Announcement

Collapse
No announcement yet.

Towards A Real Business Model For Open-Source Software

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #91
    Originally posted by yotambien View Post
    Let's avoid generalisations. If you say it's the opposite, you can probably cite a lot of examples where this was the case. I myself wasn't trying to generalise.
    I didn't generalise anyting. I noticed specifically Firefox KDE and Gnome.

    But Mozilla was cited as an instance of an open source product "break[ing] the server/consulting ghetto". I'm pointing out that it's not the best example, given the amount of money poured over it every year. Of course the particular circumstances of KDE and Gnome don't take or give anything to Mozilla, but it's no wonder why Firefox is a killer application and KDE or Gnome are not.
    The bad example is not Firefox but KDE and Gnome, coz they can not be killer applications since they are not applications at all.

    The origin of Firefox is Netscape navigator, which had been under closed development for 4 years when the Mozilla project started. When Firefox 1.0 was released, in 2004, Google already provided 75% of Mozilla's income (around 90% nowadays). Cause and effect are deeply intertwined here.

    But actually Mozilla is a great example in relation with was has been discussed in this thread. It is undeniable that Mozilla is successful and makes a lot of money. It is also notable that the money doesn't come from selling any piece of software, but from striking a deal with an advertising company that uses that software as an avenue to sell their services.
    Again, what's the problem with that? As we pointed before, Mozilla was able to find a great sponsor. A lot of other FOSS apps have sponsor deals as well but not as good as this one.
    The fact is that a FOSS app was able to sign a good deal getting a certain income like that. RedHat sells colsulting and support etc
    It's obvious that there are ways for FOSS to get income.

    Comment


    • #92
      Originally posted by Hoodlum
      I don't think you understood my comment.
      I'm talking about that business model more than the licence.
      You are absolutely right. I thought yours was an argument against BSD licenses.

      So, focusing correctly this time: the point of the BSD-like license is clear. You need such a permissive license to close parts of the project without worrying about legal consequences. I think your concerns about the proposed model are valid, although there is nothing suggesting that those problems would always and automatically arise. This is, it would be perfectly possible to have a project delivered in this way that didn't have the majority of its development behind closed doors, but only critical parts that have the potential to generate revenue. This doesn't automatically translate into a useless piece of code with all the goodies closed (the article mentions 50-80% of the source to remain open at all times, which I find on the low side). As long as a usable part of the code is open, I don't see any fundamental problem here. The time gap between the commercial release of the code and its source, as mentioned in the article, doesn't have to be huge, and it can be adapted to the particular rate of progress in the field of application. Sure, you have reasons to be wary about such promises, but it all depends on who is doing what.

      It is not a black or white situation. You say you don't see this as a way forward for free software, although you acknowledge it could work. Most of us agree there are some gaps in the open source application universe. I don't discard a general change of paradigm in the future, but so far it seems software companies are not dying to fill those gaps for free. If this, or other development model, can help to expand the number of open source applications, the delay between the release of the code and that of the source is, in my opinion, a small price to pay. If Id software released the code of Q3 in 1999 it would be, well, astonishing (it would also mean that Id wouldn't be what it is today). That they released it in 2005 was only awesome.

      Originally posted by Apopas
      The bad example is not Firefox but KDE and Gnome, coz they can not be killer applications since they are not applications at all.

      Again, what's the problem with that? As we pointed before, Mozilla was able to find a great sponsor. A lot of other FOSS apps have sponsor deals as well but not as good as this one.
      The fact is that a FOSS app was able to sign a good deal getting a certain income like that. RedHat sells colsulting and support etc
      It's obvious that there are ways for FOSS to get income.
      Well, I was arguing something the original post wasn't arguing, as you read above. However, within what I wrongly assumed it was being discussed, let me clarify a couple of points.

      First, I don't get what you are trying to say about KDE and Gnome not being applications at all. Of course they are. The distinction between "applications", "sets of applications", "desktop environments" or whatever you had in mind when you wrote that is irrelevant. I named the first two open source projects that came to my mind and compared their success to that of Firefox. Firefox is a killer application and the other two are not.

      There isn't any problem. There are facts. Like the developers of Firefox didn't start from scratch to write a browser. Like when the first version targeted for general use was released, a giant company provided millions of dollars due to their own strategic reasons (despite the browser having less than 4% usage share). Like the fact that this flow of cash continued year after year until today, representing 90% of Mozilla's revenue. So, given all this, I find it amusing that you somehow are able to make a neat distinction between cause and effect, explaining Firefox's sucess on their ability to find a good sponsor or the quality of the application. I don't, and that's why I don't consider Mozilla a representative example of how to make money from open source software or how to create a top notch application.

      You say that "The fact is that a FOSS app was able to sign a good deal getting a certain income like that. RedHat sells colsulting and support etc. It's obvious that there are ways for FOSS to get income". Please, don't stop at the et cetera. I want to hear about all those successful projects that are making money of which Mozilla is a representative example.

      Comment


      • #93
        Originally posted by yotambien
        Well, I was arguing something the original post wasn't arguing, as you read above.
        Forget that sentence, Hoodlum and o0max0o aren't the same person.

        : O

        Comment


        • #94
          I was reading this flame war and thought I had better chime in to try and fight the fire.

          Different people will license their software using whatever means they feel is best for that piece of software. I am personally a hardware / software engineer, in my spare time I write code for several GPL'd projects but when I work for money it's very proprietary, though the code would be completely useless except when running on my hardware doing my jobs.

          Generally if a programmer programmes and is in it for the money he or she will use the proprietary route and patent their software to protect themselves, whereas if they like the community effort they will use something else like the GPL.

          I personally like the combination of the two, as I like to work and earn money but I also love my open source operating system and would like to see it everywhere doing everything I want!

          Personally I like the GPL it means that code I write is kept safe and can be used by anyone in the future. This is what I like to see my code under that I do in my free time, I like trying to get my community desktops working how I like and I like adding to the community.

          I completely disagree with the OPs ideal of a hybrid of open source and closed source projects as the closed source ideal is the complete opposite and parallel to the open-source ideals. The GNU project was started to liberate computing and provide a FREE and open source OS alternative! (note the word free)

          I would have thought that if you tried to fork and make money off of a project that particular project would fork and carry on in another direction causing your modifications to be rendered completely useless and pointless after six months let alone your 5 years.

          5 years is an extreme length of time, think how far the Linux kernel has come in the last 5 years. Basically OP if your software idea is good enough you should piss off to the closed source land and make your millions, if it isn't any good (which I assume is the case) you should open source it and have it fail in this realm instead so we can all watch and laugh!

          Comment


          • #95
            kraftman, I wrote a concise economic analysis of why open source fails in my last comment and all you want to do is blather about your definitions of freedom and anarchy. The GPL is a choice? So is communism. As I already said, the GPL also allows others to do what they want with the code, just as long as they run it on their own server. BSD code competes with proprietary products better than GPL code ever has, hence Mac OS X with BSD code killing desktop linux. It is funny how your emphasis on which freedom is important is constantly changing: with GPL, you say it's all about the code but with BSD, you forget about the code and say the developers don't have freedom to decide whether the code can be used in proprietary apps. Ultimately there are many balancing freedoms here, between the developers and the users, and this hybrid model is best because it balances those concerns the best. GPL guarantees there will be contributors? What are you even talking about?

            o0max0o, I'm well aware of the two different labels of "Free software" and open source, I only use the latter because I find the first nonsensical. A single product like Mozilla means nothing for open source on the desktop, particularly because of their weird model of depending on Google's search sponsorship so much, that is not generally applicable for other software. Chromium is about to kill off Mozilla anyway, so Google finally realized a better way to spend their money.

            Apopas, who's the dogmatist: you, who only believes in pure open source and gave no argument for it, or me, arguing for a pragmatic mixed model that I've written lots of arguments for? Please, take your aggrieved incoherence elsewhere.

            ModPlanMan, a free and open market doesn't imply a communist disavowal of private property. Open source can't encourage anything if there's no money behind it, most innovation still comes from closed source, because that's where it's paid for. Closed source software is far from stagnant: there is much competition out there, which a hybrid model would only accelerate. Precisely what about the BSD license allows one individual to encroach on another's freedom? It's the GPL doing that by forcing people to contribute source when they redistribute the software binary. I have no problem with people choosing the GPL, but I believe this hybrid model shows just how bad a choice that is.

            timrichardson, I explicitly said success is how much the software is used, so your implication that I said it's about money is clearly fabricated. Money helps you get to ubiquity however, like no other factor can. I didn't call this model open source, I called it hybrid source, another fact you got wrong. It's not only on app stores that open source makes no dent, it makes no dent most places, because of the often purist views of its proponents.

            Hoodlum, redundancy is how a market works and is precisely why it works better than any planned system you would prefer: Linus understands this very well. As I've already pointed out, 5 years in medical software is not that long: the released source will certainly be usable, if a bit behind. There will also be development on the open source core that will be funded by these closed-source components, development that will be open source from day one. Rapid development simply implies using a smaller time limit for hybrid source, as I already pointed out with the 18-month time limit for web browsers. As for your claim that "free alternatives to any closed project would be created and improved before the code could ever be released," that is complete and utter nonsense. If it's so easy to clone closed-source code, why don't we have pure open source versions of all closed-source software now? Resources are always limited, I have no fear of such pure open source forks. I don't think you understand the article at all, my whole point is that the service model is a failure: open source has failed precisely because of its service model. The goal is to move to a product model using hybrid source.

            yotambien, good points about Mozilla and generally.

            danwood76, I don't care about the FSF and their four freedoms. I only talk about the practical benefits of open source and how they can be blended with the beneficial aspects of closed source. Is the linux kernel medical software? Since it isn't, what makes you think it would have a 5-year time limit? If you had actually read the original post, you'd know that I'm already using this hybrid model myself. I merely published the concept so that others could also use this better hybrid model and benefit from it. Funny how you admit that if I made it open source, it will fail, (or alternately, that only crappy software is fit to be open sourced ) glad to know that open source proponents like yourself assume failure.

            Comment


            • #96
              Originally posted by Sprewell View Post
              I don't think you understand the article at all, my whole point is that the service model is a failure: open source has failed precisely because of its service model.
              Open source hasn't failed in the slightest. The desktop market is every expanding and the linux uptake in all of this is also increasing.
              On the server side of things linux is also kicking ass!

              I don't use any closed source software on my desktop PC and I will never again. Every piece of software I used to use in windows has an open source GPL'd equivelent that is usually much better.

              On all the servers I deploy I also use linux due to the huge amount of usefull free software available. I also donate to several open source projects that I use most frequently.

              For example one of the most widely used PC media players on the market, VLC, is GPL and is cross platform.

              The problem you have is you are preeching to a lot of people who like the open source and free software they are using.

              Comment


              • #97
                Originally posted by danwood76 View Post
                ... The problem you have is you are preaching to a lot of people who like the open source and free software they are using.
                of course they/i/we like it; everyone loves something for nothing. in fact, i like it so much that i often get seconds when free food is offered to me... delicious ;-)

                ----------

                i'm not hugely proponent of this time based idea, but i think its absolutely critical to merge proprietary with FOSS in a way that can work on a large scale. it needs to be thought about, and yes, it is possible.

                GPL is not the magic bullet to rocket the concept of cooperative software mainstream; it's too purist is some regards (look @ history, uncompromising/purist peoples are never progressive... look at the static GPLv3 has caused), it offers no means for an arbitrary developer's/company's effort to improve a software to receive compensation, nor does it offer a method to generate ANY capital WHATSOEVER for the project using the license. GPL has done great things; we should learn from both its strengths and it's weaknesses. Same for BSD. the service model sucks so much fckn donkey b*lls; to think anything different means you probably have no experience on the matter.

                (service model) != (reoccurring capital)

                it's like working 9-5 (service model, you're only paid when you physically do something) when you could be on vacation and getting paid just for being you (selling a COMPLETED proprietary app again and again and again...) seriously, which would you do given the need to provide for your family?

                in the end, there is little incentive to write FOSS commercially.

                thus, how do we generate this incentive?

                back to licenses. i'm not a legal expert, and the following idea may not be legally possible/enforceable. it all came to me about 10 years ago, when i was 13 or 14, and i was downloading a pirated copy of Autodesk Inventor, a stupidly expensive solid-modelling program. all i wanted to do was design a BattleBot. i thought to myself, "why can't i just use it [Inventor] for whatever i want, and only pay the company [Autodesk] if i actually MAKE MONEY WITH THEIR SOFTWARE? wouldn't they want lots of people to use their software and give it more exposure? What gives, isn't software basically free to copy? i'm only 14 i can't pay $800 to try the damn thing".

                such a concept seems perfect for open source.

                in my mind (maybe not legally, idk) all software is actually free. it's the LICENSE you pay for, and it's the LICENSE that grants you rights to how that software can then be used by you.

                so, we get together, and develop a license that requires companies/individuals who use open source software commercially to give back to the project in the form of a small, proportional donation to an overseer foundation. the more you make with the free software, the more you would have to donate. HOWEVER, unlike investments in traditional proprietary applications, you are GUARANTEED to always be in the positive; you will never have to "pay" for the free software if you were unable to generate any capital by using it. it is risk free. the hundreds of thousands of micro donations would generate the capital needed to pay ANYONE INTERESTED to:

                ) implement features
                ) fix bugs
                ) write better documentation (we all know OSS fails hard here)
                ) fund a fledgling related project
                ) anything else needed in the current ecosystem

                and heres the kicker... instead of fighting BSD's proprietary friendliness, we ALLOW it! if someone wants to fork an OSS app, hack on it, and sell it as a closed sourced product, we let them, and we make them pay SIGNIFICANTLY-ish more for the privilege to make money off it privately. this way, until the proprietary app actually rewrites the ENTIRE codebase, the original OSS app continues to benefit from any success the proprietary app generates. everyone wins.

                if it was setup so an overseer foundation (like FSF) allocated the resources, any old or obsolete OSS apps in the wild still generating revenue (because people out in the world are still making money with it, even though development on the app has stalled/ended) could be used to fund new projects, new features, and the full bellies of anyone involved.

                we need a licence that provides full circle solutions.

                Comment


                • #98
                  danwood76, linux has negligible desktop market share, every company that has tried to get into that market has got out of it or dropped off: linspire, xandros, mandriva, the list goes on and on. The great hope of all the linux fans, android, is actually a mostly Apache-licensed stack (though with a GPL linux kernel), which mobile vendors close up and use in a hybrid model like mine, albeit cruder as they don't employ my time limits to release source. GPL code does okay on the server only because companies like Google or Yahoo can modify whatever source they want on the server and don't have to share those modifications with anyone, as they're not redistributing the server software. I also don't use any closed source software on my FreeBSD desktop, mostly because my needs are fairly light and open source suffices. However, you and I use many services every day, whether Google or most other web services, that are hardly open source. VLC market share is almost non-existent, though not because it isn't good. The problem I have is that people like you would prefer to live in a purist open source ghetto rather than have a lot more open source code produced and subsidized by a hybrid model.

                  extofme, I think your idea might be legally enforceable, but it is a nightmare to implement the kind of monitoring it would require. How would developers track when users made money with their software? That kind of monitoring software doesn't exist today and will take awhile to implement. Software may be free to copy but unless you charge for its sale or use, you won't get back the money to pay for all the time you spent developing it. The license is merely the legal document to help developers get paid, it's not what users are really paying for. As for letting closed-source developers fork an open source app, there are already companies that follow a similar dual-licensing model. The problem is that requires dual-licensors to get copyright assignments for every contribution, so that they can offer a proprietary license also, which adds another layer of work. I believe my model is a "full circle solution" (whatever that means ), what don't you like about my time-limited hybrid source idea?

                  Comment


                  • #99
                    Originally posted by Sprewell View Post
                    danwood76, linux has negligible desktop market share, every company that has tried to get into that market has got out of it or dropped off: linspire, xandros, mandriva, the list goes on and on. The great hope of all the linux fans, android, is actually a mostly Apache-licensed stack
                    What a bull. The license matters (and GPL fits the best here), but there are far more important things like video acceleration, games and specialized applications to Linux have much more market share. It was showed BSD license sucks a lot, so now you're propagating Apache one? It's a real shame Phoronix posted your bull. Where's Apache, MIT or BSD licensed desktop? Nowhere, so why would we like to go back to *BSD market share? At first, we just need GAMES and not crappy bsd or apache licensed crap. Why your stupid theory isn't working for *BSD right now, oh great fan? :>

                    Comment


                    • kraftman, there is probably no point in arguing with an unthinking GPL zealot like yourself, but I enjoy demolishing dumb arguments like yours so I will. What exactly does GPL fit the best? All the GPL distros that focused on the consumer desktop gave up. The reason nobody invests in things like video acceleration or specialized applications for desktop linux is because they can't make their money back from such a communal marketplace: economic models matter. The Apache license is basically a permissive BSD license with some extra provisions about patents and listing contributors, so no, I'm not changing anything. The fact that you're so ignorant about licensing that I have to explain this to you says a lot. There is a BSD-licensed desktop that uses a hybrid model similar to mine: Mac OS X. They release the source for a lot of BSD-licensed code while building their own closed osource stack on top of it. Mac OS X desktop market share is many times desktop linux market share, so my hybrid theory is working great for BSD code: I just want to make it even better by adding this time-limit modification. Of course, all this was already mentioned earlier in this thread, funny that you've forgotten it already.

                      Comment

                      Working...
                      X