy. You seem to have some funny notion that hybrid source code would lose money, with no rationale for why that would be the case. Not only do I think this model will become significant, I predict that it will kill off the GPL model first, because that is the weakest model, then it will kill off the pure closed model also. Let's see how soon that happens.
Hoodlum, it would be a waste of time to debunk every fallacy in your post, so I'll just hit the ones that stood out. If a hybrid source vendor has a contract with a customer saying they have to deliver the source in 2 years, nothing future shareholders or CEOs decide can change that, unless they want to get sued, just as companies who use BSD or GP-licensed code and then don't follow the terms of the license can be sued.
You still fail to see the point. If at any point you became successful all future projects would become entirely proprietary at the behest of shareholders. There is no incentive to your model. It amazes me you think this isn't important.
PS. This is why the proprietary and GPL models work, they have incentives.
You actually make a relevant point about how the time limit has to be long enough for companies to recoup their investment, but that's not hard to do. All you'd have to do is set a higher limit like 3 years and then let competition from other hybrid-source vendors drive the time limit down.
This is all academic as there is no incentive for this model.
I am not marketing this idea to purist OSS zealots, I'm writing about it for the pragmatists who will use whatever model works best.
Which is why this model isn't used. There is no incentive in giving the competition your IP while taking limited advantage of the partial F/OSS nature of it.
Glad to read that you know there's no money behind desktop linux, but I disagree that Windows can't be beat.
I never said you couldn't be beat. I'm sure it could, eventually, with massive time and investment. I'm saying it's no longer relevant. Your obsession with the desktop paradigm is very 1990s.
Mac OS X made a dent with their cruder hybrid model, my time-limited hybrid model can go the rest of the way.
They didn't start from scratch giving away their IP for no reason. That would be stupid, they took someone else's.
All the people using Chrome to access Amazon, Google, etc on their Windows computer are using much more closed-source lines of code in Windows and in Google's proprietary search algorithms every time they do something on the web, when compared to the open source lines of code in Chrome or in the linux kernel on Google's server. So if you do the math, closed source still dominates on the web, as I alluded to in my original article.
I never said proprietary code would become open, I said hybrid source code would put closed-source companies out of business.
So, let me get this right....You think that a model where you give away your IP without getting much benefit from the F/OSS community would allow you to beat closed source companies doing the same thing but retaining their IP? Ludicrous. You've intentionally handicapped yourself. They benefit from your code, you don't from theirs, you lose.
As for incentives, I laid them out already, perhaps you don't understand enough economics to follow?
I get the impression you have never worked in a business before. If you only help your competitors and not yourself by your actions your model is a failure. You missed the most fundamentally important aspect of the model entirely.
Hmm, so you think proprietary code always wins ("everything becomes incredibly proprietary")? Interesing position, maybe true if not for hybrid models.
Models that provide a technical advantage, with better resources usually win. This depends on the lead too, of course.
You're not in touch with reality if you don't know that most of what drives supercomputers is hardware, not software.
If you really think we spend years at huge expense building these systems only to go with the least expensive option you're hugely mistaken. We go with the best thing for the job. There would be no point otherwise. The more efficiency we can get the better.
Science, specifically supercomputing, is a very tiny niche, please get in touch with reality.
Science is not limited to supercomputing, sorry to burst your bubble.
The internet is actually composed of mostly closed-source software, from Cisco's IOS on their routers to Ebay and Amazon's closed-source webapps.
*cough* Founded on open tools for use with open protocols. Forgot that part, did we?
I never said this model would solve all resource problems, I said it would be more efficient.
How? there is still no incentive to give your competitors an advantage. Do you really, honestly believe business people will do this out of the kindness of their heart? It is clear you have no business experience.
I've already made the case for the open part of a hybrid model, ie sharing and innovation, if you can't understand those benefits, that's your problem.
Except you forgot to give anyone a reason to give away IP for free. It doesn't benefit you, it benefits your competitors. Your model is essentially proprietary with "goodwill". Microsoft could do this too. Why don't they? because there is no advantage to it.
You're probably right that the F/OSS purists wouldn't contribute to a hybrid codebase, but I think the OSS pragmatists probably outnumber them by at least 4:1. Those are the people who will contribute and who don't waste their time yammering about "freedom" like Stallman and you, while Linus was actually pushing a working kernel out the door.
You relate me to stallman because I understand the need of an incentive in a business model? Wow...talk about defending bad ideas to the death.
Firstly, without stallman is hardly "all talk". Without GCC, Linux wouldn't exist. Secondly, Linus picked the GPL on pragmatic grounds: the requirement to reciprocate. You seem ignorant to the need for an incentive.
I never said linux only failed on the desktop. Yes, you are making up shareholders and CEOs who don't see the benefit, as Apple is wildly profitable with a mixed model and keeps their shareholders very happy.
They took, and give a little back. You're suggesting starting from scratch and giving most of your IP away. Apple don't do this, sorry.
You seem to have some funny notion that hybrid source code would lose money, with no rationale for why that would be the case. Not only do I think this model will become significant, I predict that it will kill off the GPL model first, because that is the weakest model, then it will kill off the pure closed model also. Let's see how soon that happens.
You're suggesting a model without an incentive to use it. This is the problem. There is no benefit to it.
If you actually read the thread, I already gave proof of two hybrid models that have done great in the past, Ghostscript and Mac OS X.
Ghostscript did work with dual licensing and a strict "n releases behind is open" model (in this case, n=2). I'll give you that, though I note that it's past-tense (which is weird, by the way. Usually, the conventional wisdom says "if the shit fits, wear it" or similar). But OS X, I cannot allow you to use that as an example. To do so is disingenuous in the extreme. You can say that Apple's Mac OS X is open source because of Darwin, but people buy Apple devices with OS X because of the proprietary software layered on top, not for Darwin underneath. The key features of OS X tend to relate to their GUI and the applications that come with it; commonly summed as "Mac Experience". Applications that cannot run in Darwin.
Comparing Darwin to OS X is like comparing an apple and an orange to a fruit salad with whipped cream and a cherry on top. If you know as much about economics as you allude to, then you know at least a little about marketing: presentation is god . Trying to pretend that Apple isn't a tower of artifice is at your own peril. Their open strategy serves the twofold purpose of 1) ensuring the ubiquity of their APIs and 2) software that plays nice with their systems . One could probably also make the argument that it's a "friendly" feather in their cap that allows them freedom to take pot-shots at "less open" competitors. In short, it's a nice power play that I can respect and revile in the same breath.
Oh that's right, you're the high-handed asshat who can't be bothered to read the thread, then demands exacting data that has already been linked to. Anyone who would call my behaviour trolling clearly doesn't know what trolling is, but you're probably right that most people who use that term, like you and kraftman, don't know what it actually refers to and use it as a meaningless epithet, just like kraftman will randomly throw out the words "straw man."
Your data shouldn't be buried in a forum thread. Your data should be in the original writing. Supporting your assertions with readily available sources is one of the very basic skills that must be present in any piece of writing that you intend people to take seriously. That's why I say you are high-handed.
You make a lot of words, but that's about it. We tend to call people that do so in a provocative manner "trolls" (non-exclusively, of course). If we really must be pedantic about it, you're probably closer to a kook I'd say.
I am not really defending my idea here nor spending much effort,
That's truly unfortunate. You are aware that Phoronix forums are indexed by search engines, yes?
I'm destroying the idiotic justifications that people give for the alternative, purist open source model. What did you think would happen, that people would actually have worthwhile objections to my hybrid model?
You've never done a thesis defence, have you? I hope for your sake that you wouldn't take this sort of attitude with your panel. It wins you no allies and alienates people that wouldn't even care otherwise.
First off, Sprewell, have you considered the fact that, on a simple ideological level, having a hybrid source model will inextricably make contributors to the open source part of your software "afraid" that their code will become closed source, or that they will be contributing to a piece of software that will later become closed source without compensating them?
These fears may be totally unfounded, and they probably are. But that doesn't change that perception. There is definitely a stigma associated with commercial software in the open source community. I find that projects which distance themselves from commercialization dogma tend to get the most community contributions. Those projects which embrace commercialization dogma tend to have a largely in-house contributor base. In other words, you will be doing 99%+ of the development work yourself, because others will not trust your motives (which is completely irrespective of whether their mistrust is justified or not).
The prime example is OpenSolaris. They had a very nice open source development model, but since they remained friendly to the commercialization of OpenSolaris into Solaris (including blobs and all), they didn't get nearly the amount of community (i.e. not Sun Microsystems employees) contributions they expected. Perhaps you can develop open source software while going into it with the assumption that you'll only ever receive a few odd patches from outside contributors; but then you are not really taking advantage of the open source concept at all, are you?
On the other hand, consider projects like mpd, Quassel, or Firefox. The diverse number of contributors, from corporations, university students, researchers, and just power users, is staggering. These projects do not try to commercialize themselves, even if other people sometimes take their work and commercialize it. The base project itself is only concerned with providing the latest software bits under an open source license.
Also, I just wanted to say, NoMachine is a completely separate example than what you proposed in your article. Their binary blob server is freeware (as in beer). If you want to make a hybrid source program and distribute the closed source parts as free (as in beer), be my guest. I'll happily use it, and I bet that some people might even pay you for the source code. The NX model isn't bad. That small distinction -- of offering their closed source bits as freeware -- is alone enough to mostly validate the hybrid source model in my mind, even if you optionally sell the source code for a price. It's a good enough compromise. Through the rest of this post, you'll see why I think that.
Below, I will be going on the assumption that any access to the closed source bits of your hybrid-source program requires a license fee -- even to download and execute the binary blobs. Please disregard if you are going with the NoMachine model.
My biggest opposition to the hybrid source licensing scheme is the exact same problem I have with pure closed-source software.
Software that is licensed for a fee is, 99% of the time, priced to a level where only corporations, or very wealthy, upper class, first-world workers can afford it. Since the prices are non-negotiable, this means that people who can't afford to purchase the software will not be able to use it. This is _waste_: the potential (used in the sense of "potential energy") exists for the information (software) to flow from its source to others for virtually zero distribution cost, but it is being blocked. The fact that the person blocking it does so because they think they need to make money is irrelevant; the amount of waste generated throughout society is greater than the potential gain that may be in store for the single developer. (Yes, go ahead, quote Spock if you must: "The needs of the many outweigh the needs of the few, or the one.")
When the hybrid source developer has to decide which parts of their code to open and which parts to close, there are two possibilities:
(a) They open-source almost all of it, even the very important parts that provide the "sweet" advanced functionality. The only parts that remain closed source are like the trim on a car: ancillary. The few people who would use this extra functionality would be able to live without it -- perhaps only with a bit of a half-frown when they realize it costs $2500, and they don't think it's worth it.
(b) The developer open-sources a common core (or "base product"), but they withhold important parts that many normal users of the program or library would want to use. The salient point is that the closed source bits are necessary for at least some users of the software.
In case (a), the developer doesn't make the money they want, because most users don't need the trim. They can get by with just the open source version, and make it work for them. Since the hybrid source developer's continued support of the software is contingent on making sufficient revenue, the model fails, and the project reverts to a community-maintained, BSD-licensed product.
In case (b), the user doesn't get the full program they want, because most users need the closed-source blob. They can't get by with the open source version, so they either buy it (if they are in the elite or a corporation) or they go without the software at all. Assuming that, in the (b) scenario, the hybrid source vendor does indeed make enough money to continue developing the software, there is still the matter of the people who go without the software because they can't afford the blob, and they need its features. This is necessarily the case in scenario (b), because if they didn't need its features, then (a) scenario would be the case.
Note that I am effectively discounting the possibility that someone could create a hybrid source program where the only parts that are closed source are exactly those which are always affordable by the only people who would ever use the features contained in the closed source portion. For almost all software domains, this possibility is so remote as to be discounted. The reason is because software (like information) frequently does not discriminate between people with means and people without means. It is entirely possible that someone who does not make a living on software (or, said another way, does not make any money from using other peoples' software) will need every last advanced feature your software offers, both open and closed source.
As a hypothetical example, consider that someone might host a forum on their website. To manage the forum's data, they use FooDatabase. FooDatabase's core daemon is open source, but the management GUI is closed source, and costs money. The management GUI costs more money than this user can afford for their project, because it is something they do in their spare time, and it does not bring in any money for them. If the user had access to the management GUI, it would save them many hours of time, learning console commands and entering them, instead of using the simpler GUI.
Now, discounting the fact that learning to use the CLI is educational and beneficial -- an ancillary artifact of my contrived example that could be eliminated by providing an example where there is no educational benefit to using the open source version -- this is an example of waste.
With closed source software that is sold for a license fee, everything is mutual between the customer and the user: either the customer and the user both benefit (when a license is purchased); or, neither the customer nor the user benefits (when a license is not purchased). This scenario is identical to scenario (b) of the hybrid source model above, which I believe will end up being the more commonly used scenario, rather than (a).
This is a problem because we are creating a society built on willful inefficiency. If I need 5 proprietary software programs to support my hobbyist/community project/small business, and each of those programs' license fees accounts for on the order of 25 - 150% of my total revenue, how am I supposed to support that?
The answer is I can't. My alternatives are either copyleft Free Software, or cheaper proprietary software. But if your hybrid source model succeeds as you propose, then the copyleft free software like the GNU GPL'ed software will go away, because everyone will want to use the BSD license, because the BSD license is better -- right?
Now, as is often the case with other products, commercial software (software that costs money) eventually stratifies itself into a position where competing software is sorted by "you get what you pay for" semantics: cheaper software offers less or is lower quality; more expensive software is more comprehensive or higher quality. This may take a few years to settle down in some markets, and subsidies sometimes confuse the issue; but in the end, quality demands a higher price. Company execs say, "Well, we don't want to leave money on the table..." and next thing you know, the "enterprise" licenses of the dominant vendor (the one with the most market share) are $50,000 per CPU.
So, what's left in a world where GNU GPL software has been all but extinguished by hybrid source and closed source software?
The little guy has to buy the low quality software, or settle with the open source bits that the hybrid source developers so generously decide to leave on the table. They can't compete with the big corporations on their top-level product (for which the software is but a means to an end) because they can't afford the software with "The Works" like the dominant player can. This means that the little guy is guaranteed to remain the little guy: his business isn't software, so he can't improve the software himself; the big guy affords the expensive software, so he leverages that to remain dominant.
And that's just considering companies that actually try to make money. With the advent of the internet, there has been an enormous amount of people using advanced software for simple purposes. Why? Because it makes their life easier. It helps them stay connected. It helps them efficiently do what they need, without giving up their life's savings for a license. The continued success and openness of so many of these projects is contingent on the copyleft model, simply because the preponderance of the users of this software are not willing to pay significant license fees. And why should they be expected to?
Phoronix is a great example. At least in the early days, Phoronix was more of a blog that Michael did for fun. I was there: before all his high-profile involvement with the AMD, Intel and Nvidia folks. Only now is it getting to be a real business with a lot of lucrative opportunities. But I think Phoronix itself is a success story that can be attributed not to hybrid source, but to full copyleft source. Phoronix is unique, and nowadays it's much more than a blog. But Michael depended on having all of the critical functionality of PHP, and maybe even some of the more advanced stuff, available for free. If certain parts of PHP were locked out to him when he was just getting started, he might not have even bothered -- I doubt he could afford a $2500 license fee back then.
Unfortunately, the way I see it, the problem of how much to charge for licenses is also binary. If you charge a fee that is affordable for every conceivable user -- something like $10 for a lifetime, unlimited license -- you won't make nearly the kind of money you need to support yourself. The big guys will not even think twice about your fee, and there won't be enough little guys buying into it to make up for the low unit price. If you charge a massive fee only affordable by corporations, you might make enough money, but you're totally locking out the hobbyists, non-commercial users, and many small businesses and startups.
I write this post from the perspective of someone who creates web applications, administers services like IRC and forums, and works as a software engineer on an MMORPG, all as a volunteer, without making a penny for my efforts. It's something I do in my spare time, for fun, and to enhance the quality of life of my users. They appreciate what I do, and what I do is fun. I don't need massive license fees getting in my way, because I can't justify their cost. When I have to choose between paying a license fee and not using the software, I always choose the latter.
In a world devoid of copyleft software, my choice not to pay for these licenses might end up reducing the quality of my product: in a closed source world, I'd have to look around for scraps of public domain and research software; in a hybrid source world, I'd have to use the community version and hope it's enough, or wait years for the needed bits to be released open source.
Fortunately, the world is not devoid of copyleft software, and it will not be any time soon. So the quality of my product is as good as the community makes the platform I depend on. And that quality is damn good if you ask me -- especially considering that I help develop some of that platform myself.
Disclaimer: This article is written under the assumption that the reader has a shared interest in supporting and providing for the interests of hobbyists, end-users, non-commercial endeavors, startups with little or no investment, and small businesses. If all you care about is the power play of existing large enterprises, none of this makes any sense. This is a big problem with many theories of economy, software-related or otherwise: they don't equally and fairly foster non-commercial, small businesses, and large businesses all at the same time.
Wyatt, yes, Ghostscript switched from their previous hybrid model to a dual-licensing hybrid model earlier this decade, your point is? I never said OS X is open source, I said it's hybrid source because Darwin is mostly open source. As for what people buy OS X for, they buy it for both the open and closed parts, as one is useless without the other. I bought Mac OS X 6 years ago because of Darwin and the unix apps that are part of it and can be run on it, as do a lot of geeks who buy Mac OS X nowadays, so you're wrong that Darwin doesn't matter. So Apple markets well, what does that change about the fact that they're using a hybrid model? The API issues you mention are irrelevant to Darwin, as those APIs are higher up the stack in the closed sections. I don't think Apple keeps code open only for marketing reasons, as I've never really heard them trumpet that much. Anyway, their particular internal motivations are somewhat irrelevant since we can only guess at them, all that matters is that they're using a successful hybrid model. After all your rhetorical meandering around the edges of this issue, you cannot come up with a single reason for why Mac OS X is not a success of the hybrid model. I hope for your sake your thesis defence wasn't so badly argued. My data is buried in a forum thread because it was irrelevant to the original post, where my intent wasn't to prove the failure of linux but to present a new idea. Hilarious how you keep labeling me high-handed when you are the one imposing irrelevant academic requirements on my writing. Yes, you're right that there are a lot of idiots (such as yourself? ) who now use the word "troll" to mean anyone they disagree with and who they feel talks provocatively. It's the inevitable consequence of how most people are too dumb to use language properly, just like how they'll label anyone with a new idea a kook, simply because they're too dumb to understand the idea. I'm not defending my idea here only because the commenters arguing against it so far are mostly too stupid to come up with any worthwhile arguments against it, so if anyone stumbles across this thread through a search engine, it'll be so they can read my takedown of the dumb GPL model that is what all Phoronix members who respond seem to want to talk about. Haha, thesis defence? How many people on this planet do you think do a thesis defence? My attitude is actually a benefit, as it filters away all the reflexive GPL defenders who would never use such a hybrid model anyway, with the exception of the few hardcore dummies who chose to argue with me, and the sort of smarter person who would actually use this idea could care less where it comes from. If they see a good idea, they use it.
As a newB in the world of Linux and the open-source software, I was pleased to find this thread to feed my recent interest for the issue discussed here. There is one question that is bothering me though.
Many people use various examples of successful companies/project to support their claim that one particular business model is superior (one company that is widely cited is Microsoft of course). What i am wondering is: are these companies successful mostly because their business model is the best one, or are they successful because they were at the right place, at the right time with the right business model for that place at that time?
In the case of Microsoft, one could argue that (1) they were among the first in a new market, (2) they had the right product/ideas for that market at that time, (3) in regard for the market at the time, they had the right business model. Eventually, they reached a position of quasi monopoly. Is it their business model that still makes them so successful today or their (so) dominant position?
The quasi-monopoly position Microsoft enjoys today, makes it extremely hard for others to succeed (Apple made it mainly because they went for a niche market and did it really well). But with internet and the new tech, market situations are evolving quickly, sometimes in unexpected ways, and companies' fortunes are changing equally fast. Instead of one Business Model, maybe we should talk about different strategies adapted to specific situations.
Hoodlum, if any successful hybrid project would become entirely proprietary, why hasn't OS X become entirely proprietary?
They took the code, taking advantage of the already existing code and gave a little back. A massive overwhelming majority of development is kept proprietary. To the extent that darwin, on it's own, is completely irrelevant. By contrast you were suggesting giving away a large majority of your work, from scratch. Not taking pre-existing code and keeping the overwhelming majority of future development closed. Big difference.
Your ridiculous assertion is so damn funny because we have empirical proof with Mac OS X that it is false.
The funny part is you actually believe that all the while your ridiculous analogy doesn't work. Apples and oranges
Clearly supercomputing clusters choose linux only because it costs nothing, as closed-source companies would be able to produce a better closed solution easily.
Then why haven't they? Infact Microsoft is investing heavily in this area right now. There is a lot of money to be gained here and yet they are still behind. This debunks the quoted FUD.
Science may not be limited to supercomputing but since you have no evidence of science being dominated by linux, that hardly matters, does it?
No evidence? Oh please. This doesn't require you to have worked at CERN or NASA or even in the field. Have you ever been to a university research lab? Clearly not.
On a personal note I'm pretty confident of this claim, you may want to check my occupation.
The internet may use some open source software, but since it uses a lot more closed-source software, my assertion that closed-source dominates still holds.
Known falsehood. Backup your claims or don't bother making them.
Funny how you've now lowered the bar to whether the internet uses any open source software at all.
These are called examples. It would serve you well in future arguments to use them.
PS. desktop machines != the internet.
Hilarious how you keep repeating "there is no incentive, there is no incentive" like a mantra, yet cannot come up with a single argument against the two incentives I gave in my original post, ie reduced costs through code sharing and allowing outside innovation to easily take place and be paid for.
Reduced costs how? code sharing with who? There is no significant OSS community that contributes (or imo, would be willing to contribute to) closed projects like this. The state of the community involvement with Open Solaris is a prime example of this fact.
As you gain no reduced costs through code sharing and your competitors can just re-use your open code and keep any improvement to themselves. You end up with the inferior product and go out of business, well done.
Come back when you can come up with an advantage to this model. Until then I guess you'll just keep on with infinite faith in your ideological, hippie view that everyone shares for the benefit of the world and no one would exploit the weakness of your model for personal gain. Naive would be an understatement.
That means either a) there are valid arguments against my hyrid model but you are too dumb to come up with them since you clearly know no economics and have no business experience or b) perhaps there is no worthwhile counter-argument to my hybrid model.
or C) You have yet again failed to understand business or even basic logical thought / comprehension.
You seem to believe giving away your investment to competitors that can build upon it and keep improvements to themselves isn't an issue, this is unrealistic. You seem to have this ideological view where everyone shares for the benefit of humanity but this is pure fantasy.
I bought Mac OS X 6 years ago because of Darwin and the unix apps that are part of it and can be run on it, as do a lot of geeks who buy Mac OS X nowadays, so you're wrong that Darwin doesn't matter.
Back to your incredibly niche target again. Geeks make up what percentage of computer users? That's right, almost nothing.
It seems as you have been unable to come up with any counterpoints that replying to you is pretty futile. If you want to show this model to be viable I suggest you try it and come back. I seriously doubt we'll hear from you when you have though