Marc Fleury Responds to Scott McNealy on J2EE Open Source

Discussions

News: Marc Fleury Responds to Scott McNealy on J2EE Open Source

  1. Marc Fleury, founder of JBoss has posted an answer on jboss.org to Sun CEO Scott McNealy's charges that Open Source is destroying the J2EE industry's revenue model. In the interview, Fleury argues that "J2EE needs Open Source in order to remain a stable market and to avoid being eclipsed by .NET".

    Read Marc Fleury's Response.

    Read the original post Sun's CEO McNealy Suggests Open Source Could Be Hurting J2EE.

    See more Marc Fleury in TheServerSide's Hard Core Tech Talk with Marc Fleury.

    Threaded Messages (96)

  2. <snip>...there is one ad from Sun that particularly made an impression on me: the "Innovators are always controversial" campaign from JavaOne 2002. If Sun really believes in continuing to embrace innovation, they need to adapt to this new Open Source software production mode. </snip>

    Remembers me of a quote from Hannah Arendt:
    "The most radical revolutionary will become a conservative the day after the revolution. "
  3. I am software developer trying to make Enterprise Software for a Living. I will show how Open Source Products can help a software developer built enterprise software at a very very low cost:

    1) Operating System: Linux - FREE
    2) Application Server: Orion/JBoss -FREE1
    3) Java IDE: Sun's Forte - FREE
    4) Database: HSQL/Cloudscape etc - FREE
    5) Free ware for Source Management (CVS - Concurrent Versioning System
    6) Unit Testing - JUNIT - FREE
    6) Log4J - Logging API, Java SDK 1.4 - FREE
    7) Deployment Tool - Sun J2EE Ref. Implementation -FREE

    At least you can use some of these tools which are all free for development of Enterprise Solution. Sun should understand the reason why Java became popular is because of developers and if developers cannot get tools to update skills then java platform would lose the market.

    If Sun wants to make money then use Services model as mentioned by Mark.

    As regards to .NET it will vanish in air as there is no chance it can compete with Open Source Community and second most important reason is it has not electrified the Developer community which is very important for any software to take off in market.

    The Key to any software market is the DEVELOPER BASE.
  4. Open Source is important even for the big J2EE shops. IBM builds software on top of Open Source. WSAD->Eclipse, HTTP Server->Apache, etc...

    Views my own, not IBM's
  5. I think I agree with Marc Fluery. Microsoft is so dominant in the software industry that open source is the only thing that can stop them. In the long run, I think it will be hard even for them to compete with something that's free. Open source cuts off their air supply in the same way that they strangled Netscape with a free browser.
  6. Why do we have too look at everything from M$ context ?

    Open Source is great for innovation and technology, period.
  7. Marc Is Right[ Go to top ]

    If JBoss did not exist, I'd be moving toward .net right now on real production work. Thanks you JBoss Group.

    J2EE will still be strong on big projects in the short-run. But, without open-source J2EE tools, .net will nibble away from the bottom until there is not much left. Sun has to embrace open-source J2EE or Intel/AMD/.net will continue to erode the market that I assume is already shrinking.

    For Sun it is time for offense, not defense. Using defense the last few years and trying to use the legal system against M$ has failed for Sun. Sun has completely lost focus. Period.

    I do see a glimmer of hope in Sun releasing the LX50. That is a sign of life. A little guts instead of playing it safe. What I want is an LX50 (and maybe a bigger brother) with a certified JBoss bundled with the system ready to run in a cluster.
  8. What i think is that the Open Source is here to stay! I am very happy with all the open source! I definitely dun agree with Scott
  9. What i think is that the Open Source is here to stay! I am very happy with all the open source! I definitely dun agree with Scott
  10. Big companies like IBM may benefit from open source because the make money in the service and hardware sector. What is about small companies? Will any of them be able to survive without license revenues? Is the open source movement pushed by big players like IBM in order to suppress small and innovative companies?
  11. Not a bad piece at all. I got worried when I started reading the bit about the "OpenSource lifestyle", capitals and all, and I was bracing myself for the blue pill/red pill schtick, but no! Marc managed to stay away from any dubious Matrix analogy. Phew.

    Re. "spending zero dollars" on marketing and the "excellent reputation", please, don't assume developers are dumb. The JBoss project (Marc) has been very successful at generating viral marketing and publicity. That cost something, even if it wasn't paid in out-of-pocket cash.

    And at the end of the day, this may be what I most admire JBoss for. I mean, hacking good software is one thing (which probably most people here can do), but hacking the hackers! <admirative whistle/>

  12. I don´t know market state of Sun´s products, but Sun is today something thanks to Java, Linux, Netscape/Web and Open Source. Without these technologies, Sun would have dead or would have gone to a narrow market niche as SGI boys (it would stay at scientific and telco niches).

    - Linux: has supposed a rebirth and an impressive improving of UNIX-like technologies. Without Linux, UNIX (Solaris) would be very rare thing and very ugly to the end user in the enterprise, because NT/2000/XP (with very increased stability) would throw away. And remember: Sun sells UNIX machines. Why Linux is so important? it´s free.

    - Java: a S.O. on top of other S.O.s, other great opportunity to sell UNIX machines. Why Java is so important? it´s free.

    - Netscape/Web: minimizes the platform dependencies, and hides the server side technologies. other great opportunity to sell UNIX machines. Why the Web is so important? browsers are free (thanks to Netscape initially).

    - Open Source: promotes innovation, transparency, and **minimize the risk of adopting a new technology**. Java is so important today because a myriad of Open Source programs and tools. It´s the first step to sell versions with the best of bread as Sun does with Star Office, Sun wins a dolar thanks to Open Office version. Why Open Source is so important? it´s free.


      Free/Open Source software **promotes broad adoption of technologies**, then if you need the best of bread, buy propietary software. Propietary software sells confidence and support, confidence and support is very dificult to sell with Open Source. Open Source is a risk to mission critical applications, because nobody ensures that work will continue, will be done correctly ... There are very compelling Open Source projects (JBoss is one) but there are many buggy and dead Open Source projects too.

      McNealy remember the:
      - hardware war: Microsoft/Intel won over UNIX boxes.
      - S.O. war: Microsoft Windows won over UNIXes
      - development tools war: Microsoft VB/C++ won over Borland.
      - browser war: Microsoft IE won over Netscape.
      - Web server war: Microsoft IIS won over Netscape Server.
      - server pages processor: Microsoft ASP won over Cold Fusion.
      - J2EE & .Net war: ... ?

      





  13. Apache won the Web server war, not IIS.
  14. As far as I know Game Theory is not John Nash's. It is John von Neuman's.
  15. As for John von Neuman being the originator of Game Theory, you are correct. I think Marc is talking about the Nash Equilibrium--a certain set of solutions to Game Theory scenarios where more than one player wins, while maximizing individual gain.
  16. Isn't it funny how Sun's attitude changed from JavaOne, where they were embracing Open Source software, and now they consider it a threat.
  17. <snip>
    As for John von Neuman being the originator of Game Theory, you are correct. I think Marc is talking about the Nash Equilibrium--a certain set of solutions to Game Theory scenarios where more than one player wins, while maximizing individual gain.
    </snip>

    You are exactly right. The fact that this needs to be explained makes me wonder about the general intellectual ability of some of the posers making comments here about Open Source.
  18. Another poor decision from McNealy. In my opinion he also commited a big mistake when decided not to turn Java specs an ISO standard.

    I wonder if IBM is going to buy Sun since McNealy is doing such a good job of depreciating its market value.

    Nicholas, I think you are right but Nash's equilibrium was a really big contribution to Game Theory (he received a Nobel prize for this) and then I think Marc Fleury's citation is very acceptable.

    Best regards,
    mau.
  19. WHERE IS THE MONEY?[ Go to top ]

    MAybe most of you don´t want to listen this but...

    If Sun doesn´t make money (and it is not doing such thing), there is no Java. It´s so funny to say "yeah, open source is cool"...while the half of the java programmers are losing their jobs... because the big ones are not selling their Java app servers...because the people prefer something that is for free.

    If I was Sun I would not allow the open source "commercial" projects.

    Bruno





  20. WHERE IS THE MONEY?[ Go to top ]

    Man, if Sun doesn't make money they'll all be in deep s**t, and it wont certainly be open-source's fault, since Sun's business is, above all, hardware and support services, not J2EE licences.

    As for the "not allowing open source projects" - it's almost sad to ear your grining. Stop saying nonsense, and thank the open source community for half the software that runs your PC, your network access and your information.

    God, can't we raise the IQ for a little while?

    Hugo
  21. WHERE IS THE MONEY?[ Go to top ]

    If Sun doesn´t make money (and it is not doing such

    >>thing), there is no Java.

    I disagree. If Sun doesn't make money, there is no *Sun*. Java would live on as C++ does. IBM would swallow Java in a second and become an even better steward IMHO.

    >>It´s so funny to say "yeah,
    >>open source is cool"...while the half of the java
    >>programmers are losing their jobs...

    Disagree. Programmers are losing their jobs for many reasons, yes. But Open Source is not one of those reasons. Despite the success of open source in developer communities, big dollar business has yet to acknowledge open source as a viable alternative (with the exception of Linux, and we know that Windows developers aren't losing their jobs). Money is not being lost to open source app servers (look at the Market share).....

    >>because the big ones are not selling their Java app
    >>servers...because the people prefer something that
    >>is for free.

    Disagree. IBM and BEA don't seem to be having any problems, open source or not. Between the two they have 70% of the market (10x what Sun holds). The truth is that Sun is just looking for an excuse, a scapegoat, somewhere to point their finger...to push of the blame for the the failure of SunOne.

    If I were Sun, I would become a primary software company and START TRYING in the software space. Why??? Because there's less competition.

    In the hardware/OS space there is: Microsoft, Linux, IBM, HP, etc.... And Sun only has an *equal share at best*!

    In the enterprise software development space there is:......Sun and Microsoft. And right now Sun has the upper hand! Why not leverage that rather than *lose* that?

    Cheers,

    Clinton.
  22. WHERE IS THE MONEY?[ Go to top ]

    In the hardware/OS space there is: Microsoft, Linux, IBM, >>HP, etc.... And Sun only has an *equal share at best*!


    >>In the enterprise software development space there >>is:......Sun and Microsoft. And right now Sun has the >>upper hand! Why not leverage that rather than *lose* >>that?


    This idea sounds very interesting and I think Sun can sell much more hardware/OS for large project if J2EE beats .Net

    As I seen many messages in this forum, I think the free JBoss app server can beat .Net for the mid to smaller projects if JBoss is certified. Also, if JBoss really beat .Net, big app servers like WL and WS will become the ONLY reliable solution for the large project.

    As I haven't seen any people comment on Clinton's opinion, I just want to know if I am just being naive or missed something.

    Edmond



  23. WHERE IS THE MONEY?[ Go to top ]


     Hi ,
          Actually open source implementation of J2EE is entry level implementations of the clients who are more cost concious.One of company I know is going to deploy it's J2EE application on JBoss and as the % of usage increases...it is thinking of shifting to IBM Websphere.So I feel open source is not hurting J2ee, it's way that people are embrassing J2ee Technology.

    Ankur
  24. WHERE IS THE MONEY?[ Go to top ]

    I guess we are many that disagree.

    1) Java can easily survive without SUN. Either
        as a independent project a la Apache or IBM
        can buy Java.

    2) The number of Java programmers writing J2EE
        applications must be several orders og magnitudes
        larger than the number of Java programmers
        writing J2EE application-servers. And even though
        the last group may shrink due to JBoss and the
        like, then the first group should grow due
        to cheaper infrastructure. So JBoss is creating
        jobs.

    3) I find it a little bit funny that you blame
        JBoss for the recession in the US/world economy.

    4) If SUN kills the Java Open Source movement, then
        MS will win and both SUN and Java will disappear.
  25. WHERE IS THE MONEY?[ Go to top ]

    If Sun doesn´t make money (and it is not doing such

    >>thing), there is no Java.

    If Sun doesn´t make money, it is their own fault. They need to make applications and products that people want.

    People buy WebSphere, Weblogic or Oracle because the products are good, has a level of professional support that the customer is comfortable with, and is well integerated into the vendor's suite of products. I think reputation also plays a big role in a product's success.

    Time for Sun to get their rear in gear and write good applications.



    Their recent releases of Linux servers is a step in the right direction. Cheap Linux servers running on Intel are what alot of people want right now. They can bank on their reputation and experience in the server market to create a standards based product that offers a better value than any of their competitors.

  26. As for the original topic, well, Mark is taking a high-stake shot at Sun. It may look wild, but, strange as it seems, it's probably the best shot for both companies.

    Better or not that it's competition, JBoss is undoubtly the most widespread open-source app server in the arena, and certainly the most elegantly designed and executed. I have been following (and using) JBoss since it was called EJBoss, and had it's lists on e-Groups - anybody remembers that? - and the amount of quality work put on the server equals or suplants the work put on any commercial server on the planet.

    If Sun wants a standard that can uphold a strong brand, they need to show the market that the standard can hphold itself - even against it's major commercial implementors: BEA and IBM. Otherwise, the standard easily bends in favor of the vendors' preferences, instead of those of the market.

    JBoss is crutial here. It has a strong presence already, it's rock-solid quality and does follow the standards BLINDLY. It's my reference implementation, allright.

    I wonder when will it be Sun's. For our own good.

    Hugo
  27. Take a marketing perspective[ Go to top ]

    If one wants to defend J2EE from being surpassed by the .NET platform, to my mind the most crucial factor will *not* be the cost of the diverse server implementations.

    It is the ease of maintenance of available tools and/or the services (and quality thereof) which are offered to companies for making applications within those servers work. Compare a one time spent 50K (for an app server) to the cost of maintenance for one year (plus the costs of down time and other software related problems) and you will see what route many companies being able to do the easiest cost calculation may take.

    With open source software developers can even enhance the quality of services because they are able to get to know the implementation better or even improve on given solutions.

  28. Open Source is bad for business, the industry, innovation, and J2EE.

    The comparison with microsoft is very relevant - Open source will put commercial software companies (e.g. SUN) out of business just as M$ put netscape out of business.

    While this sounds like market forces at work, it is really two different models for software development. There are many issues with the Open Source model that make it undesirable:
    1.) No central point of responsibility. (who is responsible for patent infringement, or liability?)
    2.) Decoupled from user economics. (No economic incentive to enhance functionality.)
    3.) If community income is based on support/service/hardware then there is a bias against stronger products in favor of a hard-to-use product.(e.g. IBM)
    4.) De-values intellectual property. (like China bootleggers)
    5.) Open source does not have the resources to drive standards and adoption. (e.g. Java/J2ee,Web services, SQL, etc have all been driven commercially)

    Matt
  29. Matt: “Open Source is bad for business, the industry, innovation, and J2EE.”

    Whose business? Open source has been very good for the business of Yahoo, Google, Amason, IBM and thousands of others. This very site uses a number of Open Source products, was that bad for their business? I would also argue that Open Source projects are great sources of innovation and help the industry (and users) to reduce costs and be more competitive.

    Not to be rude, but your message sounds like a typical case of MS’s FUD.

    -- Igor
  30. Hi Matt

    I don't see your problem. I guess you won't say that science is bad just because they share information publicly without patents etc. so that the science community can grow and produce better results. The same applies to open-source.

    Some comments to your points:
    1) Responsible is how wrote or publish the code. Lately I heard many CEOs saying "I didn't know" and so there is no central point of responsibilty.
    2) Strange but JBoss, Apache, Linux are driving and adding new features as we speak.
    3) Have you ever used WebLogic. JBoss is very strong because it is much easier to use than WL.
    4) You are kidding, right? Everyone how wrote a piece of code in an open-source project has the intellectual property on it even he/she allows others to use it for free. It is unbearable for me to see how you compare honest open-source developers with criminals like bootleggers.
    5) Get real, man. Check out JDom, JSR-77, JMX etc. and especially for JBoss we created standards like hot-deployments (copied by BEA WL), JMX as an intra-application communication bus (which puzzled the Sun's JMX team that this was possible), service archives to deploy like EJBs etc. On the other hand SQL is not a standard, it is more a mess and if you don't believe me try to use date or times. The so called SQL-92 standard is not usable today.

    The only thing which is bad for business are biased developers/managers who just need a scapegoat to save their butts because they are incompetent or lazy.

    Have fun - Andy
  31. I don't see your problem. I guess you won't say that >science is bad just because they share information >publicly without patents etc.


    You can't patent a scientific discovery for good reason. Inventions are different.

    > so that the science
    >community can grow and produce better results. The same
    >applies to open-source.

    No. It doesn't. Discovering how the world works, while sometimes helpful, does not necessarily have any economic value. With Open Source you take something that has some economic value and make it freely available. The effects of this are long-ranging and, from my perspective, disasterous. Especially since in providing something for nothing you have short-circuited the normal economic processes that help to encourage future technological innovation and adoption(engineering, marketing, evangelism, standards work, sales, etc). To think that all of this would continue based on services or hardware revenue is delusional.

    (If you need another analogy think of governments that print money to stay solvent. It works initially with no bad effects, then inflation starts, and finally the whole financial system crumbles.)

    Just as the patent system was designed to protect innovators' investments, some sort of restriction should also be placed on Open Source.

    Matt (having a blast!)

  32. <snip>
    some sort of restriction should also be placed on Open Source.
    </snip>

    Your comment betrays your lack of understanding of Open Source. Open Source will succeed or fail based on the economics of the situation. No artifial restrictions are needed. If it succeeds, then you are proven wrong. If it fails, then there is no need for restiction. The fact that you think there should be a restriction betrays your real belief that is will succeed !!!!
     
  33. Matt: “The effects of this are long-ranging and, from my perspective, disasterous.”

    Again, disastrous for whom? It definitely has not been disastrous for Google, Yahoo, IBM, Amazon, eTrade and thousands of others. In fact, they have saved millions by using Open Source.

    Matt: “Especially since in providing something for nothing you have short-circuited the normal economic processes”

    Pure FUD. BEA charges for the server and gives you documentation for free, JBoss charges for the documentation and gives you the server for free. Both are great products and, as a user, you have more control and more options to choose from.
     
    Matt: “To think that all of this would continue based on services or hardware revenue is delusional.”

    IMHO, it’s delusional to think the license driven model will be the only one to survive (Hint: Linux has been around for more than ten years).

    Matt: “... some sort of restriction should also be placed on Open Source.”

    You are contradicting yourself. If you believe that it is “delusional” to think that Open Source will survive, why do you need restrictions?

    -- Igor


  34. Money, money ... ok money is important.

      I think is compatible Open Source and propietary/closed they must live with.

      Open Source doesn´t break the market per se: certain FREE applications yes:

      - Browser Market: was breaked by Microsoft offering for free MSIE against Netscape Browser, but MSIE is NOT Open Source (so it´s a strong problem to security), is a bet to control the Web, that minimize platform dependency on the client, money was secondary.

      - Web server market: was breaked by Microsoft offering for free IIS against Netscape Web Server, but IIS is NOT Open Source (so it´s a strong problem to security), is a bet to control the Web, that minimize platform dependency on the server, money was secondary.

      - Server Pages Processors market: was breaked by Microsoft offering ASP for FREE against Allaire´s Cold Fusion (now Macromedia), that minimize platform dependency on the server. Again is not Open Source.

      - Streaming on demand market: was breaked by Microsoft offering Windows Media Player for FREE against Real Player, key technology on content delivery. Again is not Open Source.

      - Enterprise Server market: Microsoft is offering .Net framework for FREE against J2EE! It doesn´t includes Visual Studio .Net, but it includes all necessary to develop .Net applications. Again is not Open Source.

      These FREE not Open Sourced products of Microsoft are backed by a VERY STRONG enterprise with >90% of S.O. and Office tools market share, are a strategy to destroy all kind of significative technology that suppose a threat against their monopoly. Of course strategic not free products use these free products.

      The result is: monopoly consolidation. Microsoft only software is not good to software market, one product of each kind of problem, not competence: a small bunch of programmers, managers, distributors, controling totally the market. It is the TRUE problem, not Open Source.

      Is very very difficult to compete with free Microsoft´s programs because:

      - These tools are good enough (ok not ever) and ready to production.
      - Is very difficult to make a better tool.
      - You can´t sell, make money to pay programmers.

      Besides: piracy of Microsoft programs have promoted strongly their products as they were free. Microsoft initially was not worried because it promoted broad adoption of their products. Windows is today considered as free. It has changed with Internet registration etc because the monopoly is a fact.

      Open Source: is the one and only hope to open the market. Linux-Windows, Open Office-MS Office, NetBeans/Eclipse-Visual Studio, Mozilla-MSIE, Apache-IIS, Tomcat/JBoss-.Net Framework.

      Note that these open source projects are backed by enterprises including JBoss.

      And making something of money if you can: Star Office (Open Office based), Forte (NetBeans based), WebSphere Studio (Eclipse based), WebSphere (Tomcat based), OEone´s Dektop/ActiveState´s Komodo (Mozilla based), consulting (JBoss guys), support (Linux-Red Hat) ...

      And many people developing end user custom solution with low cost to middle and small enterprises.


  35. Sun should drop its application server and sell JBoss. They could enhance it and integrate it into SunOne.
  36. Hi Matt

    The question is now is Software a science or not. How much code is entered into commerical software based on open-source code or development like HTTP, DNS, SMTP and their implementations like Apache web server, bind, sendmail etc. The entire internet would not work if it would be open. A patent on one of the core components would have destroyed the Internet.

    Let's see if Sience has no economic value like Superconductors, Lasers, Rockets etc. On the other side biological companies started to patent Gens and genetic altered DNA which is a bad thing in my opinion as well as Software patents are.
    I just remember the lawsuite of Apple against Microsoft about the Look and Feel of windowing OS. Of course Apple stole it from Xerox. So companies are stealing from others as well as from open-source as well.

    Back to your argument (again) that open-source is something criminal. But check out other parts of the economy where companies are giving away free stuff (you get a free cell phone if you sign a year long contract) to get other services/products sold. In open-source this is similar. We give away the code to create better software and getting on top of this service through services or additional software. So open-source helps to promote new business through better and robuster code (see the Internet as example).

    Finally I want to talk a little bit about quality some people used here often and easily. Quality is the ratio between what you pay and get in return. So high "quality" paper to print color photos is not high quality if used in a restroom as toilet paper.
    So open-source is the highest quality because you don't pay a dim (no commerical software can match this). But the open-source-ness increases the quality because it provides faster and more frequent reviews and tests so that bugs are found and fixed earlier.
    I don't say that all open-source software is high quality but the process itself has the potential to do so.

    I hope you enjoy the party. The good thing about open-source is that our party continuous and we will prevail. As Lutries and HP stopped their J2EE development and Sun is desperatly seeking to stay in business. Sun has more stopped projects and succeeding ones.

    Andy
  37. I'm not really an advocate of open-source or closed-source. However, the viability arguments that Rolf and Matt are making seem to be based a very limited knowledge of market economies.

    I have a somewhat interesting perspective on this since I was once an economist (working for the "bad guys") doing economic and financial analysts for the Maritime industry which, for a long time, has been protected from antitrust regulations. Now I'm a software developer. I left my career in finance to work in software development because of the appalling nature of finance, especially in antitrust and legislation (I'll speak to this more in a moment).

    On top of that, I started at Unisys doing proprietary development, and worked a lot with NT, IIS, ASP, SQL Server (I even had the oh so lovely pleasure of being a presentor at condex New York for the release of SQL Server 7).

    Now I work in software research, and I work quite a bit with open-source stuff.

    Of course many people have vested-interests in their companies' well being (and the industry segments within which those companies operate). And this is certainly understandable. They don't want to see their niche go away or become mitigated in any way. However, over time, it's just a simple fact that structural/cyclical economies change. Whether or not this applies here, it's still a fact of economics over time.

    From an economic standpoint, people who push one model over the other are really the problem. The models themselves are not the problem. By legislatively pushing one model over the other, you are creating an artificial economy that serves only you.

    Again from an economic standpoint, this is bad for society as a whole. Only in a (relatively) free and unregulated economy will society as a whole make the greatest economic gains, and the economy will operate more efficiently (this isn't marxism, it's capitalism at its purist: efficient market hypothesis, semi-strong form. You can look it up).
    Special interest groups who advocate for special economic benefits only end up benefitting their group. This is simply protectionism, whether by international boundaries, or by legislation within an industy.

    This creates the situation (as in antitrust) where Marginal Costs can operate above Marginal Revenues (which by economic definition is inefficient: look that up too). It happens in special interest legislation all the time. For example, with the lumber tariff, the national (US) market for lumber is selling at an artificially higher price (I know, the price of the deck I'm putting on my house went up by 27%). While it doesn't come under the heading of the Sherman Act, it is sort of an antitrust protection, allowing US marketers/manufactureres an unfettered share of the market. The same holds true for regulated software, or requiring somebody to use one thing over another (I'm not speaking here to technical or other societal mores).

    Those of you who fear open-source have this sense that your traditional economies are in some jeapordy. Whether these fears are well-founded remains to be seen. But by enacting legislation, all you are doing is creating a false economy in your sector. It does end up causing the economy to be less efficient, and therefore hurts the economy as a whole (as witnessed in the Timber example: people in the timber industry save their jobs, but other americans pay out the nose. We're robbing peter to pay paul in these instances).

    It's not the legislature's job to protect your job from a new economy. You, of course feel that it is. But the legislature's job is to protect the economy as a whole. Of course this is wishful thinking. Our good senators from Washington and California have donations and constituents directly tied to these interests, and therefore are compelled to protect the interests of their constituents, above the interests of our nation as a whole.

    So, your economic arguments are effectively bunk (of course I don't think it's a great idea to require governments to use open-source either, at least from an economic standpoint. There may be other governmental and societal reasons for that as pointed out in an the Peruvian dude's letter to MS).

    The thing to do is let these economies play out without interference from either side. Over time, the industry will recognize which is operating more efficiently, and will slowly move towards that one. Look at IBM, they've gone from an extremely proprietary model to a mox of both. Their doing OK. AND, they'll be ready to jump one way or the other should one of these things shake out.

    As to MS specifically, they are a convicted monopolist, and as an ex-antitrust-economist (who worked for the bad guys) I know what that means, and none of it is good. They are, quite simply, bad for the economy. Monopolism stifles competition and innovation (yes that's true), and a whole lot of other Bad Stuff.

    I think open-source _is_ changing the software economy a bit, but it's hard to say what the net change is. It's hard to say that in the long term, the two can't coexist, or if one will win. I have personally witnessed some interesting economies gained through the use of open-source.

    Some people seem to fear that software will become a "service-oriented" economy, but why? This isn't a bad thing. Economists have been worrying about this for ages in the US, but things seem to be OK in that respect. In some basic financial ways, service oriented economies are great. Asset structures tend to be quite a bit different though, and there are all kinds of theories on that.

    So are the legislative types saying that it's our government's job to prop up your industry? If so then whey isn't it the government's job queally to prop up an open-source developer's jobs.

    I guess my final flaim-bait statement is that special interests who gain economic advantages through legislation, monopolism, or oligopolism are the real problem here, not open-source, or closed-source models.

    -Newt
  38. Thank god someone put some sense in all this. Thanks for the excelent post.
  39. <snip>
    Only in a (relatively) free and unregulated economy will society as a whole make the greatest economic gains, and the economy will operate more efficiently.

    It's not the legislature's job to protect your job from a new economy.

    The thing to do is let these economies play out without interference from either side.

    I guess my final flaim-bait statement is that special interests who gain economic advantages through legislation, monopolism, or oligopolism are the real problem here, not open-source, or closed-source models.
    </snip>

    Yes, yes, yes. Very well put.

  40. Hi Jason,
    I am not sure where to start -there is so much to criticize about the validity of your post....

    Econ 101:
    To start with, a history lesson for you: The proven model is capitalism (American Dream, Wealthiest nation, etc.). The Ideal form of Capitalism is laissez faire a la Adam Smith. However, since people are not ideal this caused Monopolies (not good) and Depressions (worse). Hence the Smith approach gave way to the Keynesian philosophy (in the 30s) of a regulated economy based on maintaining certain conditions in markets that smooths cycles, reduces risk, and encourages investment.
    (Patents are the same idea as this but put in place by Thomas Jefferson circa 1800.)

    Your belief that:
    >From an economic standpoint, people who push one model >over the other are really the problem. The models >themselves are not the problem. By legislatively pushing >one model over the other, you are creating an artificial >economy that serves only you.

    is flawed. With most any law (even patent laws and anti-trust laws) there are those who will benefit and those who will not. However, the goal is to implement a policy that is going to create the most value for the economy as a whole.

    There are many examples of this:
    Anti-dumping laws (selling goods at below cost).
    Farm subsidies
    Zoning for buildings
    Doctor and Engineer licensing
    CAFE standards for automobiles
    and on and on and on.

    Even though these "economies" may not be optimized for efficiency. They are unquestionably superior to the un-managed alternative.

    Matt
    'In the long-run we are all dead' JMK
  41. Matt, I don't think you have a good understanding of economics in the real world. You use anti-dumping laws as an example, yet anti-dumping is a prevention to antitrust.

    Mostly what I'm arguing _is_ Kenynesian economics.

    What you say about any law benifiting one group is simply not true. How about the Securitues act of 1933, or the Securities and Exchange Act of 1934, or the Sherman Antitrust Act. All of these laws designate things like due diligence, information releases for securities; in essence a level playing field.

    I agree with you that Farm subsidies aren't a great solution, they _are_ the type of legislation I'm referring too. I already noted the unfortunacy of localized constituencies on US economic policy and legislation, which the farm subsidies are largely a result of.

    Of course there has to be a balance on rules and efficiency, but you're Smith argument doesn't hold water.

    And yes, public safety sometimes overrides efficient markets. Doctors and licensing certainly applies, leveraged speculation in the derivatives markets is dangerous also.

    NONE of the things you have just pointed out really apply to the subject at hand anyway. Software business and publicly traded companies do work fairly well in the semi-strong efficient market arena. Government intervention only causes inefficiency here, and is usually at the behest of localized constituents. The software business does not require issues of national security to protect it as the Maritime industry did.

    It is a well accepted economic fact that efficiency comes where Marginal Revenue = Marginal Cost. And it is well documented that these legislations and monopolies make this imbalanced. That is in fact a core piece of Keynesian economics.

    I think you have a lot to learn about how finance and economics works. You're spouting some stuff it sounds like you marginally understand. Heck, I only marginally understand some of these things, but it seems I understand them better than you as I actually worked in these arena.

    You seem to be arguing for and against JMK in the same posting.

    -Newt
  42. I think we are close to agreeing...

    >You use anti-dumping laws as an example, yet anti-dumping >is a prevention to antitrust.

    ...in the same way that open source controls would be a prevention of "the open source equivalent of a monopoly" (a very likely possibility if too many ignore the risks of Open Source.)

    >The software business does not require issues of national >security to protect it as the Maritime industry did.

    Maintaining the viability of our software industry and keeping it safe for investment IS of paramount importance to our national security. The export restrictions on encryption technology attest to this.

    Thanks for the discussion,
    Matt
  43. No, giving something away for free is not dumping by economic or constitutional standard. And, there is still consumer choice, to buy or not. With this choice, there is now application of the dumping standards, _or_ anticompetitive practices.

    Dumping specifically implies that the violator must be selling/releasing the product with the intent to undermine competition at a place where they are losing money (MC < MR). Meither of these standards applies to the development and release of the software. How do you measure either the marginal cost or revenue? How do you prove any anti-competitive methods. I don't see any open-source projects price colluding, dumping, tying or doing any of the traditional anti-cometitive practices.

    I think you need to study the dumping standards a bit more.

    -Newt
  44. oops, that should have read (MC > MR)

    -Newt
  45. WE ARE NOT COMMIES!

    Matt et al are misunderstanding Marc Fleury's main point. The whole point is that OSS introduces a different business model based on services revenue rather than licensing revenue. IBM is already reaping huge revenues based on services and OSS. If you look at the revenues of any big or small ERP system you will see that 90% of their revenues come via installation and consulting charges, NOT licensing fees.

    If Sun is dying, it really is their own fault. If they had built a services business around their Java technology, they wouldn't be whining incorrectly so much about losing revenue to OSS.

    Also, this model of giving the software away for free and selling services around the software really ends up with more innovation and a higher quality product. The core developers of JBoss have been grouped together into a for-profit company offering training, consulting, blah blah blah...Since we are all out in the field getting direct input from customers we tend to write better code and to provide our customer base with features they want to see.

    The licensing model tends to isolate the developers of the software project from the actual customer base and there is a huge disconnect. How many core developers of WLS or Websphere actually answer support calls, provide consulting, teach people? With JBossGroup the core developers provide these services. Who knows the code better than the developers themselves? (sorry for the sales pitch, but hopefully you're getting my point here.)

    Bill


  46. Actually I am only worried about my own company, not Microsoft, they will always manage. It is not difficult to foresee a future where MS and OSS together has viped out all other commercial software..

    Regards
    Rolf Tollerud
  47. Actually I am only worried about my own company, not

    >Microsoft, they will always manage. It is not difficult
    >to foresee a future where MS and OSS together has
    >viped out all other commercial software..

    Actually, it is very easy to see a future without Microsoft. Companies are starting to dump MS Office in favor of cheaper more sensible solutions. See the following.

    http://news.com.com/2100-1001-955383.html

  48. Dream on, buddy.
  49. "Actually I am only worried about my own company, not Microsoft, they will always manage. It is not difficult to foresee a future where MS and OSS together has viped out all other commercial software.."

    Unless your company produces core infrastructure software I don't think you need to worry about OSS wiping out its market.

    In some ways, infrastructure does operate best as a natural monopoly (like the fact that there is usually one power company per region or that most roads and highways are maintained by local and national governments).

    Not that I'm a Microsoft fan, but I do recognize the benefits they provided by standardizing the PC OS. It was when they began to abuse that monopoly by using strong-arm tactics to grab everything around and on top of the OS that they crossed over to the dark side, in my mind. However, OSS is not about software licensing revenue and does not have the growth pressures shareholders place on commercial companies. Nor they don't have the power to lobby governments or strong-arm their competition. OSS is much more likely to stay in the infrastructure domain where its free and open nature is more likely to benefit the proprietary tier on top, so I think you will see more peaceful coexistence there. After all, JBoss has proprietary software partners like AltoWeb and Gemstone.


  50. Chip,

    So? In fact a competitive product to ours is already no 2 most active in sourceforge, and Microsoft are soon to release a similar product that probably will be selling for a price somewhere around 20% of norm.

    Regards
    Rolf Tollerud
  51. "In some ways, infrastructure does operate best as a natural monopoly (like the fact that there is usually one power company per region or that most roads and highways are maintained by local and national governments). "


    Do you consider the "core infrastructure" being an implementation(BEA/JBOSS/Websphere), or just a set of standards(J2EE&WebServices)?

    I am skeptical that a single implementation would actually work for everyone. It would likely become as unresponsive as a government bureaucracy. Also, what would be the revenue stream for maintaining such an implementation (services?) What would prevent them from abusing their monopoly?

    Matt
  52. Matt,
    I consider the basic features of a web application server to be infrastructure. If the market is down to 2 players, let's say BEA and IBM, they can both set the standards and they would have every incentive to NOT offer portability. If an implementation has a strong enough marketshare and they can offer leading-edge features, they will dictate the standards.

    But of course there's no danger of that happening with JBoss, since in your estimation it is currently filling up junkyards.

    No one is saying a single implementation should work for everyone. If a product does become a monopoly like Microsoft and it becomes abusive, it should be regulated. However, JBoss, Linux, MySql and Apache are not monopolies. They all face strong commercial competition. If anything, they prevent monopolistic companies like Microsoft from attaining an abusive position.

    Services and branding are the revenue stream JBoss appears to have chosen to maintain and grow their implementation. RedHat is also about branding and service. I think MySql takes a different twist and sells an Enterprise license in addition to services.

    As for Open Source projects, if the revenue stream doesn't come in and the implementation stagnates, at least the code base lives on in the public domain. When proprietary software like HP-AS goes belly-up, the users are stranded with fewer options.
  53. Hi Bill,

    Sorry I don't get your point. Here is why.

    The better the software the less services are required.
    (this is the true definition of software quality -think of why japanese cars became so popular in the 80's).

    A services-biased revenue model is dysfunctional and counterproductive to this goal for the product involved.

    WLS wouldn't be the leading J2EE app if it was only as good as websphere.( BEA is not a marketing company) And no one would pay for a J2EE server if JBOSS was any where close (in quality) to WLS or even websphere.
    Matt
  54. Matt,

    Sorry I don't get your point. Here is why.

    The software to car analogy is quite flawed. You are confusing maintenance with service.

    When you buy a car, it fulfills your need right out of the box, er dealership. You put the key in the ignition, fire it up, and it takes you where you want to go. Do you see people dropping $2K to spend a week getting trained on how to use a Honda? I don't think so.

    Software is not like this. When you buy software, you get a shrink wrapped box, a file, a disc, whatever. This does you no good until it is installed and configured. And no matter how "easy" the software is or how smart the purchaser is, some time and knowledge is required get things up and running. Even more knowledge is required to get it running optimally.

    So, would I, a business owner, be better getting an app server and spending my time/money tinkering with it so get it running, or paying an expert to give some help?

    Also, your argument of WL/WS vs JBoss is flawed as well. If WL/WS were so spiffy and easy to use, why have my employers shelled out thousands of dollars to send people to training classes? Why should these classes even exist? Why are their WebSphere consultants? By your argument, commercial app servers should not need any services, they should just kick ass right out the box, right?

    The software service model is out there with the big dogs already. Why shouldn't open source developers get a piece of the pie, too?

    Ryan
  55. This is a good discussion....

    >Why are their WebSphere consultants? By your argument,
    >commercial app servers should not need any services, they
    >should just kick ass right out the box, right?

    Ideally, yes. The state of the art is not there yet but flexibility, ease of use, intuitiveness, simplicity, and freedom from defects are all important qualities for a product to have. (However, they are unprofitable qualities if the product is depending on a services revenue model.)

    You are right, the car analogy is flawed (as are all analogies). If App servers were cars you would find that the local "j2ee junk yard" would be full of late model JBoss's and Websphere's.

    For example, look at Dell:
    http://www.bea.com/press/releases/2002/0514_dell_print.shtml

    Matt
  56. Matt,
    "You are right, the car analogy is flawed (as are all analogies). If App servers were cars you would find that the local "j2ee junk yard" would be full of late model JBoss's and Websphere's."

    This was a good thread...however, your latest post drags it into the gutter. Before it was speculated that you worked for MS, now it looks more like BEA. Back to meat-head vendor salesmen posts :(


  57. And no one would pay for a J2EE server if JBOSS was any

    >>where close (in quality) to WLS or even websphere.

    Depends a lot on how you define quality.

    As far as I know, then JBoss is ahead of WebSphere
    in functionality.

    I have not seen any evidence that JBoss should
    have more bugs than WebSphere.

    The "quality" you pay for in WebSphere is:
      - the IBM reputation (which can be good to
        persuade managers)
      - more (and maybe better) documentation
      - IBM's support capabilities
      - the fact that IBM is very likely to be around
        in 5 or 10 years - and they usually keep
        supporting their stuff


  58. Maintaining the viability of our software industry and

    >> keeping it safe for investment IS of paramount
    >> importance to our national security. The export
    >> restrictions on encryption technology attest to this

    An interesting point that has all sorts of implications for all of us who live OUTSIDE the hegemony of the US.

  59. I agree completly.

    Let’s take a company (A) that has developed a new software product through an expensive trial and error process as is normally the case with innovation.

    Now they have a business and each year they release a new quality version of the product with a completely new code base, very often there is not a single line of code is exactly similar to the previously version.

    But as the case is, without the originally expensive development process, + all the years accumulated experiences of real life deployment, they could not have released any new quality product. So the valuable assets of the company are not really the code base, ***but the dearly costly earned knowledge***. Example for dummies:
     
    Cost for a new company (B) to make a new similar product ***without*** the accumulated knowledge of company A = $10M.

    Cost for a new company (B) to make a new competitive product ***with*** the accumulated knowledge of company A = $1M.

    Now consider that an employee finish his job in company A and sets up his own business to make a competitive product. With the knowledge he have from company B he is able to make a new product for only $1M. And he can say: “It is totally reimplemented, there is not a single line of code that is similar”. But his savings is $9M.

    As long as some code is public is does not matter what kind of license it have, patents or whatever. By studying the code all really valuable knowledge can be extracted and the code can be “reimplemented”. Or stolen, which is a more fitting word.

    Of all the hypocrites in the world none is more worse in my opinion than the open source community.

    Regards
    Rolf Tollerud


  60. Rolf,

    What are you talking about?!?

    First of all, all most all of the open source projects out there are centrered around infrastructure - app servers, web servers, OS (Linux). That said, let's apply this to your economic "Company A and B" example.
  61. (Ignore previous post)

    Rolf,

    What are you talking about?!?

    First of all, all most all of the significant open source projects out there are centrered around infrastructure - app servers (Tomcat/JBoss), web servers (Apache), databases (MySQL), OS (Linux). I don't know of too many that develop business-centered products. That said, let's apply this to your economic "Company A and B" example.

    I think most people agree that WebLogic and WebSphere are quite similar in nature - both J2EE app servers that offer the usual suspects - caching, clustering, EJBs, servlets, JMS, etc. Now, let's say that each of them has 50 developers working on their respective app server. They both produce quality app servers they can make money with, right?

    Now, let's suppose they concentrated on an open source app server instead. What if each company dedicated 10 developers each, along with 10 developers from Sun and 20 other developers from various large companies, each of whom needs a high qualily app server.

    In this model, BEA and IBM both have 40 developers left that they can farm out to make money servicing clients who use this open source app server. They still MAKE MONEY with much less investment. Plus, a greater variety of contributors will almost always produce a better product.

    There more to making money in software than through selling products!!!

    What is wrong with this model?

    Ryan


    P.S. Please clarify this rather bold and confusing statement:

    Of all the hypocrites in the world none is more worse in my opinion than the open source community.

    What exactly is hypocritical about open source?
  62. Rolf,

    After seeing a plethora of your previous "J2EE sucks, .Net rocks" posts here at the SS, I know exactly where you are coming from. No need to clarify anything...

    Ryan
  63. Rolf: "Cost for a new company (B) to make a new similar product ***without*** the accumulated knowledge of company A = $10M. ... Cost for a new company (B) to make a new competitive product ***with*** the accumulated knowledge of company A = $1M. ... As long as some code is public is does not matter what kind of license it have, patents or whatever. By studying the code all really valuable knowledge can be extracted and the code can be “reimplemented”. Or stolen, which is a more fitting word. Of all the hypocrites in the world none is more worse in my opinion than the open source community."

    But I thought you defended .NET's re-implementation of Java. ;-)

    Peace,

    Cameron Purdy
    Tangosol, Inc.

  64. Ryan

    First of all I do not mean *all* open source developers and especially not any from Jakarta, they don't do "reimplementing".

    We are making proprietary products, not contract programming. My concern is that if OS succeeds totally maybe they are destroying the (proprietary products) market completely, not only for Microsoft.

    I think that you can not deny that many OS projects are "reimplementing" projects and this is the area that they excel most.

    "P.S. Please clarify this rather bold and confusing statement:

    Of all the hypocrites in the world none is worse in my opinion than the open source community."


    I am referring to that "reimplementing" or "stolen" sometime is the same thing.

    Cameron,

    "But I thought you defended .NET's re-implementation of Java. ;-)"

    Rotor + submission of core functionality to ECMA + "no tainted statement from MS" make it pretty clear that MS want OS to copy .NET. Not because they are kind of heart of course but because they want to hit Java. And the bait was taken!


    Regards
    Rolf Tollerud
  65. <quote>
    I think that you can not deny that many OS projects are "reimplementing" projects and this is the area that they excel most.
    </quote>

    I'm not sure about that. I think Linux and Apache were both born out of developers trying to improve on existing software, not really "reimplementing". However, I don't know enough about the history of these two, so I can't speak intelligently on them.

    As for the Java OS projects...

    Tomcat - definitely NOT reimplementing. It is the RI for servlets and JSP

    JBoss - not reimplementing, as they tend to be on the cutting edge of app server features, hot deploy and support for J2EE 1.3 (JBoss 3.0 has been out for almost three months and we are still waiting on WebSphere 5!)

    Sturts - a framework filling a need that I don't think was being met at the time when it was released

    Log4j - there was nothing like when it was introduced

    JUnit - the de facto testing tool

    Ant - a Java make

    Most, if not all of these OS developments are invaluable to a most any java developer. Yet they are not a knock off of somebody else's software. They are new creations or an improvement on existing software.

    Plus, nobody's bread-and-butter is coming from developing software that would compete with these tools, except for maybe JBoss and Tomcat. If any of these "stole" something, please set me straight.

    Ryan

  66. Ryan

    You are not reading my post. I was explicitly excepting Jakarta which I respect very much. They don't use the infamous GPL license, are completely pragmatic and even accept .NET projects.

    All your examples were from Jakarta!

    Regards
    Rolf Tollerud
  67. You say in one breath that you hate the idea of open source or companies stealing alread-implemented ideas, yet you support .NET which is an obvious rip of Java.

    Let me help clarify it for you: You are against open source and competition in general in which it could hurt your economic situation, and for it in all cases in which it can help you.

    Peace,

    Cameron Purdy
    Tangosol, Inc.

  68. Cameron,

    Sorry, excuse my latest post I was to fast.

    "But I thought you defended .NET's re-implementation of Java. ;-)"

    Microsoft re-implementation of Java is of course the same dirty "business as usual". The difference is that you don't expect anything else from a capitalistic company. The Open Source community on the contrary is trying to look like Saints. If you have started a successful OS project the rewards are often substantial.

    Regards
    Rolf Tollerud
  69. Nonsense.

    re 1) Even open source projects use source control
          and it can be traced who entered the illegal
          code. What is more: because the source is
          availble then it is likely to be discovered
          unlike for a commercial product delivered in
          binaries only.

    re 2) I think that in general open source products
          are more improved than commercial products.
          JBOSS is leading WebLogic and WebSphere not
          following. People just have tons of good
          ideas and with open source they can them
          implemented.

    re 5) There are many standards that originate from
          the commercial environment, but there are also
          a lot of standards that originate from the
          free environment. HTML and HTTP just to mention
          two very significant.
  70. <Matt Quote>
    Open Source is bad for business, the industry, innovation, and J2EE.
    </Matt Quote>

    Open source isn't going to put Sun out of business - Sun will put Sun out of business - Microsoft didn't put Netscape out of business at all - the worst thing that happened to Netscape was to be bought by AOL. Innovation stopped. Development stopped. And the world, including Microsoft, passed them by.

    Your understanding of Open Source isn't very strong. Let's go through your points:

    -- No Central Point of Responsibility
    That's not strictly true (there's always a central group that oversees the product) - and if there were issues with patent infringment or some other sort of liability, the issues would be resolved by existing laws - Open Source doesn't mean a bunch of law breakers.

    -- Decoupled from User Economics (No incentive to enhance functionality?????????)

    Open Source is no different from the Big-Company mentality that you espouse - they want their product in as many hands as possible. They have a strong incentive to enhance functionality, performance, stability, etc - often more so than the Big-Company. And since users can also be developers in open source, they will want to make the product as top-notch as possible.


    ---If community income is based on support/service/hardware then there is a bias against stronger products in favor of a hard-to-use product.(e.g. IBM)

    Sorry - but I've never seen that in the real world with Open Source. Name an open source product that is hard to use and that the group running it makes its money off of hardware,support, etc and it always remains hard to use. (Hint: JBoss isn't the answer)

    --De-values intellectual property. (like China bootleggers)

    What open source product has devalued intellectual property? How so? Open source might help drive prices down to a reasonable level (China bootleggers do not drive down the prices of software around the world) but that is a good thing. If BEA can't sell their server at a billion dollars/cpu and have to sell it at half that because of open source - this is a very good thing.

    -- Open source does not have the resources to drive standards and adoption. (e.g. Java/J2ee,Web services, SQL, etc have all been driven commercially)

    I might agree with this but that's hardly a reason to call open source the end of commercial business as we know it.



    I wouldn't run and hide from Open Source - it's not going to run companies out of business. Big clients are still going to go with the big vendors and spend ridiculous amounts of cash on them. If these companies go out of business it's their own fault not that of open source.

    Cheers
    Ray


  71. Hi Ray,

    Sounds like the only thing we disagree significantly on is whether I have a strong understanding of Open Source or not.

    <quote>
    Your understanding of Open Source isn't very strong....

    Open Source doesn't mean a bunch of law breakers...

    They want their product in as many hands as possible. They have a strong incentive to enhance functionality, performance, stability, etc - often more so than the Big-Company....

    If BEA can't sell their server at a billion dollars/cpu and have to sell it at half that because of open source - this is a very good thing...

    I might agree with this[diminished stds activity due to OSS] but that's hardly a reason to call open source the end of commercial business as we know it....

    If these companies go out of business it's their own fault not that of open source.
    <quote>


    The remaining points to consider are:
    1. Does the increased use of Open Source Software cause funds to be spent elsewhere than on software? (Thus minimizing the market for software.)

    2. Would this in effect reduce software revenues for comparable products? Wouldn't it also tend to reduce the selling price of comparable products? (you already pointed this out)

    3. Wouldn't this tend to reduce investment in such comparable products? (If the company is well run, right?)

    4. Would the quality and number of product choices go up or down in this scenario?

    5. So, is this " shrinking investment" scenario good for customers, the industry, you?

    6. Will this happen?
    Are the flaws and risks of the OSS model obvious enough to prevent it? (points I made previously)
    Will the OSS product inferiority in terms of functionality and support continue and prevent it? (e.g. Windows/Solaris vs. Linux, Weblogic vs. JBoss, Iplanet WS vs. Apache)

    Or the worst scenario, that the flaws are dismissed, inferiority continues and it still DOESN'T prevent it.

    regards,
    Matt








  72. Hi Matt -

    I'll respond to your points:

    <matt>
    The remaining points to consider are:
    1. Does the increased use of Open Source Software cause funds to be spent elsewhere than on software? (Thus minimizing the market for software.)
    </matt>

    In my experience, it drives the market to other software segments - in other words, instead of spending a ton of money on J2EE servers, you spend money on applications that sit on J2EE servers, that kind of thing. I work in data warehousing, etl, and eai - the money being spent on software that utilizes j2ee app servers (in this case, the name-brand types like BEA) is reasonably large and growing. Money does not decrease (at least for the reason you suggest) but gets distributed to other software segments.

    <matt>
    2. Would this in effect reduce software revenues for comparable products? Wouldn't it also tend to reduce the selling price of comparable products? (you already pointed this out)
    </matt>

    Yes - it would. This is, in fact, a good thing. This causes the industry to move on and look at other ideas, etc.

    <matt>
    3. Wouldn't this tend to reduce investment in such comparable products? (If the company is well run, right?)
    </matt>

    If the company is well run, they are continuing to think of ways to enhance their own products - maybe they can't make a lot of money off of their J2EE container any more, but maybe they can develop various add-ons that set them apart from others, for which they can add revenue. If the company is a one-trick-poney - then the situation is certainly different.


    <matt>
    4. Would the quality and number of product choices go up or down in this scenario?
    </matt>

    The quality would depend on the company, of course - and there's no reason, with investment in add-ons and completely different products, that both the quality and product choices go up.



    <matt>
    5. So, is this " shrinking investment" scenario good for customers, the industry, you?
    </matt>

    As noted above, it's not a shrinking investment scenario - so its good for everyone.

    <matt>
    6. Will this happen?
    </matt>

    No.

    <matt>
    Are the flaws and risks of the OSS model obvious enough to prevent it? (points I made previously)
    </matt>
    Sorry - I need to go read your points more closely.


    <matt>
    Will the OSS product inferiority in terms of functionality and support continue and prevent it? (e.g. Windows/Solaris vs. Linux, Weblogic vs. JBoss, Iplanet WS vs. Apache)
    </matt>

    How are the products you mention inferior in terms of functionality and support? I don't buy your argument. There maybe things where they don't compare line-by-line, but that's not a mark of inferiority in any way.

    <matt>
    Or the worst scenario, that the flaws are dismissed, inferiority continues and it still DOESN'T prevent it.
    </matt>

    I don't buy your argument on so-called inferiority, so this doesn't follow.


    Cheers
    Ray

  73. Ray,

    Well, I am certainly not going into this business with service/support/contract programming. For the first installation and configuration seems to take a large part of it and I want to write code, system administration is for me aburrito and uninteresting. Second when you do contract programming IMO you never are allowed the time you need to deliver quality. I want to work on a product, refining, trimming, seeing myself more as an artist. I have more that 10 years successful production behind me, and if I can not go on in the lifestyle I am used to I rather gets out of the computer business altogether.

    It seems strange to me that this GPL business have originated in USA, land of the “Americam dream”. OSS have all the marks of the usual left utopia movements including leaders who get rich and workers that are not paid or paid very little and someone that are made to look like the great Satan guilty or not.

    Regards
    Rolf Tollerud
  74. "It seems strange to me that this GPL business have originated in USA, land of the &#8220;Americam dream&#8221;. OSS have all the marks of the usual left utopia movements..."

    The US is hardly the refuge of left utopia movments and the developoment and promotion of OSS is hardly limited to the US. Among the founders of projects mentioned in this thread, Linus Torvalds is Finnish, Marc Fleury is French and Compierre's founder is German. The American dream part you are referring to (getting rich quick) had to do with the last days of the dot-com boom. Linux briefly became the flavor of the moment for the same people talking about "paradigm shifts," the "New Economy" and the VA Linux IPO was making headlines.

    In the end when the hype dies down, it's all about practicality, economics and market shifts. Bill Gates and Richard Stallman (founder of the Free Software Foundation) began their debate on software licensing at the moment when hardware costs came down enough and the computer market broadened enough so that it was economically feasible to make money purely off software licensing. Because software licensing subsequently made a ton of money, it earned its holy water of legitimacy in the 1980s, but Open Source could never take off until a little over a decade later when the Internet made it possible for widespread software collaboration and distribution.

    Forget nationalism and ideology. If Open Source succeeds, it's because it successfully competes in the marketplace. It provides the right product at the right price. Also, when an Open Source project becomes widespread and popular enough, no proprietary competitor can compete in terms of stability. The number of users and speed with which they can contribute bug fixes far exceeds what any proprietary software company can achieve in their QA process.

    I believe proprietary companies will continue to provide good development jobs in higher-tier software. Open Source can commoditize infrastructure and even ERP-lite (ie Compierre) but I just don't see it being successful with software that sells to smaller, more specialized markets. Then there's the whole market for developing customized applications, which won't go Open Source.
  75. Pierre,

    "The number of users and speed with which they can contribute bug fixes far exceeds what any proprietary software company can achieve in their QA process"

    There are much more to software development than bug fixes. Trial and error, rebuilding again and again learning from mistakes. It takes stamina and determination and a lot of money. But of course when it is finished, it is fairly easy to re-implement.

    People never learn, when ruthless old capitalistic practice is set up against usual left utopia movements the results is always the same, applauding of the left movement. It doesn’t count how many times in history such movements have failed, often with disastrous result. But it is the capitalistic systems that have generated the welfare of the western society.

    It also never fails to amaze me to see how those blindly followers think for themselves. For example since we are in this thread why don’t we examine the jboss.org. There are many things that are clearly suspicious with this organization, for instance it seems to have become a family company. I challenge you to get out information of how much revenue is coming from "jBoss certification". What was really happening with Richard Öberg who was as I understands it the person who wrote the most important things in the kernel? Why is jBoss not part of the only incorruptible body in this business, the Apache/Jakarta group?

    Look the true face of Open Source:
    http://smoothwall.org/community/home/articles/dickmorrell/20020322.time.html

    Regards
    Rolf Tollerud
  76. Excuse me - but you have got it all wrong.

    Capitalism is about consumer free choice and
    survival of the fittest among the producers.

    That JBoss are surviving and iPlanet are not
    is capitalism. JBoss is cheaper and better, so
    the consumers prefer it.

    As a side note: WebSphere and WebLogic are
    also surviving, because apperently the consumers
    like them too (they think they get value for
    the money they pay).

    But trying to ban free software is the government
    artifcially keeping inefficient producers alive.
    That is not capitalism. I do not necesarrilythink
    it is particular "left oriented". It is more
    "plain stupid".

  77. Excuse me - where do I propose that the government should ban free software? I only advocate a modest proposal, that people should get a grip on themselves and not succumbing to mass hysteria. Its similar to religions – the world would be a better place without it. Nothing would be better if what the socialist and leftist parties says it true, but it isn’t. Similar for OSS.

    Regards
    Rolf Tollerud
  78. Rolf,

    This post of yours exposes you as the one that needs to get a grip.

  79. Sorry, but the whole OSS movement and articles like "The Cathedral and the Bazaar" is inherently disgusting to me. Diderot: "I can stand all scoundrels, ruffians and thieves in the world, its all part of life. The intellectual hypocrite I have more problem with"

    Regards
    Rolf Tollerud
  80. Rolf,
    I applaud your intellectual honesty.

    For groups to be able to radically disrupt the model of software development to one that is apparently inferior and possibly seriously flawed and where innovation is based solely on the psychological needs of developers to show-off is tragic. (Esp. when they are not even able to articulate a defense)

    To clarify, I am not concerned about the welfare of any particular company or product. It just seems irresponsible to join the siren song of "no license fee, free, free, free" and not acknowledge the short-term and long-term risks. I only advocate regulation if necessary. (Many other industries need protection from fanatics, why not software?)

    Matt



  81. For groups to be able to radically disrupt the

    >model of software development to one that
    >is apparently inferior and possibly seriously
    >flawed and where innovation is based solely on the >psychological needs of developers to show-off
    >is tragic. (Esp. when they are not even able
    >to articulate a defense)

    If open source does not meet your need then go pay $10,000 or whatever conpanies charge for your software. Nothing is stopping them form innovating and charging for it and nothing is stopping you from paying for it either.

    Open source software has shown to be quite superior. JBoss, Tomcat, Postgresql are some examples of quality open source software, yet Oracle and IBM are still able to sell their software. If a company is having difficulty competing then they should write better software.

    Open source software is here to stay -- live with it.
  82. Rolf,

    What exactly is hypocritical about proponents of OSS? You mentioned earlier about something about "reimplementing" and "stealing". Is that what you are referring to? If so, what are some examples of this, please.

    Specifically, do you see JBoss developers as hypocrites? If so, why?

    I really don't see where all of this anti-OSS venom is coming from. The two "anti-OSS" args I have heard in this thread is that it will destroy the software economy (total bullshit) and that OSS steals from others. You are clearly in the "anti-OSS" camp - why is this so? Please be specific.

    Ryan
  83. <Rolf>
    There are much more to software development than bug fixes. Trial and error, rebuilding again and again learning from mistakes. It takes stamina and determination and a lot of money. But of course when it is finished, it is fairly easy to re-implement.
    </Rolf>
    I agree somewhat. Software does take, trial, error, stamina and determination. I disagree it takes money. JBoss' success has been built mainly on free contributions and effort.

    <Rolf>
    People never learn, when ruthless old capitalistic practice is set up against usual left utopia movements the results is always the same, applauding of the left movement. It doesn’t count how many times in history such movements have failed, often with disastrous result. But it is the capitalistic systems that have generated the welfare of the western society.
    </Rolf>
    Most OSS developers do NOT consider themselves a member of a left utopia movement including myself. As Marc Fleury has already stated in his response to McNealy's comments, it is just a different business model and approach.

    Although I have never read John Nash's general dynanics theory of economics, I will quote from the movie. "The best outcome is when the individual works for both the best interest of himself as well of that of the group." (or something like that, but you probably understand what I'm trying to say.). Capitalist/communist, they both fail. (Great depression for capitalism, fall of Soviet Union in early 90s). You need a balance from both ideals. Marc Fleury's vision for JBoss/JBoss Group is an attempt at that balance IMHO. Will it succeed? I have personally bet my well being on it doing so.
     
    <Rolf>
    It also never fails to amaze me to see how those blindly followers think for themselves. For example since we are in this thread why don’t we examine the jboss.org. There are many things that are clearly suspicious with this organization, for instance it seems to have become a family company. I challenge you to get out information of how much revenue is coming from "jBoss certification". What was really happening with Richard Öberg who was as I understands it the person who wrote the most important things in the kernel? Why is jBoss not part of the only incorruptible body in this business, the Apache/Jakarta group?
    </Rolf>
    Since I don't see anybody else from JBossGroup responding to this, I will do my best (Marc, I hope you don't mind...)

    Family company? Besides, Marc, his wife, and his father's marketing mind. The other 15 of us at JBossGroup (including our CFO and director of sales and biz dev) are not related to Marc. (as far as we know ;). Besides who fucking cares? The New England Patriots are a family owned and operated company and they're a top notch organization. (Sorry, I have football on my mind).

    Revenue from "JBoss certification"? All I can say is business is booming for JBossGroup.

    Rickard Oberg? Rickard has not been a JBoss contributor since I've been a core developer (Feb, 2001). His ideas were important to the current JBoss kernel, but it is really the current core development team that has brought these ideas into reality. Honestly, although Rickard's ideas were good, they really were not revolutionary. Iona's Orbix2000's had a pluggable component design and Aspect Oriented approach long before JBoss ever existed. (I should know, I worked on the project.)

    Apache/Jakarta incorruptible? Follow the money trail dude. You'll find IBM. All Apache/Jakarta is, is a brand. That's all they got. JBoss is doing fine and doesn't need Apache/Jakarta. Again, since I claim that Apache/Jakarta is just a brand, what does JBoss and JBossGroup stand to gain by joining Apache? Nothing more than destroying the JBoss brand and the revenue model of us trying to make a living doing something we love? Plus Apache does not provide an organization and business channel for those of us who would like to make a living from the OSS contributions we love to do. JBoss has a strong community, a strong development process, talented dedicated engineers, and a growing brand name. Apache gains us and the industry nothing.


    <Rolf>
    Look the true face of Open Source:
    http://smoothwall.org/community/home/articles/dickmorrell/20020322.time.html
    </Rolf>
    I read the article. Must say that JBoss itself dwarfs the SmoothWall project in complexity and users. JBoss currently boasts 180,000 downloads per month (1.8 million over 20 months). Sure, forking in OSS is always a possibility (and some say a danger), but if the project has done the right thing and built its brand this is not a problem. Think about it....What if ANT was forked from Apache/Jakarta? Nothing, ANT and Apache have a strong brand and the forked project would gain little visibility.

    If you want to make a business with OSS, you have to be smart:

    1. Trademark the project name (JBoss is trademarked).
    2. Protect your trademark. (JBoss has a distinguished lawyer with an OSS speciality (Larry Rosen), and has protected its trademark and copyrights on numerous occasions).
    3. Take care of your core developers. (Marc Fleury has been instrumental in this from my own personal experience. If he can't find the money for you, he tends to find other ways to compensate(book and magazine article opportunities)
    4. Create top-notch services around your project. (JBossGroup provides top-notch training (on-site an open), consulting, support (including 24x7!), and documentation. What? Pay for docs? $10 is nothing. Plus the $$ provides incentive to the developers to maintain the docs.)
    5. Create an accountable organization. Providing services such as support.

    Bill


  84. Very good luck with your company then.

    Regards
    Rolf Tollerud
  85. Matt -
    You have an unfounded fear of open source software. You call that which you don't understand 'inferior'. You call the proponents terrorists. You give simplistic and unrealistic arguments as to how some sort of (what you call) worse-case scenario will come about because of the evil OSS. Instead of making these sorts of silly statements, show how Open Source has affected your company's bottom line and how your company is going to go out of business because of OSS. Please do not respond by telling me I do not understand economics, or like Rolf, assume that myself or others are part of some sort of left-wing movement bent on destroying capitalism. That's just ridiculous, of course. I quite enjoy making money as I am sure do many people on this thread. How about some real examples as to how OSS has traumatized your life or that of your company. Do you advocate legislative protections from Open Source? What is it that you want Open Source to do for you so that your company can better compete? Why is it that your company can't compete when there is an open source product in the market place?

    Cheers
    Ray
  86. Ray,
    I hope my fears are unfounded too.

    <Ray>
    Instead of making these sorts of silly statements, show how Open Source has affected your company's bottom line and how your company is going to go out of business because of OSS.
    </Ray>

    This whole thread has been about SUN's dire business situation and McNealy's comment about Open Source. Doesn't that qualify?

    Matt
  87. Matt -
    <Matt>
    This whole thread has been about SUN's dire business situation and McNealy's comment about Open Source. Doesn't that qualify?
    </Matt>

    No, I'm afraid it doesn't - Sun's dire business situation has to do with market forces, a participant of which is OSS, but no one can turn around and say that OSS is driving Sun out of business or that it is even responsible for Sun's poor performance. The fact that business spending is on the rocks has to do with why Sun isn't making money.

    Most will agree that the fact that iPlanet isn't (or at least wasn't the last time I dealt with it) a good product is why people aren't flocking to it in droves. McNealy is blaming the poor response to iPlanet on something like jBoss, which isn't terribly bright on his part. If you have a poor product, you'll get a poor response.

    Cheers
    Ray
  88. Hi Ray,

    <Ray>
    In my experience, it drives the market to other software segments - in other words, instead of spending a ton of money on J2EE servers, you spend money on applications that sit on J2EE servers, that kind of thing
    </Ray>

    The insidious aspect is that why wouldn't some group start doing open source Siebel, Peoplesoft, or SAP -type apps. My point is that we are at the edge of a slippery slope. Where OSS-"Terrorists" will systematically destroy all major commercial software markets and replace these commercial products with inferior substitutes. (which leads to...)

    <Ray>
    How are the products you mention inferior in terms of functionality and support? I don't buy your argument. There maybe things where they don't compare line-by-line, but that's not a mark of inferiority in any way.
    </Ray>

    To start with Quality is very poorly understood by most in software. It is not, as someone said above, the ratio between what you get and what you pay. It also is more than just a line-by-line comparison of features or the number of bugs in the code base.

    Software is unique in that defects (bugs, limitations, missing features, complexity, inefficiency, low performance, etc.) are not readily visible at the beginning of a project, but have a greater negative impact on the success of the project (i.e budget, schedule, features) the longer they go undetected.

    So the following examples are why you can consider current OSS implementations (Jboss in particular) inferior to their Commercial counterparts.
    1. Performance : “JBoss also shows increasing processor asymmetry as load increases, so multi-processor machines can only provide a certain level of load balancing. This explains why JBoss performs almost identically on the current h/w environment (peak of 600TPS, dropping to 350TPS, quad cpu server) compared with the previous dual cpu servers (peak of 512TPS, dropping to 357TPS).” http://www.cmis.csiro.au/adsat/jboss.htm

    2: Limitations: Tool and 3rd party product support is Weak. Only Togethersoft and Altoweb support JBOSS. Does BMC have a plugin for JBOSS? Does Siteminder have an Agent? How do you interoperate with a Microsoft or .NET Application with JBOSS?

    3. Complexity: “All and all, we have found that even though JBoss/Jetty does not have many clustering features in its current release, our hardware and software architectures can make up for these deficiencies.” http://www.onjava.com/pub/a/onjava/2001/09/18/jboss.html

    4. More Defects: “…open source participants rely heavily on source access, taking it on faith that the “many eyes” of open source developers equal a more secure operating system. Recently, a TechRepublic article comparing security between Windows and Linux reported that up to its publication date in the autumn of 2001, Windows had 24 reported security vulnerabilities. In comparison, Red Hat Linux had 28 vulnerabilities. When you consider the difference in the size of the installed base of Windows 2000[along with the code base/functionality size differences] to Red Hat Linux, the percentages indicate a [much] higher degree of security vulnerabilities for Linux.” http://www.microsoft.com/windows/embedded/docs/SAKComparison.doc

    and I could go on and on. So to use a very understated quote by Terry Hicks of Gartner: “Free Software is not Risk Free”.

    I'll ask you again:
    What will prevent the worst-case scenario mentioned in the previous message?

    regards,
    Matt
  89. Matt,

    I have yet to see the inferiority of linux/unix (on the server) to Windows. In fact, no really experienced developers/admins that I know would choose to run, say Oracle, on Windows over Linux. And certainly over Solaris.

    You're "insidious" comments are almost ridiculous. How about SAPDB? Apache? Struts? Linux? (Ironic that you mentioned SAP since they support SAPDB open source project wholely). These things are not inferior products, and they seem to be doing pretty well. Almost every aspect of networking has some relation to some open-source product or other.

    Finally, your lack of economic knowledge is showing again. You need to learn about investment cycles, structural and cyclical economies, economic specialization. The investments not made into closed-source will certainly, over time, move into new areas for profitability. IBM again has proven that a services model makes a whole lot of money. Even if the products were inferior (which definitely isn't the case with MS, but maybe other companies..), does the ROI justify it since more can be spent on services? The ROI is going to be higher without paying for heavy licensing fees.

    You're looking at the values from only one perspective. But the clients perspective is important. If they see heightened ROI from this structure, they're going to push for it. The vendors will see opportunities to sell services, and "accessories" that create value for them for both.

    I get paid plenty to work almost entirely with open source products. There's value for me, _and_ value for my clients (I hope!).

    You keep implying that there is a value destruction in open source, but that just isn't true. It's only true from the licensing fee standpoint. The value is made up for in other places by the client and the seller. Economies shift slowly to new places over time. Like I said in an earlier post, the US economy has been becoming more service oriented for some time, and it seems to be working.

    In fact it may be a good hedge against a down economy. People always need services and often have to pay for them anyway, but they do curb manufactured good investments. Hmmm. I'll have to look into that.


    -Newt
  90. Matt,

    Why don't you just write your Congressman: "Open Source is a threat to the American way of life," maybe you can connect it to September 11 or distressed American Farms while you're at it. Your argument about Open Source being a threat to US National Security holds as much water as your prior technological and economic observations. If you truly care about this issue, go to Mitre corp's site and read the pages and pages of their research report on Open Source and US military systems.

    You may not like Open Source, you may think it's crap, but if you didn't think Open Source had a chance of being successful, it's clear you wouldn't waste your (and presumably your company's) time by writing on this thread.

    When you cut to the chase, what you and Scott McNealy are really crying foul about is that Open Source offers a successfully competing product that endangers profitable existing revenue streams. As Jason McKerr stated, that's just part of market economics. Ultimately the only thing you can do about it is find another market niche where you can compete.

    The honest letter to the congressman/woman would read as follows. "Dear Sir or Madam, please remove developers' right to develop software that competes in my company's market, remove their right to give away this software for free and make the source code freely available. While you're at it, please remove consumers right to use this software."

  91. <Matt>
    2: Limitations: Tool and 3rd party product support is Weak. Only Togethersoft and Altoweb support JBOSS. Does BMC have a plugin for JBOSS? Does Siteminder have an Agent? How do you interoperate with a Microsoft or .NET Application with JBOSS?
    </Matt>
    You forgot to mention OptimizeIt from Borland, JProbe from Sitraka, Eclipse, Gemstone, SonicMQ. There are probably many others. OptimizeIt, JProbe, Sonic have supported JBoss integration for more than a year. And this is without a partnership. In fact our conversations with Sonic have revealed that most of their customers are using JBoss in development and/or production.

    <Matt>
    3. Complexity: “All and all, we have found that even though JBoss/Jetty does not have many clustering features in its current release, our hardware and software architectures can make up for these deficiencies.” http://www.onjava.com/pub/a/onjava/2001/09/18/jboss.html
    </Matt>

    Hey, thanks for quoting me! Look at the article date. Almost a year ago. Since then Sacha, others and I have been motivated to implement the clustering features of JBoss 3.0. Please check out my most recent article on JBoss 3.0 clustering.

    http://www.onjava.com/pub/a/onjava/2002/07/10/jboss.html


    <Matt>
    4. More Defects: “…open source participants rely heavily on source access, taking it on faith that the “many eyes” of
    </Matt>
    Software quality really depends on the organization/management of the project whether closed or open. JBoss for instance does realy on "many eyes", but we also have a growing testsuite of over 2000 tests that is run nightly and reported to the world every day(actually it gets triggered by user commits as well, so we know instantly if somebody has inserted "interesting" code). Since even our testsuite is OSS, you KNOW how well we test our stuff and can make this a criteria in your evaluation. With closed source, you as a software selector, don't have this advantage.

    Bill
  92. Open source isn't going to put Sun out of business -

    >Sun will put Sun out of business - Microsoft didn't
    >put Netscape out of business at all - the worst
    >thing that happened to Netscape was to be bought
    >by AOL. Innovation stopped. Development stopped.
    >And the world, including Microsoft, passed them by.

    I agree that Open source isn't going to put Sun out of business, however Microsoft does resort to underhanded techniques to leverage their monopoly, and Microsoft's monopolistic leveraging did put Netscape out of business. Court testimony proves it and to deny it is dillusional.

    Microsoft did resort to monopolistic leveraging by distributing a version of the JVM that is obsolete and incompatible with Java and they need to be held accountable for their shady business practices.

    As far as innovation, I believe Sun and the open source community has produced the most innovation in the past decade. Microsoft Windows is not an innovation -- they merely copied Apple which copied the work at Xerox PARC.


  93. Hi Peter -

    <Peter>
    I agree that Open source isn't going to put Sun out of business, however Microsoft does resort to underhanded techniques to leverage their monopoly, and Microsoft's monopolistic leveraging did put Netscape out of business. Court testimony proves it and to deny it is dillusional.
    </Peter>

    I agree with you about Microsoft, but I should have qualified my own statement. _I_ stopped using Netscape because the product became stagnant and unstable (I do use Mozilla now). For a long time, whenever I bought a computer with Windows, I disabled IE and installed Netscape. But it became an unstable product so I stopped using it. I would have happily continued using Netscape had it remained a viable product. It was the first browser I ever used and I had a high level of loyalty for it. I agree that Microsoft went a long way to destroying the user base of Netscape with their particular illegal business practices. However to say that Microsoft is _entirely_ responsible is somewhat delusional as well - you still need a good product which Netscape wasn't. Even after the cash infusion from AOL (in fact it got worse in my opinion, since AOL was only interested in the Netscape user-base, not the product).

    Cheers
    Ray
  94. Some reality checks[ Go to top ]

    - Apache didn't "win" any web server war. They have a greater marketshare than IIS. IIS' marketshare isn't something to sneeze at.

    - Why does Microsoft "need to be stopped"? And how is it that they can't compete with open source? Do you realize what customers value -- and that open source code is but ONE of many considerations in any purchase?

    Let's frame this: Linux is perhaps the most widely used piece of open source software in existence. Sales for Linux were $80 million during 2001 -- which is what Microsoft makes in OS revenue in 2 DAYS. (Source: http://news.com.com/2100-1001-948678.html)

    Many of you might suggest that the majority don't "buy" Linux or Apache, or whatnot -- but please realize that in the software development area the bounding scarcities are in time and people. And there are three forms of economic resources: Labour, Capital, and Knowledge. OSS solutions are short on labour and capital, but high on knowledge. They're at a clear econonmic disadvantage. And don't give me any "new economy, OSS destroys the concept of capital" neo-Marxist rehetoric, lest you want to look silly.

  95. Some reality checks[ Go to top ]

    Stu Charlton wrote:

    "Let's frame this: Linux is perhaps the most widely used piece of open source software in existence. Sales for Linux were $80 million during 2001 -- which is what Microsoft makes in OS revenue in 2 DAYS. (Source: http://news.com.com/2100-1001-948678.html)"

     What is the revenue of MSIE, IIS, ASP, Windows Media Player, .NET framework if they are free?

      Microsoft offers several key end user technologies for free and **sells related technologies** : starting with the OS that runs them.

      Are you going to see other .Net IDE better than Visual Studio .NET?
      Are you going to see other non MS S.O. running .NET?
      Are you going to see any .Net application better than MS counterpart?
      Are you going to see any kind of competition?

      Linux and Java (free both) are promoting many bussiness around them that it wouldn´t be possible without them. Microsoft competitors (IBM, BEA, Sun, Oracle ...) are porting their applications to Linux and Java because they are really opened and is really possible to compete and gain market share.
     
      Sun must support to JBoss to compete with .Net, and develop software to middle-big enterprises.

      


  96. I don't get you people who are against Open Source. In one breath you claim that Open Source is but a drop in the economic bucket, that it does not make business sense, but then in the next breath you want to outlaw it and restrict it and call it Un American. If its so small what are you afraid of? The fact of the matter is you are afraid of it because it is larger than a drop and thats because it works as a viable way of doing business.
  97. Frankly, considering the success and quality of Sun's own appserver they should buy jboss group and pool their resources toward making jboss their app server. It would fit their strategy of marrying linux and solaris to marry jboss and iplanet.