Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Software GNU is Not Unix

Seven Open Source Business Strategies 97

Openstandards.net writes "One of /. posters' favorite discussions is the value of open source as a business model. OSDN has an article on IT Manager's Journal that highlights seven business strategies for open source. " Slashdot and ITMJ are both owned by OSDN.
This discussion has been archived. No new comments can be posted.

Seven Open Source Business Strategies

Comments Filter:
  • by Wesley Felter ( 138342 ) <wesley@felter.org> on Monday May 17, 2004 @11:22AM (#9173657) Homepage
    One thing I noticed about that article when reading it yesterday was that only two of those business models actually include writing open source code. This fits with my thoughts that there's plenty of money to be made from open source, but not necessarily from creating open source.
    • by fbrain ( 758421 ) on Monday May 17, 2004 @11:24AM (#9173676)
      My company implements open source solutions for a small fee, and when asked to install some OSS we always donate a percentage of the fee towards the projects we use. It would be great if everyone 'played nice' but I can't see that happening.
      • by tanguyr ( 468371 ) <tanguyr+slashdot@gmail.com> on Monday May 17, 2004 @11:30AM (#9173732) Homepage
        Forget the money, send the patches you had to write.

        Open source shouldn't have to count on people "playing nice" - it's about enlightened self interest: sharing your patches decreases your costs in the long run, because you can apply other people's patches more easily.

        Convincing MBA weilding bosses of this is more than half the fight.
        • I'm newly MBA-weilding, and hopefully former-sysadmin.

          The good news is that business people ARE starting to hear about F/L/OSS. Within my own classes, I tried to inject presentations about it whenever possible. In my last class, a business strategy class, my prof allowed me to deviate from the normal "find a way to help a business" project to doing one on "what every manager needs to know about OSS".

          It will take some time, but the word is getting out!

          It's not the MBA that makes people stupid. Stupid p

          • The good news is that business people ARE starting to hear about F/L/OSS.

            I, for one, am overjoyed to hear this!
            Business people with things stuck between their teeth are the scourge of the Earth!
          • I think the point that was being made is that a lot of stupid people who wield MBAs use those certifications as proof that they're right.

            In other words, stupid people made dumber by arrogance.
            • That reminds me of something someone else wrote on Slashdot:

              A wise manager once said: "Obviously you want smart, productive people on your project. Note that dumb, unproductive people are relatively harmless, because they are not productive enough to cause much damage. What you need to watch out for are dumb, productive people."
        • by torpor ( 458 ) <ibisum.gmail@com> on Monday May 17, 2004 @12:16PM (#9174134) Homepage Journal

          I think the problem is that its so difficult for 'accountants' and 'evaluators' to really put a value on any open source which a company may end up contributing to with .diffs ...

          With software, there are a number of different approaches for 'valuating' a company codebase and sticking that figure in a spreadsheet along with all the company's other assets, such as account balances, etc.

          With OSS though, how does that valuation occur? Its a public trust sort of scenario - as if the tax which all employees paid the government each year was 'also valuable' to the company, whose cash it was originally before payroll was paid.

          OSS 'contributions' are a strange beast to an accountant, and unfortunately, many companies these days rely on valuations and assessments from 'traditional bankers' for things (such as getting loans to cover payroll, or new inventory for sales seasons, etc).

          I know that EFF donations and all those 'tax-free writeoffs' are also valuable too, but these don't get thought of as 'investment return', generally. So if you put your main codebase development out into OSS, and your accountant wants to write all your primary code off as 'donations and contributions, non-return expected', then it gets a bit weird...

          I think groklaw really ought to spend some time on this sort of thing; the more boring side of supporting linux/OSS on the legal fronts may well lead to a solution to this accounting dilemna, and that would surely be nice for a lot of companies that want to get into OSS ... and still keep the books in order.
          • I have never thought about that, so thanks for the brain jolt. In my job (contract developer for medium-large company), i'm more likely to wind up arguing with the marketing guy who thinks "free software" = "people working for you for free" or the development manager who still thinks "open source means no support" than the accounting guy.

            Over the last year to eighteen months, i've seen the words "open source" become more buzzy around here, but no real understanding of what they mean. Like i like to say: "i
          • Book OSS development expenditures as advertising. There is ample precedent for corporate sponsorship of arts and sporting events. Such investments have a goal of increasing brand recognition and other metrics of corporate reputation. If this approach were adopted by more companies, standard metrics for evaluating brand returns on OSS investment would emerge.
    • by Anonymous Coward on Monday May 17, 2004 @11:26AM (#9173691)
      Indeed you are correct. Amazon and Google are but two testaments to the idea of using open source for business purposes, and neither is exactly a "software" company.
    • by platos_beard ( 213740 ) on Monday May 17, 2004 @11:53AM (#9173932)
      Huh? All except one involve distributing open source code:

      Optimization - give away good, sell better code.
      Dual license - give away good code, sell same code to anyone not wanting to GPL their modifications.
      Consulting - give away code so people will hire you to customize it.
      Subscription - give away code, but get people to pay for more convenient distribution of it.
      Patronage - find someone else to pay you to give away code.
      Hosted - (this is the one) write code based on free software that only you use.
      Embedded - Give away code tailored to hardware you sell.
      • The Optimization, Consulting, Subscription, Hosted, and Embedded strategies don't require writing any open source code; they work better if you just use existing code.
      • No, I agree with him. While they all involve open source software (how else would they get on the list), only one of those business methods actually involved making money off of open source software that you wrote.

        Optimization: Align your proprietary software with someone else's open source software to make the total package less expensive for your customer without giving out any code yourself.

        Dual license: Subsidize open source development with licensed version.

        Subscription: Subsidise open source develo
        • In all of the rest of these the open source software is not the primary money maker...

          Well, yeah. That's the whole point. You don't make money by giving away stuff for free. Ever. These business models are ones where it is to a company's advantage (usually indirect) to develop and distribute open source software. Of course you can be a consultant without writing open source software. I would think that point is so obvious, no one would misconstrue the article as suggesting anything else.

          It isn't hard

    • The ability to modify open source software is part of the value proposition for those organizations that can take advantage of it, in other words those that have in-house programming capability. The advantage becomes less clear when you have to pay someone from outside to do the modifications but may still be less than proprietary code (which, in some cases, may also need customization). So you're right on about the money angle, even if the per-job income is smaller, you can do a lot of customization jobs
    • > only two of those business models actually include writing open source code.

      You know what this means, though: you write the code, and somebody else gets the profit. Is it any wonder Open Source is so popular? It lets everyone live off the geeks for free. Most geeks are just interested in writing code (and good code, mind you, not some boring database screens for those customization scenarios) and are not likely to actively pursue any of the mentioned strategies. They don't even have enough time to try
    • Good SOS BS examples, and all have proven examples of ROI value.

      However, the examples only show ways that businesses can exploit OS "software products" for ROI. Businesses using these SOS BS will outlast/outperform other business making the proprietary "stovepipe" (one choice, no options) mistake.

      The big businesses of the future will discover ways to use/exploit the real international value of Open Source which are the methods, values, processes, ... of the Collaborative Open Community (COC) that promotes
    • I don't have code contributions to return to the community, so I give money to the The Perl Foundation" [perlfoundation.org]. I then get the company to pay for it via our reimbursement system.
  • Data Strategy? (Score:5, Interesting)

    by Mr. Sketch ( 111112 ) * <<moc.liamg> <ta> <hcteks.retsim>> on Monday May 17, 2004 @11:25AM (#9173685)
    It was a nice read, but I was curious about one additional option, the 'Data Strategy'. This is specific to games where you may develop the engine using open source tools and release the engine as open source, but charge for all your data files the engine uses for your particular game. Or has this just not been proven as being a viable model?
    • I think that's a version of the optimization strategy.

      The problem with any strategy that involves giving away code and making money on something else is that there's a first-mover disadvantage. You're better off waiting for someone else to give away a game engine and then building games based on it. (Although I could be wrong, considering how little I know about business.)
      • Well, the question is whether the first-mover also gains credibility beyond all the others in customizing / optimizing the codebase.

        The question is whether the credibility and knowledge gains outweigh the cost of the original development - and that varies from product to product. In games software, it probably doesn't- but in monitoring systems, it may.
    • by A. Pizmo Clam ( 779689 ) <apismo_clam@nOSpAm.yahoo.com> on Monday May 17, 2004 @11:47AM (#9173892) Homepage
      This is specific to games where you may develop the engine using open source tools and release the engine as open source, but charge for all your data files the engine uses for your particular game.

      Let's take a look at gaming development under the GLP. It offers an excellant case-study of how GNUlatic-ism will destory America.

      1. NetHack: This game promises to connect your computer to a "net"work and expose it to "hack"ers. This is not a good model for a law-abiding company

      2. Frozen Bubble: I'm not up on the street-lingo these days, but I'm 101% sure this is a drug reference; perhaps to "crystal methane" or "the acid". Do you want drug users working for your firm? Is that a way to be profitable? Also, I believe it comes from France, that notorious hangout for Maoists and ne'er-do-wells.

      3. Klotski: Like many pinko names, it begins with a "K-" and ends with a "-ski". It's also affiliated with a sinister group of known East German "hackers" [kde.org] whose avowed goal is to "conqer" the American workplace.

      I could proffer further examples, but any healthy-minded citizen will clearly see that GLP games encourage roguery of the worst kind and are unsuitable for emulation by the large, dependable corporations upon which our Constitution is based.
    • Considering that nobody has really done this (there's one that I know of, can't remember the name, but the one that ran Tribes) with an engine that anybody would actually use, I think it's safe to say it has not been proven. It is certainly attractive, since game companies spend a ton of money reinventing the wheel by creating yet another game engine, but I can't imagine someone open sourcing one of the very popular ones because they derive a ton of revenue selling that engine to other game companies.
    • KDE's KVisio program is free with no stencils, and you must buy stencils to do anything useful.

      See here [thekompany.com]

  • don't follow ours
  • Adapt... (Score:5, Insightful)

    by carvalhao ( 774969 ) on Monday May 17, 2004 @11:30AM (#9173730) Journal

    I found this article quite insightfull. One of the paradigms of modern business is the outsourcing model, and that is due to a recurrent need to reduce overhead in any kind of business (overhead always looks like bad news for stock investors). And software aquisition and maintenance IS a major overhead in any IT oriented enterprise.

    The obvious solution is to get the resources as they are needed. And that business frame fits perfectly on the OS business model. That, and not Open Source fundamentalism, is what may make or mars OS.

    Therefore OS must continue to focus on enhanced flexibility and customization, not offered by other non-OS platforms, as a way to thrive. Then, let us consultants do the rest of the dirty work :)

  • by G4from128k ( 686170 ) on Monday May 17, 2004 @11:31AM (#9173743)
    Unless you count WalMart/Lindows as embedded, the only business model that seems remotely geared toward consumer desktop is a subscription model. I can see how Linux provides multiple sound business models for b2b, but wonder if any company can make money off consumer desktop linux.

    Any thoughts?
    • by DR SoB ( 749180 ) on Monday May 17, 2004 @11:35AM (#9173789) Journal
      Yes, with a few simple steps:

      1. IBM dumps tons of money into marketing Linux (any distro..)

      2. IBM upgrades OO..

      3. IBM allows blatant pirating of OS, and offer's free (or next to free) training. Pirates = Free training..

      4. IBM reduces price of x desktop model that comes with Linux.

      5. IBM sell's business licenses because now it's desktop version is well known to market/consumers/PHB's..

      6. Profit!
    • by idfrsr ( 560314 ) on Monday May 17, 2004 @11:42AM (#9173862)

      I would imagine that in order to successfully do this you need be a hardware company. If you can provide commodity hardware with commodity software that presents a viable consumer alternative (read marketable to average joes as the best thing since a toaster for this new sliced bread stuff) then you might be able to make it. This is really a marketing problem rather than anything else. Generate a kick ass device by intergrating open source solutions with commodity hardware and your product development is much cheaper.


      For example the L600 [neoseeker.com].

      In the end, to be successful with this strategy you have to be a kick ass marketer to deal with the big guns at the top of the food chain.

    • by pavon ( 30274 ) on Monday May 17, 2004 @12:07PM (#9174047)
      I don't think there is much money in developing a Free home desktop OS. There is money in packaging and supporting it, and you can certainly subsidize some development using that money. But more importantly, there is money in the business desktop, and if you look at the current state of affairs, the business desktop and the home desktop are extremely simular. Furthermore, most people aren't confortable switching to a new system once they have learned windows, but having used open source software at work will remove those concerns (assuming it was a good experiance :). So I really don't think there will be much of any demand for the home desktop until OSS becomes more prevelent in the workplace anyway.

      In other words, not having a viable business model for the home desktop is not important, because it will ride on the coat-tails of the business desktop.
    • It's the "optimization" strategy for WalMart. You build desktop computer hardware, and sell it. If you put Linux on it, all of the software is free; if you charge a little less for Linux than for Windows, your margin is still higher for Linux. For Lindows, it's the subscription model.
  • by 16K Ram Pack ( 690082 ) <(moc.liamg) (ta) (dnomla.mit)> on Monday May 17, 2004 @11:32AM (#9173764) Homepage
    Does anyone have any more details on this?

    Like if I submit a fix/enchancement to MySQL in the GPL version, can they as the 'owners' put that in their commercial license which their customers can release without the source code?

    • by Anonymous Coward
      No. They would require a copyright assignment from you, as seen in projects such as GCC and Cygwin.
    • Quick answer: No (Score:4, Informative)

      by pegr ( 46683 ) * on Monday May 17, 2004 @11:38AM (#9173821) Homepage Journal
      Like if I submit a fix/enchancement to MySQL in the GPL version, can they as the 'owners' put that in their commercial license which their customers can release without the source code?

      Assuming your fixes are GPL, they cannot put your changes in the commercial version without first negotiating a separate license from you.
    • by Wesley Felter ( 138342 ) <wesley@felter.org> on Monday May 17, 2004 @11:40AM (#9173836) Homepage
      Like if I submit a fix/enchancement to MySQL in the GPL version, can they as the 'owners' put that in their commercial license which their customers can release without the source code?

      If you (foolishly) assign the copyright on your code to them, then they can do whatever they want. If you keep copyright to your patch, they can only use it under the license you have chosen.
      • I think they rewrite all submissions to keep their codebase 'clean'. If they take rough ideas from the submitted code but rewrite it from scratch then AFAIK they have copyright.
        • That sounds reasonable. Of course, with bug fixing, identification is often more important than cure.

          Question is, do you think many developers would allow their code to be dual-licensed?

          In other words, if I made a piece of software that was quite useful, but dual licensed it, do you think a lot of people would be happy to have their GPL code dual-licensed (for reasonable payment).

          Personally, if I were selling a dual-licensed product, I'd probably rather just pay people for their fixes and assignment o

          • I suspect thats why organisations that dual-license probably have to rewrite contributions or get explicit assignment of copyright. Given the administative hassles of keeping track of the copyright issues with a large number of contributions I would expect that this explains the 'rewrite contributions' model.

            I would not want to be involved in surviving legal due diligence for a company that was careless about re-using contributions without managing the copyright issues.

      • by Anonymous Coward

        Like if I submit a fix/enchancement to MySQL in the GPL version, can they as the 'owners' put that in their commercial license which their customers can release without the source code?

        If you (foolishly) assign the copyright on your code to them, then they can do whatever they want. If you keep copyright to your patch, they can only use it under the license you have chosen.

        I don't see what is foolish per se about allowing them to use their license on my code, considering the huge benefit I got from gettin

    • Your fix will not get implemented, is my guess, unless you agree to the potential licenses that they use. Since you used it as GPL, you really couldn't claim copyright on it any more than they could. It is an interesting question, though, cuz by nature the fix you make could not be licensed in a dual nature unless you originally had the non-GPL version, which I'm guessing you would not. If you started with the GPL you couldn't relicense it as dual-license for them to reincorporate back into their dual-li
      • As someone else commented, the only way would really be to analyse GPL changes, rewrite them and implement them in a non-GPL version, and then release the amended non-GPL code as the new GPL code.

        Sounds complicated, and a pity as I can see dual-license as a really good option from all perspectives.

  • D'oh! (Score:1, Funny)

    by deadgoon42 ( 309575 ) *
    There's an open source business model?
  • Cygnus (Score:5, Informative)

    by Henrik S. Hansen ( 775975 ) <hsh@member.fsf.org> on Monday May 17, 2004 @11:52AM (#9173927) Homepage
    Cygnus was actually the very first company to deal exclusively in Free Software (back then the term 'Open Source' was not yet coined). The company's founder, Michael Tiemann actually got the idea from the GNU Manifesto [gnu.org], which outlines ways to make money from Free Software.

    Cygnus primarily provided support, but I think they also did some development. Maybe someone can elaborate?

    In my mind, Cygnus is a good example of how a small company can survive on just dealing in Free Software. Many people need support, or perhaps need custom-made changes to Free Software.

    AFAIK, Cygnus is now part of Red Hat, and Michael Tiemann is CTO of Red Hat.

    • In RevolutionOS, Michael Tiemann talks about porting GCC, working on Emacs, and doing some other GCC related work.
    • Re:Cygnus (Score:3, Informative)

      by PhilipPeake ( 711883 )
      I was a Cygnus customer - yes, they did do development work and provided support for the GNU compiler which was much better than you would get from any commercial compiler vendor -- particularly hardware vendors (SUN, HP, IBM, etc).

      One of the problems that they always had was that they were "customer driven", and tried to evolve the GNU compilers in the direction that favored their customers. This put them in direct conflict with the "official maintainer" of the GNU compiler suite, and, from time to time,

    • Michael Tiemann's name is in at least one file in the gcc source code [gnu.org]
      Judging from this changelog [gnu.org] there have been various people from cygnus who have contributed.
    • AFAIK, Cygnus is now part of Red Hat, and Michael Tiemann is CTO of Red Hat.
      And Red Hat seems to be backing away from the whole Open/Free concept. FIFO, I guess.
  • by turnstyle ( 588788 ) on Monday May 17, 2004 @11:57AM (#9173969) Homepage
    Demand more allowance!
  • software dumping (Score:2, Interesting)

    by lkcl ( 517947 )
    "Contributing software to the open source community alone was not sufficient to save the successor to the Netscape browser."

    very very interesting. recently i just ran Microsoft Office under Crossover-Office (Wine with codeweavers improvements) and on a pentium 400 with 128mb of RAM it took 5 seconds to load a word document.

    the SAME document took OVER A MINUTE to load with OpenOffice.

    open source software does not automatically mean better.

    as a community we are almost entirely dependent on the goodwill o
  • OSS Support company (Score:1, Interesting)

    by Anonymous Coward
    My company (www.idreus.com) supports OSS projects and most everything we sell and support is OSS based. Even down to the firewall/vpn appliances from snapgear based on uclinux. We have and will continue to donate to the projects we make money off of.
  • by bzipitidoo ( 647217 ) <bzipitidoo@yahoo.com> on Monday May 17, 2004 @01:44PM (#9174952) Journal
    From the article:

    The Mozilla project continued to deliver buggy, late releases

    That can be said of most successful software projects. Why is the article picking on Mozilla especially? Because a superficial look at their Bugzilla database makes it seem like Mozilla has lots of problems? IE is worse.

  • I want to start a company writing F/OSS but do not want to go broke. This is a lifesaver for me! Better than a starburst, which is what I had before and it does not last as long as a lifesaver. Anyway thanks for the info.
  • After reading this, I finally understand why my company is not putting Linux on the desktop. It's because all the linux IT people are making such a killing for the company in the server space, they don't want to waste their resources on the desktop, which everyone needs to use, but in terms of revenue generation (for my company at least) means little.

  • Comment removed based on user account deletion
  • Serious question, Why do articles linking to OSDN owned sites ie: NewsForge, ITMJ, etc have to always have a remark like: "Slashdot, , are owned by OSDN" ?

    • Serious question, Why do articles linking to OSDN owned sites ie: NewsForge, ITMJ, etc have to always have a remark like: "Slashdot, , are owned by OSDN" ?


      It's common practice in journalism, so that the reader is aware of any potential bias caused by the relationship. So, I'm not sure why slashdot does it.

2.4 statute miles of surgical tubing at Yale U. = 1 I.V.League

Working...