Discussions

News: BEA and Open Source

  1. BEA and Open Source (29 messages)

    A lot of heat (though not much light) has been generated on TSS of BEA CEO Alfred Chuang's comments in a recent article in The Register. Alfred ostensibly makes two points when referring to the current state of open source. First, that "open" is a sliding scale, and that those who conflate the notion of open source and total freedom are missing an important nuance. His second point is that there are people in this world who (Gasp!) could not care less about open source. The first point has to do with control. I would posit that control in open source is properly maintained by effort you place into a project. BEA has a modicum of control over the WTP project, not because of its positional authority, but rather the people who earn the right to commit to the project. Similarly, (and it pains me to say nice things about them) IBM's control over Eclipse has nothing to do with their founding status, or board seats, but rather their thundering hoards of developers who contribute to the project. It would be disingenuous of JBoss to suggest that their community is fully-based on a meritocracy. Things that happen in the JBoss community happen to meet JBoss' business needs, and that is ok. But equate this community with "total freedom" implied by the Free Software Foundation verges on the mendacious. It is akin to saying Myanmar/Burma has a free press, so long as you do not write about the government and run against their interests. JBoss is a free community, so long as the work you are doing does not run against their interests. That notwithstanding, RedHat's acquisition is bound to have an effect on JBoss' behavior, as they have a good track record of working within the open source community. The second point has to do with common sense. For customers with specific customer needs, like companies that just want an efficient call center, or an inventory management system, they could care less about the distribution rights of the software. They do care about lowering call times and reducing cost per Customer Service Representative. What matters to these folks is not where the code comes from, but how good it is. This is the real metric of value. Open source offers no intrinsic value to someone who wants to sell more insurance, or make sure customers are notified when their plane is late. Quality software, regardless of origin, is what matters. All in all, I agree with Alfred's statements, and not just because he signs my paycheck. I agree with him because his comments are just common sense.

    Threaded Messages (29)

  2. Re: BEA and Open Source[ Go to top ]

    they could care less about the distribution rights of the software
    They could care less... so they care a little? http://www.incompetech.com/gallimaufry/care_less.html
  3. Re: BEA and Open Source[ Go to top ]

    His second point is that there are people in this world who (Gasp!) could not care less about open source.
    That seems pretty specious. I mean if they don't care either way, they are essentially opting out of the argument. Not caring about open source is not a argument against it nor is it an argument for it. It's just irrelevant.
  4. Re: BEA and Open Source[ Go to top ]

    Bill will find the following link helpful: http://www.fsf.org/licensing/essays/free-sw.html This is the FSF definition of "Free Software", and has nothing to do with stuff Alfred and Bill make up over dinner.
    It would be disingenuous of JBoss to suggest that their community is fully-based on a meritocracy.
    But equate this community with "total freedom" implied by the Free Software Foundation verges on the mendacious.
    Nowhere does the FSF definition of Free Software mention "total freedom" - whatever that means - or "meritocracy". In fact, as far as I am aware, the only open source community which talks about "meritocracy" is the Apache Foundation, which is _not_ "Free Software" according to the FSF definition. Secondly, my understanding is that the way "total freedom" is (extremely rarely) used in this context is to differentiate GPL-ish licenses from BSD-ish licenses. It has absolutely nothing to do with how a project is led or managed. To quote from the FSF:
    Free software is a matter of the users' freedom to run, copy, distribute, study, change and improve the software. More precisely, it refers to four kinds of freedom, for the users of the software: * The freedom to run the program, for any purpose (freedom 0). * The freedom to study how the program works, and adapt it to your needs (freedom 1). Access to the source code is a precondition for this. * The freedom to redistribute copies so you can help your neighbor (freedom 2). * The freedom to improve the program, and release your improvements to the public, so that the whole community benefits (freedom 3). Access to the source code is a precondition for this.
    So of course JBoss is Free Software. And of course BEA is not. There is plenty else here that is misleading. For example, BEA's influence in Eclipse is due to the fact that BEA pays Eclipse a whole bundle of money every year, and so has little to do with "meritocracy":
    BEA is joining the Eclipse Foundation as a Board Member and Strategic Developer. As such, BEA can play a lead role in helping to guide the Foundation’s technology innovations. While Eclipse currently counts 91 companies in its membership, only eight of these companies are Strategic Developers. In addition to joining as a Board Member and Strategic Developer, BEA has also been elected to lead the Web Tools Platform (WTP) project with the appointment of a BEA senior architect to the WTP Project Management Committee.
    http://www.bea.com/framework.jsp?CNT=pr01422.htm&FP=/content/news_events/press_releases/2005 FYI, it is _extremely_ expensive to become a Board Member of the Eclipse foundation.
    . I would posit that control in open source is properly maintained by effort you place into a project.
    No, this is not true of "open source". This may be true of the Apache model. Please stop equating open source with Apache. Many open source projects are led by companies such as MySQL, JBoss, Sun, IBM, etc. (Of course, it is certainly true that BEA does not lead any significant open source community.)
    It would be disingenuous of JBoss to suggest that their community is fully-based on a meritocracy. Things that happen in the JBoss community happen to meet JBoss' business needs, and that is ok
    Since BEA are continuing to repeat these outlandish claims in public, I would like to challenge Bill Roth or Alfred Chuang to an open public debate on this issue, somewhere where journalists are present. They get to name the time and place. I can be contacted at gavin dot king at jboss dot com. As a guy who has now spent most of my working life developing Free Software, I find it outrageous when people like Alfred and Bill - neither of whom, AFAICT, has contributed a single line of open source code in their lives - want to lecture real, living open source communities about how they should govern themselves.
  5. Re: BEA and Open Source[ Go to top ]

    For example, BEA's influence in Eclipse is due to the fact that BEA pays Eclipse a whole bundle of money every year, and so has little to do with "meritocracy"
    Gavin, I really don't want to get in the middle of what looks like a fun debate with Bill and others at BEA. But I did want to correct or at least clarify this statement by you. First, we at Eclipse do value "meritocracy" in ways very similar to Apache, so they are not alone in that regard. Second, BEA's influence at Eclipse is definitely not just a result of the funds they provide to the Eclipse Foundation. They have also sent a lot of really good developers to contribute to Eclipse projects. Most of those people have gone through the process of achieving committer status at Eclipse. The influence that BEA --- or any organization for that matter --- has on Eclipse projects is proportional to the good people they contribute, not their membership dues. So, like I said, I don't want Eclipse to be collateral damage in your raging debate, but hopefully this clarifies things a little bit about how Eclipse works.
  6. Re: BEA and Open Source[ Go to top ]

    For example, BEA's influence in Eclipse is due to the fact that BEA pays Eclipse a whole bundle of money every year, and so has little to do with "meritocracy"

    Gavin,

    I really don't want to get in the middle of what looks like a fun debate with Bill and others at BEA. But I did want to correct or at least clarify this statement by you.
    OK, thanks for the clarification, I do not mean to take shots at Eclipse. The point is merely that if BEA are trying to draw some distinction between Eclipse and JBoss, then it is relevant that Eclipse is controlled by large corporate interests who cough up large amounts of bucks each year. (And there is _nothing_ wring with that! I am the first to defend the positive work that some corporations do for open source.) However, it is difficult to see how this model is so very different to the JBoss or MySQL model in the context of the current debate. It is different, certainly, but I don't think the differences help Bill's argument.
    First, we at Eclipse do value "meritocracy" in ways very similar to Apache, so they are not alone in that regard.
    We at JBoss also deeply value meritocracy, as do most people in most companies and organizations. Are we as meritocratic as Apache? Hard to say. Certainly Bill makes no convincing argument that a Foundation with a large community is _by_nature_ more meritocratic than a company with a large community. (He seems to assert this, but apparently he regards it is obvious, which it is not.) But my point is not that meritocracy is not, by and large, a Good Thing. Rather my point is that "meritocracy" is not part of the definition of "Free Software" or "Open Source", which is what Bill Roth tries to claim. (It _is_ part of the definition of "the Apache model", however.)
  7. Re: BEA and Open Source[ Go to top ]

    But my point is not that meritocracy is not, by and large, a Good Thing. Rather my point is that "meritocracy" is not part of the definition of "Free Software" or "Open Source", which is what Bill Roth tries to claim. (It _is_ part of the definition of "the Apache model", however.)
    Consider the source. Both Chuang and Bill (now, in damage control mode) are seeking to redefine the definition of Open Source, throwing a bunch of stuff in the air, and hoping some of it sticks. I would not bother arguing any of their points, as they will likely keep changing with every rebuttal. "No, what Alfred really meant to say is..." The FACTS are that JBoss enjoys a vibrant community of OS developers working on a range of projects. This has not changed since the original AS project was founded. There is no difference between how JBoss projects are run and how other OS projects are run. The *only* difference, is that our projects compete with BEA. The java community is smart enough to place FUD where it belongs (/dev/null), and IMHO, BEA is doing itself more harm than good by taking this stance. (I wont charge you guys for my strategy consultation, in this case). ;-) STAY METAL! Roy Russo
  8. Re: BEA and Open Source[ Go to top ]

    ...The point is merely that if BEA are trying to draw some distinction between Eclipse and JBoss, then it is relevant that Eclipse is controlled by large corporate interests who cough up large amounts of bucks each year.
    Last I checked RHAT has a market cap of about 5 billion which compares to BEA's. I think JBoss is being controlled by large corporate interests too.
  9. Re: BEA and Open Source[ Go to top ]

    Gavin, this is an interesting debate and lets all stick to real facts:
    FYI, it is _extremely_ expensive to become a Board Member of the Eclipse foundation.
    Everyone can read Eclipse membership statements, what that means (in different terms such as financial investment but also assigning employees and skills to Eclipse projects) on the Eclipse.org pages (read the "Membership Agreement". I am not posting the figure and details here as it is a legal document and only this document has authority so go read it). http://www.eclipse.org/org/documents/ I assume this is a strategic choice every company can do and the financial terms sound affordable to a company like RedHat/JBoss but your choice was different (at least so far). (granted, this is a bit off topic but I did not wanted your vague statement misinforms the TSS community). Alex
  10. Re: BEA and Open Source[ Go to top ]

    Gavin, this is an interesting debate and lets all stick to real facts:

    FYI, it is _extremely_ expensive to become a Board Member of the Eclipse foundation.


    Everyone can read Eclipse membership statements, what that means (in different terms such as financial investment but also assigning employees and skills to Eclipse projects) on the Eclipse.org pages (read the "Membership Agreement". I am not posting the figure and details here as it is a legal document and only this document has authority so go read it).
    http://www.eclipse.org/org/documents/

    I assume this is a strategic choice every company can do and the financial terms sound affordable to a company like RedHat/JBoss but your choice was different (at least so far).

    (granted, this is a bit off topic but I did not wanted your vague statement misinforms the TSS community).
    Alex
    I've no idea what your point is. Of course RedHat could afford to buy a seat on the board of the Eclipse Foundation - RedHat is a 5 billion dollar corporation! But individuals and most small companies certainly cannot afford it. Again, there is nothing wrong with the Eclipse model. The Eclipse community is led by a foundation controlled by a group of large corporations. The JBoss community is led by a single corporation with a deep commitment to open source. Both are perfectly valid models. Both fit the definition of "Open Source". Both have advantages and disadvantages. For example, I could argue that a disadvantage of the "Foundation" model is that it is not well adapted to creating the kind of disruptive change in the marketplace that JBoss has created. It is highly unlikely that any not-for-profit foundation would have been able to break the IBM/BEA appserver duopoly as JBoss has done, benefiting consumers with greater choice and downward pressure on pricing.
  11. Re: BEA and Open Source[ Go to top ]

    Of course RedHat could afford to buy a seat on the board of the Eclipse Foundation - RedHat is a 5 billion dollar corporation! But individuals and most small companies certainly cannot afford it.......The Eclipse community is led by a foundation controlled by a group of large corporations.
    Gavin, I'm sorry to be a pest, but I need to clarify this statement as well. The annual membership dues for strategic developers are based on a percentage of the revenue of the company. So we most certainly have small companies on the board. Two good examples are Scapa and Zend. In all cases, the commitment to lead a project and provide eight full-time developers on Eclipse projects far outweighs the financial cost. Also, there are individuals who are elected to the board to represent the committers and the add-in provider membership class. Many of these elected representatives are very active board members. I actually think that the Eclipse model does a really nice job of balancing a diverse set of stakeholders. And I really believe that it is factually incorrect to say that Eclipse is "...controlled by a group of large corporations". Well, at least at the Board level. If you look at the committer population, since 85% of our committers are employees of member companies, you could say that there is a control point there. But we are comfortable that a diverse committer population from multiple companies is a good thing for what we're trying to do at Eclipse. You may now return to your previously scheduled debate :-)
  12. Re: BEA and Open Source[ Go to top ]

    The Eclipse community is led by a foundation controlled by a group of large corporations.

    Gavin, I'm sorry to be a pest, but I need to clarify this statement as well.
    Eeesh, apparently this is a touchy subject. OK, I'll be more precise: The Eclipse community is led by a foundation with a board consisting of 13 large corporations, together with 2 or 3 smaller companies who were able to afford to make a commitment of > 1 million dollars per year (approx: dues+8*developer), together with 8 seats set aside for representatives of the rest of the community. With 13 board seats out of 24, combined with the $$$ these guys are spending, it looks to me like the balance of power sits with the large corporate interests. Not that there is anything wrong with that. Really, I don't see this as a bad thing at all. It is perfectly natural. Unlike BEA, I have a big-tent understanding of "open source community". All I am saying is that - just like JBoss - this is not a community which is led and controlled by individual contributors. (As an aside, it looks like the JBoss community has historically had a higher percentage of individual contributors, ie. > 15% of committers.) Of course, in both the JBoss and Eclipse communities, individual contributors are respected and valued.
  13. Re: BEA and Open Source[ Go to top ]

    All I am saying is that - just like JBoss - this is not a community which is led and controlled by individual contributors....Of course, in both the JBoss and Eclipse communities, individual contributors are respected and valued.
    Fair comment. Thanks.
  14. Open email to BEA[ Go to top ]

    My email to Bill and Alfred:
    Hi Bill, Alfred since you guys are interested in continuing the discussion about control of open source projects, I would like to invite either or both of you to debate this issue with me in a public forum, preferably with press present. Since I am the one proposing this course of action, I'll give you the chance to pick the date and place :-) I feel that this would be a great opportunity to clarify both sides of the debate, and will also provide awesome entertainment! Faithfully, Gavin King
  15. Re: BEA and Open Source[ Go to top ]

    A lot of heat (though not much light) has been generated on TSS of BEA CEO Alfred Chuang's comments in a recent article in The Register.
    This is rubish -- your boss Mr. Chuang may be accostumed to yes men agreeing with him, but when he posted those comments in The Register article, he opened himself up to public scrutiny and too bad if he or you don't like the result.
    First, that "open" is a sliding scale, and that those who conflate the notion of open source and total freedom are missing an important nuance.
    This is flat out misleading -- in fact this is not what was said; from the article (http://www.theregister.co.uk/2006/05/24/bea_open_source_sun_jboss/): "According to Chuang: "JBoss is open source software for selected people who are approved [and] can participate. [Participation is] selected by Marc Fleury. That's not an open source process. An open source process is an open community process. You've got to have an open source process and open source community."" Mr. Chuang specifically targets JBoss, as well as his perceived understanding of what their rules are behind who is allowed to commit.
    The first point has to do with control. I would posit that control in open source is properly maintained by effort you place into a project.
    You may posit all you like -- but open source projects are not modelled after your law regarding how they are run. There are many projects which are not free for anyone to commit to, and IMO I think this is perfectly fine.
    Open source offers no intrinsic value to someone who wants to sell more insurance, or make sure customers are notified when their plane is late.
    And if this is who you're trying to sell your software to, then maybe you need to reconsider your business strategy?
  16. Re: BEA and Open Source[ Go to top ]

    The article in question is here : http://www.theregister.co.uk/2006/05/24/bea_open_source_sun_jboss/ (I don't know why it wasn't in the original post...) I think that the key problem is that Chuang conflated the licensing of the software and the governance process of how that software was created. "Open source", like "free software" is really about licensing, not control. "Open development" or "meritocracy" or "professional open source" or "peace and love open source" (whatever that is) is about governance process and control, not licensing. I have my likes and dislikes about the various development models, just like I have my likes and dislikes about various opensource/free software licenses. I choose to participate in the govenrance process and license that I prefer. In the end there always is a governance model. I can't think of any serious open source project that just lets random, anonymous, unknown people add code. There's always some kind of gating mechanism to ensure that code contributions are aligned with whatever metrics or objectives the community around the code has chosen to establish. Ultimately, I think that with any software under an open source license, the control issue is irrelevant- by definition, people are free to take the code and do whatever they want, including trying to establish another separate 'thread of development', aka forking. The fact that the codebases we've talked about in this thread haven't seen that happen means that to a reasonable degree, the control models established around those open source codebases are working. I don't really like all of them personally, but it's hard to argue with the results. geir
  17. It would be disingenuous of JBOSS to suggest that their community is fully-based on a meritocracy.
    Have they suggested this? If not this seems to be kind of a sneaky statement.
    The second point has to do with common sense. For customers with specific customer needs, like companies that just want an efficient call center, or an inventory management system, they could care less about the distribution rights of the software. They do care about lowering call times and reducing cost per Customer Service Representative. What matters to these folks is not where the code comes from, but how good it is. This is the real metric of value. Open source offers no intrinsic value to someone who wants to sell more insurance, or make sure customers are notified when their plane is late. Quality software, regardless of origin, is what matters.
    Said the expensive software company ;-). I'd suggest that anything that dramatically affects any company's bottom line will be cared about whether it is in the IT department or not. BTW: Doesn't every customer have "specific customer needs"? Seems like all this bluster boils down to one thing. BEA hopes that people look at features before price/licensing freedom. I think that is a great, somewhat obvious, idea. But I think it would be disingenuous of BEA to suggest that open source licensing should not play into technology decisions. _____________ George Coller DevilElephant
  18. Re: BEA and Open Source[ Go to top ]

    What matters to these folks is not where the code comes from, but how good it is. This is the real metric of value. Open source offers no intrinsic value to someone who wants to sell more insurance, or make sure customers are notified when their plane is late. Quality software, regardless of origin, is what matters.
    IMHO when a project is open source and the authors are able to admit and discuss the limitations of their code, the quality of that product can quickly become much better than with any closed source product. This also applies to projects where I say "this is a bug" and the author of the code denies that fact. In addition, the risk associating with using an open source product with critical mass is much less than the risk associated with using a commercial, closed source product. Guglielmo Enjoy the Fastest Known Reliable Multicast Protocol with Total Ordering .. or the World's First Pure-Java Terminal Driver
  19. Re: BEA and Open Source[ Go to top ]

    What matters to these folks is not where the code comes from, but how good it is. This is the real metric of value. Open source offers no intrinsic value to someone who wants to sell more insurance, or make sure customers are notified when their plane is late. Quality software, regardless of origin, is what matters.
    +1 I agree with this notiion. "Quality software, regardless of origin, is what matters."
  20. Re: BEA and Open Source[ Go to top ]

    I agree with this notiion. "Quality software, regardless of origin, is what matters."
    I don't agree. There's always a point where quality becomes too expensive. I wouldn't want to pay for a software testing regime of the kind Boeing must surely employ. My server is allowed to crash once a year. Can't say the same about the plane I travel on.
  21. Re: BEA and Open Source[ Go to top ]

    I agree with this notiion. "Quality software, regardless of origin, is what matters."


    I don't agree. There's always a point where quality becomes too expensive. I wouldn't want to pay for a software testing regime of the kind Boeing must surely employ. My server is allowed to crash once a year. Can't say the same about the plane I travel on.
    IMO Your comment is out of context to the intent. Having worked on apps that needed FDA approval, I realize all software quality requirements are not equal. However, I am not sure this is the same use of the word. This is quality comercial software... Quality is not always equal to QA. It can also mean applicability. I believe in the best tool for a job. If it is OS, great. If not, oh well. The cost of adoption for OS is lower, but I have paid for comercial products when OS was avaiable b/c the comercial products were better. I use the best tool for the job (if possible). Rick Hightower (linked in),blog JSF, Spring, and Hibernate training and consulting
  22. Re: BEA and Open Source[ Go to top ]

    IMO Your comment is out of context to the intent.
    Indeed. And the reason is that I feel the same about your post. The point is that some people argue that open source is a quality in itself. So what's the logic of your "quality software, regardless of origin" stance? I take it that you do not regard open access and reusability of source code is a quality in itself. I don't think that is true. The real question is, however, whether these qualities are more important than the qualities a commecial closed source software vendor can offer. It's a tradeoff and the answer may not be the same for every situation, but in any event it doesn't make sense to play it naive and act as if the way software is produced, made available, funded, etc wasn't relevant. It is very relevant.
  23. Re: BEA and Open Source[ Go to top ]

    IMO Your comment is out of context to the intent.


    Indeed. And the reason is that I feel the same about your post. The point is that some people argue that open source is a quality in itself. So what's the logic of your "quality software, regardless of origin" stance? I take it that you do not regard open access and reusability of source code is a quality in itself. I don't think that is true.

    The real question is, however, whether these qualities are more important than the qualities a commecial closed source software vendor can offer. It's a tradeoff and the answer may not be the same for every situation, but in any event it doesn't make sense to play it naive and act as if the way software is produced, made available, funded, etc wasn't relevant. It is very relevant.
    Hmmm.... I am not an OpenSource bible thumper. Hey its okay. I believe in freedom of religion. I get value from OS. I get value from comercial software. It does not have to be OS to be valuable that is all I am saying. I am not against OS. I am not against comercial software.
  24. Re: BEA and Open Source[ Go to top ]



    Hmmm.... I am not an OpenSource bible thumper. Hey its okay. I believe in freedom of religion. I get value from OS. I get value from comercial software. It does not have to be OS to be valuable that is all I am saying. I am not against OS. I am not against comercial software.
    I feel very much the same. This seems to be somwhat of a misunderstanding though. My point was simply that the debate about open-source cannot be reduced to the quality of the software product as seen by the user at one point in time, but has to include the economics software development. Because as a (business) user I have to be aware that I'm not only buying into a static piece of functionality but also into a development process. I depend on the people who create and maintain the software, on their motivatoins and interests and plans, etc.
  25. Re: BEA and Open Source[ Go to top ]



    Hmmm.... I am not an OpenSource bible thumper. Hey its okay. I believe in freedom of religion. I get value from OS. I get value from comercial software. It does not have to be OS to be valuable that is all I am saying. I am not against OS. I am not against comercial software.


    I feel very much the same. This seems to be somwhat of a misunderstanding though. My point was simply that the debate about open-source cannot be reduced to the quality of the software product as seen by the user at one point in time, but has to include the economics software development. Because as a (business) user I have to be aware that I'm not only buying into a static piece of functionality but also into a development process. I depend on the people who create and maintain the software, on their motivatoins and interests and plans, etc.
    Then we agree.
  26. Read the comment again[ Go to top ]

    I agree with this notiion. "Quality software, regardless of origin, is what matters."


    I don't agree. There's always a point where quality becomes too expensive
    Rick said "Quality software, regardless of *origin*, is what matters." What do you mean by expensive? Are you saying you are willing to use a free product because you can afford a crash once a year? Open source products aren't necessarily buggy. Even commercial servers could crash once in a while. I have never seen "let's use free product because we don't have too stringent SLA requirements" argument being used in product selection. C http://chintanrajyaguru.com
  27. I agree with this notiion. "Quality software, regardless of origin, is what matters."


    I don't agree. There's always a point where quality becomes too expensive


    Rick said "Quality software, regardless of *origin*, is what matters."

    What do you mean by expensive? Are you saying you are willing to use a free product because you can afford a crash once a year?
    Yes, if I have the choice between a free product and a product that costs $100,000 and the only difference is that the free product crashes once a year, then I'll take the free product, thank you very much :-) Seriously, what I'm referring to is quite simply the notion of "good enough".
    Open source products aren't necessarily buggy.
    I didn't say that.
    Even commercial servers could crash once in a while. I have never seen "let's use free product because we don't have too stringent SLA requirements" argument being used in product selection.
    Well, I have.
  28. Re: BEA and Open Source[ Go to top ]

    Quality software, regardless of origin, is what matters.
    Cost matters too! The TSS viewpoints expressed about Open Source are generally going to be along the lines of the perspective of enterprise software developers. The argument posted by Bill is pulling away from the developers view and moving towards the customer view and by doing so is validating the statement that there are people who don't care about open source. If the discussion is from the customer perspective then the concerns are: - will the product serve my needs? - how much does the product cost? - how long will it take to get the product up and running? If I have a team of IT professionals and JBoss will serve my needs and setting up JBoss can be accomplished in a timely fashion, I am going to go with JBoss every time because of cost. Also, as a customer I don’t like being forced into having to interact with BEA to make sure my licensing is correct to make sure I am not going to be sued. JBoss takes a less invasive approach and gives the products (and source) away for free and gives a number incase the customer decides they need them. Due to the complexity of setting up an enterprise system, a great number of customers will call. Having used BEA and JBoss over the last 4 or 5 years, comparing BEA to JBoss is like comparing a nice Mercedes-Benz to a Honda Accord. Taking BEA products for a spin is a sweet ride, easy to install and configure, lots of nice features. BEA has been able to focus on achieving corporate and industry goals in a way that open source projects have not had the financial resources to do. As a result, open source projects have typically lagged behind what BEA provides (with some exceptions of course). However, several dedicated and talented people at JBoss have overcome the initial hump of getting a product out for free and achieving critical mass. JBoss is the new 800 lb gorilla in the enterprise software arena and the execs at other companies have to learn how to deal with it. If BEA decided to give away their enterprise products for free and established a business model like JBoss, it very well could be the Enterprise Universe according to BEA. Danny http://www.soamodeling.org
  29. When we did some TCO studies, we found JBoss higher than BEA. But that is only part of the picture. We try to select the platform depending on DNA of the development teams. If the teams can make their had dirty when they encounter any problem or think beyond API into implementation, OSS is good. But if the teams are used to nice wizards, it is difficult to execute a project without nice tools. And our marketing/ product people dont care about what technology we use. But they do care if wrong selection of technology brings downtime. And our experience with JBoss is just great. One recent collection system we architected (with 200 concurrent users and lots coarse grained ops) is working seamlessly. Yes, I love JBoss, Hibernate (to appease Gavin) and most Apache products.
  30. What this argument is really about[ Go to top ]

    Once upon a time, there was a little software company that let you download their application server from their website, try it out for 30 days, reregister and try it for another 30 days, and even hack out the license manager in about 30 seconds. If you wanted to see the source code for the product, there were 1/2 a dozen different decompilers. If you had a problem with the software, you could send an email to support@.... and get a response from one of the developers of the software in about 2 minutes. The response to this kind of model was nothing short of an avalanche, and an acquisition-hungry company came along and bought the company.

    What did the acquirer do? they made the downloads more difficult, inserted 3 levels of support between the customer and the developers, sidelined the low-cost ($1500) version of the product, and pushed "enterprise sales" of the 6-figure variety. This created an opportunity for another company to come along and satisfy the need developers had to get their hands on the binaries to solve their problems, have access to the source code, and get support from their peers.