Home

News: JBoss and Apache clash on shared code continues

  1. JBoss and Apache clash on shared code continues (79 messages)

    JBoss' lawyers sent a notice to Apache regarding code similarities and licensing infringements in March 2004. Apache came back pretty quickly explain that in their minds this wasn't the case, and in fact the reverse was true. Apache recently released their official response to JBoss, and the issue has started up again.

    Summary
    ASF wrote an official response to this letter two weeks ago, claiming that Geronimo project is not in breach of JBoss' IP rights, as Apache developers had originally written the code. In this letter Ceki Gulku, the founder and developer of the Apache Log4J project, one of the projects which JBoss had claimed was in breach of its copyright, said that the opposite was true -- JBoss was in breach of Apache's IP rights.

    "In summary, the source code incriminated by Exhibit A [a section of code which JBoss claimed Apache had copied] existed at the Apache Software Foundation at least 6 months before it made its first appearance at JBoss," said Gulku. "I cannot see how JBoss LLC has any valid claims against the ASF for this particular code. In fact, it appears that by neglecting to attribute the code and follow the Apache License for this example, the violation of copyright would be the exact reverse."

    But, Sacha Labourey, European general manager at JBoss, said on Friday that ASF's letter has not satisfactorily answered its concerns and claims that ASF is still in breach of its copyright. "There answer is not very satisfying -- our lawyers are due to respond," said Labourey. He said that this is an important matter as the Apache license is more permissive than the JBoss license. "We don't accept Geronimo taking code from our licence and putting it in a very open licence," said Labourey, "Our license is not for fun, its important."

    JBoss is licensed under the Lesser Gnu Public License (LGPL) which prohibits closed-source additions to JBoss. The Apache license does not have this requirement, although it does require that anyone using the code attributes the ASF.

    Geir Magnusson, the director of ASF, refuted Labourey's claim saying that the ASF letter goes into detail about why it has not breached JBoss' IP rights and said that Apache take IP issues very seriously.

    JBoss and Apache clash on shared code

    Apache's Official Response

    Alleged Code Copying in Apache Geronimo spurs JBoss response

    Threaded Messages (79)

  2. First there was Apache, then JBoss so my bet is on JBoss pinching Apache code. Admittedly JBoss have committed to Apache but that's the point isn't it?
    It's sad that these two bodies have to spend their money on lawyers, why don't we just set up a coding competition between the two bodies and the winner buys the drinks. At least the money will be well spent and we all get some more code out of it!

    Alternatively perhaps we could use the lawyer money on a drinking competition and the winner gets the IP, Geir would see off any of the JBoss lot, count me in too.

    Why am I blogging on TSS on the weekend? It's my birthday so I can do what I want today! :-)

    -John-
  3. Bye Bye Jboss & EJBs[ Go to top ]

    As a techie -- I hate litigous companies no matter how good their product. I normally don't care howm any lawsuits any company has pending but I do care when they sue other companies whose products I happen to care about.

    Thanks you , Jboss -- you were the really great once!! But now there are much cheaper and better alternatives to EJBs and Appservers. So we don't have to put up with you anymore!!
  4. How to break a license[ Go to top ]

    Here is an accepted way to break a license that is well supported by community:
    http://www.gnu.org/software/classpath/faq/faq.html#faq3_2
    Point: “You can’t look at the source code”.

    Here is illegal way the CDN(CoreDevelopersNetwork / BlueCode) did:
    http://marc.theaimsgroup.com/?l=geronimo-dev&m=106875482022176&w=2
    Point: “We (CDN) *refer* to the code”.
    Or like Compaq story, how they clean room reversed the IBM PC Bios and became a billion dollar company. "Clean rooom"!

    So I think jBoss is on great legal ground, they need to prove the source code connection took place and ideas are extracted, which they can do with CDN’s publicly abailable emails.
    ASF assets:
    http://news.netcraft.com/archives/web_server_survey.html
    jBoss would be a large company then, due to CDN/BlueCode (aka "BroughtCode") actions. jBoss *did nothing wrong*, they were damaged and are a victim. CDN premeditated this, as per their web published plan, to refer to the code step by step.
    Can you say jBoss Lucene, jBoss log4j, jBoss Struts, etc?

    A point of confusion for some is that using jBoss is perfectly legal to write your applications that you then own and use jBoss in any way. What you can’t do is copy and refactor jBoss code and call it “myBlueGlueServer” – the server code is not yours, the application is. It’s very common sense.

    I hope ASF joins them to fight CDN.

    .V
  5. How to break a license[ Go to top ]

    Geir and the others are contributing under the ASL, they are belonging to the most active developers, so they can speek the Geronimo.
    Why don't you accept that?

    Your war mongering is really annoying.

    And what do you say about the missing licence information of the logging stuff in your code base?

    And a point for Juan:
    He did not say bad things about JBoss, but about the JBoss Group.
  6. Vic.

    It is obvious that you know nothing about copyright law.
    For your information:

    1) It is quite all right to have two screens up, one with the original code, and one with your code when you develop the new code, as long as you doesn't use exact the same variable names and do some small changes to avoid it looking exactly similar. After all, that is how most of the Open Source apps is developed!

    If you don't believe me, ask Cameron! :)

    2) The Apache developers say they have originally written the code.

    Then they can of course do what they want, even using the code exact as it is. Remember, JBoss doesn't have the practice of making their developers write over their copyright to the company.

    Regards
    Rolf Tollerud
  7. 1)It is quite all right to have two screens up, one with the original code, and one with your code when you develop the new code, as long as you doesn't use exact the same variable names and do some small changes to avoid it looking exactly similar. After all, that is how most of the Open Source apps is developed!

    No, that is plagerisam and it's illegal. If this was so, tech leads would avoid open source due to liability and that would be the end of O/S.
    People that do this in the name of O/S, should be hunted down for embarasing the O/S community.
    *PLEASE SUPPORT SAFE OPEN SOURCE *
    2) The "Code Developers Network" developers say they have originally written "parts of" the code

    I am not sure if I would risk ASF on that. What if it turns out that they took ALL the code? Geir admits they took ideas. The web posted plan of Core Develoeprs Network said they plan to do take all the code. jBoss developers says the have intermixed CVS code: " file is a derivative of certain files that I wrote. This file was moved to and renamed to the Geronimo project."
    from http://www.theserverside.com/news/thread.tss?thread_id=22337#101159
    The code is comingled it appears and it's likely that they did take all the code, not just some.

    ASF was very ethical until Core Developers Network brought gifts. So this is not JBoss vs ASF, both were duped.

    This is Open Source(jBoss and ASF) is safe
     vs people that plagerise and call it O/S(Rolf aproach #1 above), creating liability.


    .V
  8. Even if you are right (and I think not!), so can not JBoss take any legal measures because they do not have the copyrights. Only the persons that have the rights can sue.
  9. Even if you are right (and I think not!), so can not JBoss take any legal measures because they do not have the copyrights. Only the persons that have the rights can sue.
    Rolf, he's right. Of course he's right. This is the whole point of licensing open source software. Any GPL-compatible license strictly forbids the use of code or a derivative thereof to be used in code licensed incompatibly with the GPL. See http://www.gnu.org/philosophy/license-list.html for the distinctions. The thought that you could simply change a few variable names and avoid violations of copyright (copyleft in this case) is truly laughable. Any elementary school teacher would fail you for such flagrant violations of copyright law.

    Yes, only the copyright holder can legally sue. Note the comment line at the top of every piece of orignial code created for JAS and other JBoss projects:
    /*
     * JBoss, the OpenSource J2EE webOS
     *
     * Distributable under LGPL license.
     * See terms of license at gnu.org.
     */
    JBoss *IS* the copyright holder. NOT the orignial developer, NOT some later developer, NOT the last developer, and NOT the general public. This is THE point. GPL/LGPL licenses protect the copyright holder and make sure that their contributions cannot be modified for use in for-fee software or relicensed in such a way that they can. Period.
  10. laws and copyrights[ Go to top ]

    /* * JBoss, the OpenSource J2EE webOS * * Distributable under LGPL license. * See terms of license at gnu.org. */

    If I read the papers correctly, JBoss is violating ASF copyrights by putting this header in instead of honoring the true origins. That counts at least for the code that obviously mirrors Log4J.
    As far as the other stuff goes, the claim is that the original developers (who are now part of CDN) are free to re-use not only the ideas, but also the code under a different license umbrella.
  11. laws and copyrights[ Go to top ]

    If I read the papers correctly, JBoss is violating ASF copyrights by putting this header in instead of honoring the true origins. That counts at least for the code that obviously mirrors Log4J.

    Nope, you aren't reading this correctly. I said that this notice was at the top of all of the **original code** created by JBoss. There is quite a bit of code in JBoss that comes directly from Apache and retains the Apache license in the header. JBoss also includes all relevant licenses in text files along with the distribution along with all appropriate links to download source, etc.

    Mirroring publicly available API's (i.e., J2EE, Log4J, etc.) is not an issue, nor is mimicking functionality (assuming there isn't a patent on the functionality). As long as the code base isn't modified directly and the results claimed as original work, there isn't a problem. I don't know the code base for JBoss's logging mechanism nor whether they do or don't borrow from or attribute to Apache, so I won't venture an opinion there.
    As far as the other stuff goes, the claim is that the original developers (who are now part of CDN) are free to re-use not only the ideas, but also the code under a different license umbrella.

    CDN is certainly allowed to take ideas from JBoss. It takes an explicit patent or some form of explicit non-compete agreement to legally prevent the export of ideas. The issue is with the code. The license differences specifically prevent CDN from using JBoss code. Period. IF they were to choose a GPL compatible license adn IF they properly attribute the code to create a competing F/OSS app server, then there is no issue. IF (capital IF) they are actually "borrowing" the code and re-issuing it under the ASF license, then this is a clear violation of the LGPL license. Again, period.
  12. JBoss is not the copyright holder[ Go to top ]

    "GPL/LGPL licenses protect the copyright holder and make sure that their contributions cannot be modified for use in for-fee software or relicensed in such a way that they can. Period."

    That is right.

    "JBoss *IS* the copyright holder. NOT the original developer, NOT some later developer, NOT the last developer, and NOT the general public."

    That is wrong.

    They do not they claim that JBoss is the copyright holder. You can be sure that they would have done that if they could.

    The law protects only the copyright holder, nobody else.

    Regards
    Rolf Tollerud
  13. JBoss is not the copyright holder[ Go to top ]

    That is wrong. They do not they claim that JBoss is the copyright holder. You can be sure that they would have done that if they could.

    Rolf, who then do you believe IS the copyright holder? The original coder? The last coder to modify the code?? FSF??? All of the above???? Society at large????? I'm really curious...

    No, by submitting code under the L/GPL and placing that header in the file, you are giving up your copyright to the code. This is NOT a bad thing UNLESS you want to use your code one day in an incompatible, proprietary, closed-source way. You are guaranteeing (assuming that these licenses do hold up in court) that your code can't be used in commercial software and that you are only contributing to the OS community, not inadvertently to the betterment of IBM or Microsoft or Sun or the like. You are also getting the "power" of a real organization and real lawyers in defending this license, which is good.

    Note that you can also retain copyright on orignial code in the L/GPL arrangement simply by claiming yourself as the copyright holder (i.e., putting your own header at the top of the file). JBoss could conceivably still use the code, because they use a compatible license. However, if the code is original and submitted to JBoss using their header, you are designating them the copyright holder.
  14. Copyright is stronger than GPL[ Go to top ]

    "No, by submitting code under the L/GPL and placing that header in the file, you are giving up your copyright to the code".


    No, here is where you are wrong and this is the reason there is so much misunderstanding around the GPL license. The copyright-holder can at any time add something to the code and declare the new version commercial!

    Listen, let us try to find out who is right here without starting a flame war.

    I will sort it out with a example.

    1) The sole copyright-holder (we call him A) Open Sources his software. In this case the GPL license works as intented (or as most people think). But, at any time the copyright-holder can add something and declare a new commercial version. Or he can start to use two versions, one commercial and one open.

    2) Another guy or company (we call him B) fork the code. That is his right. But, and notice this because this is very important, B can not have a commercial version! So in the competition with A, B is severly handicapped.

    You doesn't renonce your copyright simply because you L/GPL your code. To give up your copyright you must explicitly sign it over to someone. Also you does not loose your rights only because you forget to put the copyright notice at the top of the page. Your work is still your work.

    Regards
    Rolf Tollerud
  15. Rolf, who then do you believe IS the copyright holder? The original coder? The last coder to modify the code?? FSF??? All of the above???? Society at large????? I'm really curious...

    the original author is the copyright holder of the original work. The modified version has joint ownership of the copyright
    No, by submitting code under the L/GPL and placing that header in the file, you are giving up your copyright to the code.

    not true. Copyright can be transferred only in a written document signed by the original holder, for example the OpenOffice Joint Copyright Assignment
  16. No, by submitting code under the L/GPL and placing that header in the file, you are giving up your copyright to the code.
    not true. Copyright can be transferred only in a written document signed by the original holder, for example the OpenOffice Joint Copyright Assignment
    Okay, I stand corrected on this. I assumed that the header explicitly claimed copyright for JBoss. On further examination, though, I guess it doesn't explicitly claim as such, only identify the code as part of the larger distribution. So here's a followup question. The GPL FAQ makes the following blanket statement:

    "Part of releasing a program under the GPL is writing a copyright notice in your own name (assuming you are the copyright holder). The GPL requires all copies to carry an appropriate copyright notice."

    And, of course, the instructions on how to apply the license also specifically and repeatedly discuss the need for a copyright notice. I was (incorrectly I suppose) assuming that the first line in the comment section ("JBoss, the OpenSource J2EE webOS") was this notice. However, unlike many other GPL'd project files that I've looked at, none of the JBoss files include an explicit copyright notice or even the standard LGPL boilerplate disclaimer text, etc. There's not even a version number. I know this goes against FSF's recommendations, but does it make it any less valid? I'm also assuming then that the @author tag would probably identify the copyright holder. Must then other contributors add their name to a list of @author tags in order for them to be able to stake a claim?

    This is all confusing.
  17. "The GPL requires all copies to carry an appropriate copyright notice".

    Yes, without it the GPL can not function properly because there are no party that can litigate the case. Without the copyright clause it is nothing more than abandonware.

    Regards¨
    Rolf Tollerud
  18. Vic, read groklaw a little more often[ Go to top ]

    No, that is plagerisam and it's illegal. If this was so, tech leads would avoid open source due to liability and that would be the end of O/S. People that do this in the name of O/S, should be hunted down for embarasing the O/S community.*PLEASE SUPPORT SAFE OPEN SOURCE *

    If you'd been following the SCO vs. IBM debacle a little more closely you'd learn that you CAN legally plagerize. If what is finally released bears no real resemblance to the original, plagerized code there is no problem. It doesn't matter how it became that way (as much as SCO would like us to believe otherwise). So the fact that the geronimo version of the class in question is completely different than the JBoss version eliminates any problems. Primarily, this is to avoid cases exactly like that between SCO and IBM.
  19. please cut the hypocrite bullshit[ Go to top ]

    "If what is finally released bears no real resemblance to the original, plagerized code is no problem, it doesn't matter how it became that way."

    There you have it, nicely expressed in one paragraph. This is the way of the world folks, no matter what Vic says.
  20. As usual, it seems that the money
  21. ... finishing[ Go to top ]

    Sorry, I pressed enter to early. My comment on this issue is that whenever money gets involved, friendship ends. Its hard to belive that sometimes a single line of code can keep lawyers busy for month.
  22. Vic, read groklaw a little more oftens[ Go to top ]

    learn that you CAN legally plagerize

    I have a problem w/that.
    I have a problem that some members of ASF are taking it in that direction.

    That would lead to...
    For example I can download Source to Sun Jdk 1.5, refactor it in a modern IDE, and contribute it to Linux, so that they can distribute it. (Right now Linux is not allowed to distibute Sun Java, users have to download it).

    I can download Source to Resin ....

    So lets all obtusify and get patents.
    Not my battle, just pointing out.
    Peace out,
    .V
  23. learn that you CAN legally plagerize

    I have a problem w/that.

    Vic, I think most people do (or would) have a problem with that, whether or not one can legally "get away" with it.
    I have a problem that some members of ASF are taking it in that direction.

    I'd like to have some indication that such a thing is happening before jumping to conclusions. Remember, a lot of the people that started Geronimo were copyright holders on sources that were used in the JBoss project. If any of the sources outside their copyight domains were "translated," then I think your concern is valid.
    That would lead to...For example I can download Source to Sun Jdk 1.5, refactor it in a modern IDE, and contribute it to Linux, so that they can distribute it.

    That is the concern that leads many companies to not ship their source code. Unfortunate, isn't it?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  24. Cameron wins award[ Go to top ]

    the code was "translated"

    "translated"

    I love that term. It describes the problem exactly.
    Therefore I steal it! So sue me. hi hi

    You will get credit though, for the most innovative word in IT-industry 2004!

    Regards
    Rolf Tollerud
  25. learn that you CAN legally plagerize
    I have a problem w/that.

    Really? You shouldn't. At least, not always.

    Let's take some examples:

    Some coder called Tolkein takes some open-source IPs called Beowulf, ScandMyth, and RingSaga, takes a lot of their best ideas, mixes in a few ideas of his own, and creates something called LoTR.

    Brilliant!

    Some coder called Linus did an OS course (which was full of ideas taken from - guess what - actual OS systems!), took those ideas home, mixes in a few ideas of his own, writes Linux.

    Brilliant!

    Taking ideas from other works is essential to progress in any field, be it art, literature or computer engineering.

    So the moral issue matches the legal one: taking ideas *is* okay, directly copying every line is not, the only question is how closely you're allowed to copy other works. 'Inspired by' vs 'copy of'.


    Sean
  26. To sum it up[ Go to top ]

    1) JBoss can do nothing because they don’t have the copyrights to the work.

    2) Even if they did, ASF can just "translate" the code a la Cameron and everything is ok. Ideas can never be copyrighted.

    So if there ever was a case "written in sand" in must be the JBoss accusations.

    Regards
    Rolf Tollerud
  27. learn that you CAN legally plagerize
    I have a problem w/that.
    Really? You shouldn't. At least, not always.Let's take some examples:Some coder called Tolkein takes some open-source IPs called Beowulf, ScandMyth, and RingSaga, takes a lot of their best ideas, mixes in a few ideas of his own, and creates something called LoTR.Brilliant!Some coder called Linus did an OS course (which was full of ideas taken from - guess what - actual OS systems!), took those ideas home, mixes in a few ideas of his own, writes Linux. Brilliant! Taking ideas from other works is essential to progress in any field, be it art, literature or computer engineering. So the moral issue matches the legal one: taking ideas *is* okay, directly copying every line is not, the only question is how closely you're allowed to copy other works. 'Inspired by' vs 'copy of'.Sean

    Vic, the above is what I wanted to say. Except that I'm not that skilled in english, like Sean. I completely agree with the above statement.
  28. How to break a license[ Go to top ]

    CDN caused part of this mess together with idiots like fat boynes and dumbstrom. kind a ryhmes really
  29. How to break a license[ Go to top ]

    Vic,

    I did some paralegal work doing copyright and trademark stuff to help pay my way through grad school, and I think you're dead wrong about copyright law.

    I'm (legally) allowed to read your novel, and write another one that steals your best ideas (80% of modern fantasy authors would be in court if people weren't allowed to steal that much from Tolkien, Fritz Leiber, etc). But I'm not allowed to directly steal your phrases, or to write a direct copy.

    Likewise with code. If your code is open source then I can legally read your code to find the ideas behind it, take those ideas, and write my own code using them - copyright does not protect you there. Inevitably there is a legal grey area about exactly where you fade from 'inspired by' to 'copy of', and that's where the expensive lawyers come in.

    There are many ways to implement an EJB container, but there are not so many good ways. Taking good code source and scouring it for ideas sounds only sensible if you're writing one. There is no (legal) question about whether or not they can do that: only about whether they were too derivative of the original work.


    Sean
    ---
    Disclaimer: I'm not a lawyer, and even if I was I'm certainly not an AMERICAN lawyer (and despite the Paris Treaty copyright law does vary somewhat internationally). If you need real legal advice, go pay someone for it.
    ---
  30. How to break a license[ Go to top ]

    Apreciate it Sean and I agree that it may be legal, but if you aply the GNU Classpath as standard it's the other way.. But I said:
    " this may or may not be legal, but I think it is not ethical "
    I am not a lawyer, but I have an opinion (on everything it seems).
    Forcing ethics on people is like forcing religion on people, everyone has their value system, but to me plagerisam stinks, and I hope the people that plagerise do not meet w/ success, so I kick them when it's convinient.
    Also for the Makevelians: "Should we get patents now" ... just to assure atribution?.

    In the end, JBoss is working with ASF on Tomcat, and that is good and I hope O/S comunities work together on more things in the future.

    .V
  31. Do you want to take the Red Pill ?[ Go to top ]

    Hey this pill is brown and it seems to say -Turbolax- on the bottle .... mmmnnnnn ... drink up good buddy .. good pal !

    --b
  32. My affinity is with ASF[ Go to top ]

    My affinity is with ASF, and I think ASF should vote out the bad apples at the Apache Conference, that stand for "it's legal but not ethical". In the spirit of "Free as in free speach" let me add:
    No one asked ASF directors to refactor one license to another, which is what I think Geir's point is that they could do that.

    Where the Geronimo code came from, or at least was inflenced by (inflence is enough for a legality test):
    http://marc.theaimsgroup.com/?l=geronimo-dev&m=106875482022176&w=2
    Quote: "Exibit D" should be deleted so no record exits in CVS"

    Geir, the author of the response here, works for GlueCode as per bellow (he says "my employer is GlueCode" below)
    http://www.mail-archive.com/general at incubator dot apache dot org/msg03726.html
    but used ASF letterhead on response,
    And other members "Core Developers network" (see list) that might be using ASF as a shield for their activity also work for GlueCode
    http://www.gluecode.com/website/company/management.jsp for those that think that Core Developers Network was the one that cuase the rife btwn ASF and jBoss; the 2 O/S groups used to get along splendid.
    I hope that GlueCode developers are not incliend on takeing code from their current clients that does not belong to them or that was done for their client.

    Their action could lead more people to patent their designs for copyright protection, if Geir establishes that copyright is not sufficient protection, making it harder to develop sofware.

    Other than "Core Developers network issue", jBosss is still helping ASF w/ Tomcat, etc.
    I urge ASF members to look for a way to resolve this, even if it means puting the accused parts of Geronimo on sf.net, outside of ASF hosting, for the Core Developer network or GlueCode to defend the code or continue it's development, assuming jBoss would drop damages against ASF.

    .V
  33. Intelligence is based on ability to copy and improve upon peer's experience. This is basic law of nature and no intelligent living creature is exempt from it.
    We are genetically programmed to borrow habits, tricks etc. during millions years of evolution. Therefore all limitations on that are very harmful to the very existence of humans as species on the space ship Earth.
  34. We are genetically programmed to borrow habits, tricks etc. during millions years of evolution. Therefore all limitations on that are very harmful to the very existence of humans as species on the space ship Earth.

    But we are also genetically programmed, during the millionsyears of evolution, to potect ones' own food, mates, and territory and to kill others for these.
  35. We are genetically programmed to borrow habits, tricks etc. during millions years of evolution. Therefore all limitations on that are very harmful to the very existence of humans as species on the space ship Earth.
    But we are also genetically programmed, during the millionsyears of evolution, to potect ones' own food, mates, and territory and to kill others for these.

    I have never heard about a wolf that has killed another for copying hunting tricks.

    And there is huge difference between defending own food and depriving others from having their food.

    Carnivores abandon their food when they are not hungry anymore and do not kill other carnivores it they hunt the same kind of deer.

    Anyway,I guess copying comes first.
  36. I have never heard about a wolf that has killed another for copying hunting tricks.

    True, but what if the wolf is trying to take food from another wolf?

    Even worse, my dog keeps threatening my cat whenever the cat is trying to get near the cat's own food.

    For us in software, publishing, etc., the idea and intellectual property are our food. Nobody is you just learn programming w/o imitating the protected idea.

    And ever since life exists on this planet, competition has been the main driving force of evolution (and thus innovation).

    I'm not here to defend either JBoss or ASF. Just try to point out a simple fact of life that many here seem forgoten.
  37. Nobody is you just learn programming w/o imitating the protected idea..

    Typo: Nobody sues yu for just ...
  38. My affinity is with ASF[ Go to top ]

    Hi Vic,

    Your post totally confused me, and IDEA was unable to reformat or refactor it. Could you explain exactly what it is that you were saying?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  39. My affinity is with ASF[ Go to top ]

    Cameron,
    I think this may be plagerisam.
    No one at ASF is saying that Core Developers Network did not import the jBoss code into Geronimo, they are saying that they refactored it and that it is now different.
    I can try to explain it another way. Read page 8 of Geir's PDF:
    ""code in Geronimo is a diferent implemnation of an jBoss idea".
    and
    "the current code has kept the idea that .... "
    Geir is admititng that it is the same idea.
    This is copyright vs patent question. So the court now has to decide is this derived or inflenced by jBoss. Clearly they had acessed the code.
    It could go either way IMO, in wich case jBoss would be awarded damages, ASF has like $100K in the bank, so jBoss could end up owning the entire collection of ASF copyrighted works, and no one wants that.

    My comperhension is that Geir is saying that jBoss did not patent the idea, and since the code was changed and is not verbatum, it's ok to change the license.
    So the shades are... was this inflenced ? - which the jurry would rule on.

    If people decide to protect their idea's w/ patents instead of just copyrights, you and others would (which this case may be pre-meditated and planed by Core Developers Network) have to check a catalog of patented ideas just to avoid using them in your projects.

    Or more extreme lets say you Cameron are at a client and finish an application for company Blue. Then company Red hires you and you take the code w/ you and refactor it to remove com.blue to com.red. Are you allowed to take the code w/ you and refactor it (and IDE could do this in minutes)? What about code of other developers on the team? What about ideas and designs of other developers?
    Would companys ever use consultants if this was the case?

    My point is that this may or may not be legal, but I think it is not ethical to not atribute ideas to the origniator and most people including me, hold ASF and O/S as some of the most ethical engineers.

    .V
  40. What have JBoss to do with it?[ Go to top ]

    Vic,

    The question of "has Apache used, derived from or was influenced from this particular code in question" is a matter for the copyright owners of that code, not JBoss.

    Regards
    Rolf Tollerud
  41. My affinity is with ASF[ Go to top ]

    Hi Vic,Your post totally confused me, and IDEA was unable to reformat or refactor it. Could you explain exactly what it is that you were saying?

    I doubt that there's a translation system yet invented by man that could make sense of that one.
  42. My affinity is with ASF[ Go to top ]

    Hi Vic,Your post totally confused me, and IDEA was unable to reformat or refactor it. Could you explain exactly what it is that you were saying?Peace,Cameron PurdyTangosol, Inc.Coherence: Shared Memories for J2EE Clusters

    :)

    This should indicate that Tangosol went has the wrong IDE in their stack.
  43. i don't know for others, but personnaly, i prefer see news like this one :
    http://www.infoworld.com/article/03/12/08/HNobjectweb_1.html

    ObjectWeb, Apache team on open source J2EE

    Why not help each others Marc ? :)
  44. i don't know for others, but personnaly, i prefer see news like this one : http://www.infoworld.com/article/03/12/08/HNobjectweb_1.htmlObjectWeb, Apache team on open source J2EEWhy not help each others Marc ? :)

    Speaking as a member of the Apache Geronimo project, we're really happy about the collaboration we've had for over a year now with the ObjectWeb community. We share code and ideas, to the mutual benfit of all parties, and still maintain our independent identities.

    If you are coming to ApacheCon this weekend, or live in Las Vegas, come to our BOF on "Collaborating Communities" where we'll talk more about what we've achieved so far, and what we plan for the future.

    It's BOF01, on Monday, November 15, 2004 at 8pm

    geir
  45. As a member from Objectweb, i'm very happy to see how much peoples can succeed to collaborate... i'm proud.

    It's always a pleasure for me to see peoples set aside their problems to be able to work for community.

    I won't be at apachecon as i'm in france, but maybe i will go to objetwebcon in january :)
  46. Is it that the JBoss people want to stop Geronimo attracting more users?
  47. Motivation[ Go to top ]

    While your supposition may or may not be the actual motivation here, there is a valid legal and ethical question at work. As I understand it, the GPL and LGPL give developers confidence that their contributed work can only be used in the context of freedom -- that is, it cannot be modified and distributed under non-free terms. This stands in contrast, say, to BSD-style licenses, which did nothing, for example, to prevent Apple from modifying BSD and distributing it under non-free terms as OS-X. Some developers don't care if their contributions are used in non-free contexts. Others do. The GPL and (to a lesser extent) the LGPL give those contributors who do care a legal means by which to make sure that their contributions are only used in free contexts.

    The Apache license, as I understand it, is more similar to the BSD license than the GPL or LGPL. A consequence of this is that is possible to include Apache licensed libraries and software and such in an LGPL or GPL distributed product (like JBoss, which is LGPL), but not the other way around.

    Although the inclusion of JBoss code in Gernimo would not limit the freedom of the JBoss code, it does mean that if the JBoss code (or a derivative thereof) is distrubted in Gernimo, it would be unprotected from others taking it to non-free places. It would also mean that the JBoss code was being distributed in violation of the license agreement (LGPL) which is the only thing granting the right to distribute in the first place.

    That said, as another poster has already point out, this could be a very complicated case in which ideas were extracted but not actual code. Which puts you in a grey area between copyright law and patent law.
  48. Motivation[ Go to top ]

    Alarik :
    Although the inclusion of JBoss code in Gernimo would not limit the freedom of the JBoss code, it does mean that if the JBoss code (or a derivative thereof) is distrubted in Gernimo, it would be unprotected from others taking it to non-free places. It would also mean that the JBoss code was being distributed in violation of the license agreement (LGPL) which is the only thing granting the right to distribute in the first place.

    Please re-read our findings and conclusion. We do not believe that JBoss code or a derivative work of JBoss code was included in an Apache codebase. The ASF strongly, strongly believes in IP rights and license rights, of any license, and works very hard to ensure that any code contributed to any ASF codebase or project receives proper vetting and review.

    That is the reason the ASF has the Incubator - a mechanism to act as a gatekeeper and review point for incoming contributions - as well as our set of contributor and and software grant agreements, through which we do our utmost to ensure that contributions able to be licensed to the ASF.

    - geir
    (speaking as ASF member)
  49. Motivation[ Go to top ]

    Sorry if I implied I have any kind of an informed opinion about whether or not JBoss code is in Geronimo -- I have none. I was just sort of stating a set of hypothetical situations, hence the conditional tense. Apache is an amazing organization. My deep gratitude to you all.

    Alarik
  50. More to come[ Go to top ]

    Is there anyone out there that didn't see this type of stuff comming?

    I'm sure we will see more of this to come. I have always liked the JBoss server, but not the Jboss group.

    My dislike for the Jboss group is now stong enough for me to stop using the JBoss server. They are not the only Free J2EE server out there, and I'm not even sure they are the best anymore.
  51. JBoss should drop this and move on[ Go to top ]

    If the Geronimo team really did copy code from JBoss improperly, then the JBoss team should have been able to come up with more convincing examples than these. The whole affair smells of a dubious attempt to derail fair competition.

    It's surprising, because JBoss has made so many smart moves during the last 18 months, and, from outward appearances, they enjoy a big lead over Geronimo at implementing both J2EE 1.4 and J2EE 5. But this affair makes them look very scared of the other guys, almost desperately so.

    Both sides can handle a little competition, can't they? Move on, and may the best app server win.
  52. JBoss Position[ Go to top ]

    It seems we must post to give some further information and clear things up...

    JBoss actually sent a letter about a year ago to Apache about concerns we had. The essence of our concern is based around the potential taking of code and IP and moving it from LGPL to and ASF license. Our committers to the JBoss opens source project feel strongly about not having their code taken non-open source and a switch to an ASF license would open up that possibility.

    Our position was, and continues to be, that this is Apache's concern to monitor. We do not feel it is JBoss's job to monitor and review Apache code. We spent about 2 hours to point out some areas to Apache for the purpose of the letter a year ago. We have not spent more time on this and it is clearly not our focus. Our focus is on building out a full open souce middleware system over the next couple of years that is based on the Professional Open Source model. We were actually surprised to see a letter from Apache a year after we sent our original to them.

    JBoss is a friend of Apache. We contribute significantly to a number of Apache projects, and use a number of those projects in our offerings. All of our software is free and open source. Thousands of users incorporate our technology across the world and we have no problem with that as long as the license terms remain constant - which means it continues to stay free and open. You can read more on this topic in one of Marc's blogs - http://jboss.org/jbossBlog/blog/mfleury/?permalink=From+GPL+to+BSD+to+LGPL%3A+On+the+Issue+of+Business+Friendliness.html

    Apache's business model is different. Part of the difference in opinion on licenses is ideology - just like the ideology of the FSF backers (feels a bit like the Presidential election). But a significant factor in the Apache business model is that the large vendors that support Apache like the idea of being able to take code and then turn it proprietary for their own use (like the Apple-BSD example earlier in this thread). Because the license model is different, Apache seems not to want to acknowledge JBoss and other LGPL code. For example, the logic for creating Geronimo was to create an app server that "is licensed under the Apache License, and passes Sun's TCK for J2EE 1.4, and reuses the best ASF/BSD licensed code available today" (http://geronimo.apache.org/). Since JBoss App Server is now J2EE Certified, the only argument left to creating Geronimo is that there needs to be an Apache license.

    JBoss will continue to be a friend to Apache. On Monday we will be making a major announcement of another Apache project beyond Tomcat that we will be contributing to in a significant way. We will be making that project stronger by contributing to the community, and we will not be changing the license. We welcome Apache embracing the JBoss App Server and other JBoss projects in the future - lending their support to these large and growing communities.

    Bob Bickel
    JBoss
  53. JBoss Position[ Go to top ]

    Bob:
    It seems we must post to give some further information and clear things up...

    Thanks Bob for the nice post. I, like many, many others, am happy to move on. As Cameron says, "Peace".

    - geir
  54. Bob said:
    We do not feel it is JBoss's job to monitor and review Apache code...JBoss will continue to be a friend to Apache.

    But Sacha said:
    ...that ASF's letter has not satisfactorily answered its concerns and claims that ASF is still in breach of its copyright. "There answer is not very satisfying -- our lawyers are due to respond"...

    So, Bob, are you saying that JBoss will instruct its lawyers to start another round of this, or that you're going to drop it and be friends from now on?

    Or merely than one can keep threatening legal action and still be friends?
  55. Bob said:
    We do not feel it is JBoss's job to monitor and review Apache code...JBoss will continue to be a friend to Apache.
    But Sacha said:
    ...that ASF's letter has not satisfactorily answered its concerns and claims that ASF is still in breach of its copyright. "There answer is not very satisfying -- our lawyers are due to respond"...
    So, Bob, are you saying that JBoss will instruct its lawyers to start another round of this, or that you're going to drop it and be friends from now on? Or merely than one can keep threatening legal action and still be friends?

    As allways when asked straightforward what their position really is, aside from the official marketing bullshit (remember the reply to fake posts acusations ? mf just stated publicly that this will be prohibited in the future, thus aknowledging the guild btw, hoping all the acusations will go away fast and they can continue doing business like they were "clean") the JBOSS group won't stop their dirty ANTI FAIR COMPETITION IN OPEN SOURCE campain.

    Yes, in spite of all GOOD SUPPORT/CONTRIBUTION they have to Tomcat/Hibernate/Jboss app server (please note that I definately make the difference between JBOSS group and JBoss application server) they still play dirty against fair competition. Their competition right now is Geronimo. Mf has stated many times that he is a fan of Microsoft practices. I do not want to start another ANTI MS flamewar here but I think what JBOSS group is doing against ASF right now is just what MS is doing against their competitors. Include here everything you know about MS 'best' practices in this mater.

    JBOSS group is working agains fair competition in the Open Source arena. As it's been said in this thread before, ASF has colaborated succesfully with ObjectWeb in the SOLE interest of the OS java community. What I want to say is that when it comes about MONEY, JBOSS group f*cking LLC don't give a f*ck on the open source community.

    To JBOSS group LLC: Stop trying to full people into how positive you are to the OS community, all you care about is YOUR MONEY COMMUNITY which is fine with me AS LONG AS YOU DON'T kill another OS projects or transform the OS marketplace into a war arena that won't be trusted by anyone doing business. I hope the damage you are making now to to open source community won't stop people from contributing and won't stop big vendors from investing in it as a real alternative to monopoly like Microsoft's in the op system arena.

    To Vik: When I start writing a software I am actually evaluating what the OS has in that field. Most of the time I embrace a OS project and user it or extend it and use it. But sometimes nothing suits my needs and I start my ow software package BUT 90% I start from the good IDEEAS I found in those OS projects. Does that make me eligible for prison ? Using other's ideeas ? Wasn't OS mean to be for software inovation and evolution and other 1000 GOOD things ? We are talking about Dain Sundstrom and James Strachan and Jeremy Boines, David Jencks and another X people here. DO YOU REALLY THINK that these guys have stolen something from the OS community ? What I see is that these guys are actually GIVING to the comunity ALOT MORE that they have taken ever from it.
  56. Our focus is on building out a full open souce middleware system over the next couple of years that is based on the Professional Open Source model.

    Can I observe that by implying that the rest of the OSS community is "unprofessional" you are insulting the people who often provide the core underpinnings to your product suite. It doesn't lead to good relationships, which is unfortunate.
    We contribute significantly to a number of Apache projects, and use a number of those projects in our offerings.

    Sometimes you even contrib back, though not, I note, Apache Axis. This creates issues on those occasions we have to field support requests; they scare us as much as anything coming from Axis on WebSphere.


    Because the license model is different, Apache seems not to want to acknowledge JBoss and other LGPL code.

    Acknowledge is a complex phrase. ASF code is not allowed to link to or contain any LGPL code. The exchange of letters appears to be a complaint that Geronimo did, and a rebuttal to that complaint.

    JBoss will continue to be a friend to Apache.
    On Monday we will be making a major announcement of another Apache project beyond Tomcat that we will be contributing to in a significant way.

    Excellent! One thing both Apache and JBoss have done is brought OSS into the enterprise. JBoss have been central in commoditising -EJB, while Apache's push for OSS project access to TCKs made that TCK tested JBoss possible. They also created the set of libraries which form the underpinnings of many J2EE engines, both open and closed.

    We welcome Apache embracing the JBoss App Server and other JBoss projects in the future - lending their support to these large and growing communities.Bob BickelJBoss

    I'm sure Apache projects welcome JBoss embracing the Apache product suite - lending their support to these large and growing communities.

    The OSS community doesn't gain from a split -real or perceived- between key players. I work on both ASF/BSD-licensed projects (night) and on LGPL (day job), and think both have roles. Nobody should be enemies because they believe in a different software license, within the OSS family.

    Steve Loughran
  57. JBoss Position[ Go to top ]

    Bickel says:
    We were actually surprised to see a letter from Apache

    Uh, so when you had your lawyers send a letter to Apache and said in the letter "I am writing...to request your response" and "I look forward to your reply", what were you expecting? How was a response surprising when you had your lawyers specifically ask for it at least twice?

    Or are you saying the initial letter was just a publicity stunt and you didn't really expect a response?

     - Don
  58. The Difference? Leadership.[ Go to top ]

    [..] the logic for creating Geronimo was to create an app server that "is licensed under the Apache License, and passes Sun's TCK for J2EE 1.4 [..] Since JBoss App Server is now J2EE Certified, the only argument left to creating Geronimo is that there needs to be an Apache license.

    I believe I represent many other developers when I tell you that the main attractiveness of Geronimo is not the license - it is the focus and leadership of the project as opposed to JBoss. There is a big, big different there.

    I predict that Geronimo will eventually explode in popularity, and I believe that it will do so _despite_ not being under the GPL.
  59. Our [JBoss'] focus is on building out a full open souce middleware system over the next couple of years that is based on the Professional Open Source model.

    This is a great idea. Actually, a full open source middleware stack of high quality is currently available at ObjectWeb! Today. Not in a couple of years. All is professional open-source -- ie, open-source developed in a professional way, by professionals.

    Why not depart from the NIH syndrom and discontinue reinventing the wheel?
  60. Knowing the history of JBoss and some of it's founders (we all know who wrote the real JBoss Kernel and who stole it) I think it is a justified educated guess that it is JBoss who is guilty of plagarism not apache.
  61. Just seems odd...[ Go to top ]

    It seems to me the issue here is did the Geronimo project borrow code or just ideas? The spirit of the GPL is to foster community and "open" software development. As such, the license to me also encourages new ideas, but these ideas are shared with the community freely.

    If the GPL includes the idea(s) behind the software, it's no better than a patent and a patent should be used. A patent allows you to force whatever license you want on whomever you want. However, this seems very counterintuitive to the spirit of the GPL and open source software in general. It's made to be open for people to see it and if they choose to use that exact source code. If they choose to use that source then yes they should be bound to the license. However, if they see a great idea and feel it should be under a different license, and take the time to recreate those ideas in new code, then it seems the GPL would support this.

    If however the GPL does indeed include the scope of the ideas embodied in the code, then I must say that the GPL is counter productive to the open source movement. Likewise, if any companies would like to push or try to enforce that doctrine, then I must say that I can not support nor continue to support those projects in any means. At the core of my ontology is that free/open software is created for the express purpose of helping, growing, and/or fostering the development and/or advancement of peoples around the world. Any license that claims ownership to the ideas and not just the specific implementation is not a license I support, because in the end ideas, to me, should always be free for all to share.

    Additionally, I must agree with other posters in this thread that the money on lawyers (assuming they're not working pro bono), could be much better spent. However, I must concede if as suggested Geronimo has borrow code (not ideas) from JBoss, then the money may be well spent protecting the ideals of the developers in question and I would even commend JBoss on fighting for the rights of said developers.

    Cheers,
      Tyler
  62. Let's put it straight[ Go to top ]

    Ok, I'm a JBoss user, and I also feel that I owe some beers to the ASF guys... but I have to tell this.
    And I'm deeply sorry if someone feels offended, that's not my intention. I don't know anything about laws, I know that probably my tech skills are quite poor (as my english) so usually I just read these threads, but I know something...
    All this stuff about LGPL versus Apache license is just FUD. The real matter is that JBoss Group are deadly scared about a real competence from other Open Source EJB server.
    All we know what is the real spirit of Open Source, and sorry you guys from JBoss Group, but you don't deserve that name, you don't deserve to be called Open Source. Maybe your code has that license, your acts tell something opposite.
    I'll have to keep using JBoss at work, but at home on my own stuff? Not anymore.
  63. Let's put it straight[ Go to top ]

    All this stuff about LGPL versus Apache license is just FUD. The real matter is that JBoss Group are deadly scared about a real competence from other Open Source EJB server.All we know what is the real spirit of Open Source, and sorry you guys from JBoss Group, but you don't deserve that name, you don't deserve to be called Open Source. Maybe your code has that license, your acts tell something opposite.

    Juan, what are those acts you are talking about? My impression is that JBoss has lost a lot of sympathy in the community, not because of actions that actually matter to developers, but rather by exhibiting aggressive, ostentatious language enriched with marketing phraseology. But that shouldn't lead us to lose sight of the real issues.

    JBoss uses a license that says, you can use this code but you can't make it closed source. So developers who write the code have a garantee that they can profit from any extensions/modifications made by others to their code. Apache uses a license that allows companies to take the code, modify it so that it becomes a proprietary piece of software, and distribute it without the source code as long as they declare that ASF code is being used inside.

    I'm not saying that any of the models is not legitimate because after all, developers can choose under which license they want to distribute their code. But I find it hard to follow your conclusion that JBoss doesn't "deserve" to be called open source or that they violate the "spirit" of open source. What is this spirit and what does anyone have to do to "deserve" the honor of being called open source? For me, it's quite simple: If I get to see, modify and redistribute the source code, then it is open source. If I'm not allowed to do that, then it's not open source. There is no spirit apart from that and there is no requirement for Marc Fleury to talk nice, or to not be a business man, or to not employ lawyers. The commercial companies that use ASF code in their products have lawyers and marketing people too. I have no problem with that, because it doesn't affect what I can do with the code.
  64. Let's put it straight[ Go to top ]

    Juan, what are those acts you are talking about? My impression is that JBoss has lost a lot of sympathy in the community

    -1, and from my point of view, there are more people who got tired of all this legal issue (just my point of view)
    not because of actions that actually matter to developers, but rather by exhibiting aggressive, ostentatious language enriched with marketing phraseology.


    +1, they are really proffesional, my point is that they don't act as open source.
    But that shouldn't lead us to lose sight of the real issues.JBoss uses a license that says, you can use this code but you can't make it closed source. So developers who write the code have a garantee that they can profit from any extensions/modifications made by others to their code. Apache uses a license that allows companies to take the code, modify it so that it becomes a proprietary piece of software, and distribute it without the source code as long as they declare that ASF code is being used inside.I'm not saying that any of the models is not legitimate because after all, developers can choose under which license they want to distribute their code. But I find it hard to follow your conclusion that JBoss doesn't "deserve" to be called open source or that they violate the "spirit" of open source. What is this spirit and what does anyone have to do to "deserve" the honor of being called open source? For me, it's quite simple: If I get to see, modify and redistribute the source code, then it is open source. If I'm not allowed to do that, then it's not open source. There is no spirit apart from that and there is no requirement for Marc Fleury to talk nice, or to not be a business man, or to not employ lawyers. The commercial companies that use ASF code in their products have lawyers and marketing people too. I have no problem with that, because it doesn't affect what I can do with the code.

    What spirit? Don't get me wrong, I do know the difference between LGPL and the ASF license but before JBoss Group came into the play zone, all the attacks against OpenSource (as ASF is) mainly came from outside the community.

    Legal language, threats about sue them, oh, come on, JBoss using ASF code all along, JBoss paying OpenSource developers in order to say that they have acquired another product, JBoss all around... my boss (at my actual job) worry about if using Open Source (for example to use JBoss as EJB server) will have further consequences in the future. You know how hard was to convince managers to use Open Source, we know how easy it will be to get them back to "normal" way, as much as the license keep going down and there are option to get in troubles using OpenSource tools.

    We can keep arguing about the consequences of LGPL, if "someone" is going to pick up that code and become close source... that's the point of JBoss Group, just talk about that matter. Meanwhile a lot of people like me, getting tired of this ridiculous circus. Sure,
    there is no requirement for Marc Fleury to talk nice
    , and also there is no requirement to my, as "end user" of Open Source tools, to believe that they "deserve" to belong to the OpenSource category.
  65. Let's put it straight[ Go to top ]

    Legal language, threats about sue them, oh, come on, JBoss using ASF code all along, JBoss paying OpenSource developers in order to say that they have acquired another product, JBoss all around... my boss (at my actual job) worry about if using Open Source (for example to use JBoss as EJB server) will have further consequences in the future. You know how hard was to convince managers to use Open Source, we know how easy it will be to get them back to "normal" way, as much as the license keep going down and there are option to get in troubles using OpenSource tools.

    I agree with you that it is not good to have too much legal wrangling. It puts some people off. But on the other hand, if you look at commercial companies, there are legal battles all over the place and it has always been like that, not just in IT. I think as developers we should not see these kinds of things as moral issues where we have to take sides. If a piece of software suits my needs and comes with the right license for my busiess model, I use it.
  66. Now that I think about it more, maybe ASF is just being used here by "Core Develoers Network" developers?
    This is realy GlueCode(Core Developers Network) vs jBoss.
    Here is a theory # 142:
    The letter was posted by Geir, that works for GlueCode, maybe as publicity stunt to make jBoss look bad. Maybe they lost a deal to jBoss. A lot of maybees.
    No one is arguing that Core Devlopers Network did not copy the jBoss code into Geronimo, Geir is just saying that they refactored it. (There are apache emails saying so at mail archives).
    This way, ASF takes the heat, and they can use it at GlueCode becuase of the new license. Clever. Plus they can say "come to GlueCode if you do't like jBoss". Of course the Core Developer Network IS the bad part of jBoss, now at GlueCode.
    Maybe?

    ASF is just stuck in the middle and the only one w/ a lot to loose and no one looking out for them. The danger to ASF is real, should jBoss (and their $10MM in the bank) go after them, but I do not think they would, thus... GlueCode is out scott free, with the "derived ideas" that came from jBoss.

    Gier is saying OK, shows us what ideas are yours that are in the Apache's incubator, not Apache proper?
    and jBoss is saying: that is our code, and we were damaged, and Apache incubator is owned by ASF.

    I personaly think EJBs don't work, so I don't have a use for either. But I do use Tomcat, developing by jBoss under ASF license. I do not think that Geir or other's with GlueCode should be representing ASF.

    Now if there way for the 2 open source groups to work together and go after Core Developers Network members for the damages.... he, he. Merry X-Mass.

    .V
  67. clarification on Tomcat[ Go to top ]

    Several companies have contributed to Tomcat, including Sun, IBM, and JBoss. I don't think Apache foundation is in any real trouble here. Since the process is transparent and apache requires committers to sign a contract saying the code was written by the committer and not stolen. I don't use geronimo and i don't know the code base, but there aren't hundreds of ways of implementing an EJB container efficiently.

    I would argue there's probably a handful of ways of implementing an EJB container so it meets the compatability test and achieves good performance. That means if you were to compare the top 5 EJB container, you're probably going to find a lot of similarities beyond the base API.

    when push comes to shove, it's easy to rewrite the piece of code. It means slowing down a bit, but so what. having the latest and greatest feature isn't always good. Having a reliable and robust server is far more important to me. I do hope things get worked out so that jboss and geronimo can compete head to head.
  68. Germino release=JBOSS worries[ Go to top ]

    Lets hope for JBOSS demise ASAP.
  69. This FUD is just plain sad[ Go to top ]

    Folks, this is just crazy. This seems to be an "Attack JBoss / Love Apache" thread, seemingly based on the perceived fact that Apache is somehow a better open source community member and truer to the open source ideals. Putting any perceptions of Marc Fleury aside (IMO Marc's perceived arrogance, real or not, is irrelevant), let's really compare the two companies:

    Apache
    ------
    License: BSD-based ("free" as in "anyone can do just about anything they want with it")
    Developers: Some unpaid, many paid (by IBM and other corporate "sponsors" (including JBoss, incidentally))
    Products: Freely distributed; often modified and incorporated into proprietary, fee-for-license products (which is why they have such good corporate sponsorship)
    Source of Income: Primarily donations and corporate sponsorship

    JBoss
    -----
    License: FSF-based ("free" as in "freedom")
    Developers: Some unpaid, many paid (by JBoss itself)
    Products: Freely distributed; fee-for-service; licensing prevents direct modification and inclusion in fee-for-license products (which is whay they can't rely on corporate sponsorship)
    Source of Income: Primarily self-supported by service contracts (again necessary because they don't get corporate sponsorship)

    I'm at a loss to see what is wrong about JBoss's model that makes Apache better. Or vice-versa for that matter. They both contribute their code freely to the open source community. Both have to make money somehow and choose different means of doing so. There is room for both models, IMO. Demonizing one in favor of the other seems silly.

    Also note that JBoss isn't afraid of another open source competitor. JBoss makes their money because there is a perceived benefit for having a "legitimate" organization backing an open source product and able to provide "from-the-horse's mouth" support. Another open source competitor doesn't diminish this value in the slightest. They ARE (rightly) afraid of the work contributed to JBoss showing up "legally" in the next version of Websphere. These letters are simply a gentle nudge to remind Apache of their responsibilities to respect the LGPL license. You may not like the idea of getting lawyers involved in open source; but like it or not, there are licenses involved, and where there are licenses, there are lawyers. Rest assured, however, that nobody benefits from suing and/or destroying Apache. JBoss benefits from Apache products as much as IBM and others do. Apache's model has benefits that JBoss doesn't. The "Professional Open Source" paradigm plain doesn't work for everything, and JBoss knows this very well. JBoss has no interst in "owning" the rights to Maven or Ant or Struts. They simply don't fit with their business model.

    Stop the FUD, folks. If you're going to make an argument that Marc wants to rule the open source world and destroy Apache, at least back it up with logical reasoning. The reality is that both organizations want to live in peace. JBoss simply wants to make sure that the peace involves the mutual protection of their legal rights.
  70. This FUD is just plain sad[ Go to top ]

    Paul,

    > Also note that JBoss isn't afraid of another open source
    > competitor.

    JBoss Group has every reason to be scared. The investors who have pumped in $10M would want a return on investment soon and not 20 years down the line. There is an amount of pressure to deliver - and fast.

    In my opinion, anything that threatens or more importantly could threaten JBoss Group's revenue and ability to show a ROI in the present or foreseeable future would be labelled a risk. And Apache's Geronimo project falls plumb under the "threats" category. It should too - to any private organization (JBoss Group in this case).

    The primary objective of all private (non-charity ;) companies/corporations is to make money. Deliver to shareholders/investors. Nothing wrong with that. Just that not many people realize it applies to the JBoss Group as well.

    -krish
    Views are my own . . .
  71. This FUD is just plain sad[ Go to top ]

    JBoss Group has every reason to be scared. The investors who have pumped in $10M would want a return on investment soon and not 20 years down the line.

    Krish, I'll speak only from my second hand knowledge of JBoss, but they have been cash flow positive for quite a while. They claim that the ten million from their VC's is still in the bank collecting interest, and that the primary purpose of the funding was as a marketing trigger to demonstrate credibility. Also realize that the VC money was in exchange for what is in essence a non-controlling interest in the company's future earnings. Sure, it's in Marc's best interest to keep them happy (as keeping them happy also keeps him as the primary shareholder happy as well as keeping open the possibility of future investment). But, at the end of the day, it is still Marc's company to run and he doesn't currently have to "answer" to anyone.
    And Apache's Geronimo project falls plumb under the "threats" category.

    I disagree for the reasons I gave above. JBoss is just as free as Geronimo. Short of features or performance figures that simply blow JBoss away, all things else staying constant, JBoss has nothing to fear from Geronimo. Sure, they may lose a portion of their developer market-share. But corporations (JBoss's money-paying customers) are not going to see any real benefits to Geronimo over JAS, and since JBoss offers real support, a good customer approval rating, and a tangible and credible corporate structure, corporations will continue support JBoss.
  72. This FUD is just plain sad[ Go to top ]

    Krish, I'll speak only from my second hand knowledge of JBoss, but they have been cash flow positive for quite a while. They claim that the ten million from their VC's is still in the bank collecting interest, and that the primary purpose of the funding was as a marketing trigger to demonstrate credibility. Also realize that the VC money was in exchange for what is in essence a non-controlling interest in the company's future earnings.

    I don't mean to spoil your view of the industry, but no VC is interested in an interest of future earnings. No VC would put $10MM into a company unless there was a reasonable chance of that company hitting $100MM in product revenue in a five-year period, and thus a viable exit strategy (preferably IPO) in three years.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  73. This FUD is just plain sad[ Go to top ]

    I don't mean to spoil your view of the industry, but no VC is interested in an interest of future earnings. No VC would put $10MM into a company unless there was a reasonable chance of that company hitting $100MM in product revenue in a five-year period, and thus a viable exit strategy (preferably IPO) in three years.
    Product revenue?? JBoss has no plans to generate product revenue, period. In fact, they can't by terms of the LGPL license. By your logic, then, no VC would ever invest in any service company, which is obviously incorrect.

    Look, neither of us has access to the agreement between the VC's and JBoss. Marc claims that the VC money was in return for a non-controlling share of the company, i.e., a non-controlling share of the future profits of the company. In essence, they then are no different than any other private investor. They have non-controlling, preferred stock. Believe me, I know that this is somewhat inconventional for VC investment in the early stages of a company's growth. I've done quite a few audits for VC companies interested in investing in other companies, and if investment took place, they almost always stipulated some degree of control over the operation of the company (often even injecting their own CEO into the mix). However, this doesn't appear to be the case here, and I can point to other cases where investment occurred without stipulated control, as well. Yes, VC's like all other types of investors are out to make money off of their investments. However, it doesn't always have to involve control over the investment, and assuming the statements from JBoss are correct, in this case, they don't.
  74. This FUD is just plain sad[ Go to top ]

    Look, neither of us has access to the agreement between the VC's and JBoss. Marc claims that the VC money was in return for a non-controlling share of the company, i.e., a non-controlling share of the future profits of the company. In essence, they then are no different than any other private investor. They have non-controlling, preferred stock.
    I'll temper this a little bit. I did notice that Matrix had placed one partner on the Board of Directors of JBoss, presumably a condition of the investment. This does somewhat refute the above, as they have a vote, so the analogy to non-controlling, preferred stock is inaccurate. We still of course don't know what impact a single board member vote might have on company direction, however...generally not much unless a majority of the rest of the board agrees.

    One other not on your "exit strategy" statement in the previous post. Yes, of course they aren't simply looking for a trickle of funds from future earnings. They want a way to sell their stake, and preferrably for a profit. Also, obviously, an IPO is a quick way of turning potential future earnings into immediate cash and is an ideal "exit strategy." The point was that unless the statements regarding the nature of the arrangement are not 100% correct (and the transparency of the single board member situation is questioned), Matrix and Accel have no way of *directly* influencing JBoss towards this end.
  75. This FUD is just plain sad[ Go to top ]

    Product revenue?? JBoss has no plans to generate product revenue, period. In fact, they can't by terms of the LGPL license. By your logic, then, no VC would ever invest in any service company, which is obviously incorrect.

    Agreed. Product revenue has a higher multiplier, though .. FWIW it would take $600-800MM in services revenue to bring the same IPO valuation.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  76. Blah Blah Blah

    I've used JBoss for quite some time now and have always considered it a high quality server. Apple ships JBoss with it's Xserve boxes, now HP OpenView will support JBoss and HP will provide services based on JBoss
    http://infoworld.com/article/04/11/12/HNhpjboss_1.html

    People bashing JBoss are a bunch of playa haters
  77. This is hilarious too:
    JBoss' Labourey said that not only was ASF's response unsatisfactory, but he was surprised that ASF had made the matter public. "We sent a letter -- it was a private letter at first and they decided to publicise it," said Labourey.

    Huh ? You mean you wanted to held in private a matter that is of interest of half of the java open source sommunity (I mean ASF, the other half being Object Web) ? Isn't it that this bullshit has made far more damage to you than any good, and you weren't expecting it? Hah. You weren't expecting this to go public isn't it ? Gee, someone in the Jboss LLC schmucks 'group' didn't think of all alternatives, huh ? :-))
  78. Ok, call me dumb but Gerenimo and JBoss are open source. What specifically is the clash in the licensing that says they can't steal each others source code?
  79. I think that JBoss, pursuing the issue with lawyers instead of around a private table sharing a beer, it's going to take walk on very thin ice (without considering who is right or who is wrong). In the, more or less, open source community it is quite easy for the users (developers) to change feelings about a product, stop advocating it and switch to another implementation (not lacking alternatives). JBoss doesn't own a strong branding like IBM, Microsoft, etc and so can't bully anyone. JBoss can't really afford to piss off developers because, notwhistanding the quality of the product, without a large number of developers supporting and advocating it, the frail reality of JBoss it's going to fade rather quickly.
  80. Isn't it just ironic that one O/S provider is suing another one? I mean really, this is just plain sad. What is the meaning of code being Open Source?

    Maybe it should follow the SNL model: Looks and Feels like Open Source, but it is not? (Tastes like butter but...)

    It just looks like the Open Source community needs to be flushed out and refreshed with people who really believe in Open Source ideals and do not hide their big corporate butts behind the "FREEDOM of CODE" slogans...

    Regards,

    Artem D. Yegorov