Discussions

News: EJB3 is not Hibernate

  1. EJB3 is not Hibernate (80 messages)

    Mike Keith has addressed the common perception that EJB3 is Hibernate in The EJB 3.0 Hibernate Fallacy.

    The perception of EJB3 as being a simple clone of Hibernate is primarily based on developer familiarity with Hibernate and a similarity of naming, as well as common purpose, and that Hibernate is morphing itself into an EJB3 implementation based on the work going into the specification, not the other way around.

    Speaking as a member of the EJB3 specification group, Mike says:
    Saying that we simply copied Hibernate would be trivializing ... time and work, especially when I know full well that the conclusions that we arrived at are in most cases either the best solution, or the best possible solution given the circumstances. The spec should look a lot like Hibernate, TopLink and JDO. If it didn't then we would not have done a very good job since the whole point of this was to make use of our experience and standardize it.

    How do you look at the relationship between Hibernate and the EJB3 persistence specification?


    *********************

    Video Tutorials on Learning Hibernate
    Hibernate Made Easy WebSite
    Recommended Books for Learning Hibernate

    Threaded Messages (80)

  2. EJB3 is not Hibernate[ Go to top ]

    Interesting, five minutes ago I've answered a question here on TSS if "Hibernate will be EJB3 compliant". So, it doesn't seem everybody heard about this. Thanks for repeating it, Mike :)
  3. There isn't any such thing in the latest Hibernate beta...
  4. So, where is the EntityManager ??[ Go to top ]

    JBoss EJB 3.0 is built on top of Hibernate 3.0 and the Hibernate Annotations project. You will find a full EJB 3.0 implementation there as it pertains to EDR2.

    When the spec finishes its work on outside-of-container APIs, an EntityManager will be available in Hibernate.

    Bill
  5. http://www.jboss.org/products/ejb3

    This is the full EJB3 implementation preview, including the EntityManager. A standalone JSR-220 persistence implementation will be released very soon, we are discussing the final details. As I said in my earlier reply in a different thread, it's not going to be a big change from your existing Hibernate Session code to EntityManager, so you probably shouldn't wait if you'd like to start now. Or use the full JBoss EJB 3.0 preview.
  6. grrrr, copy paste from Mike's blog which is a copy/paste from the hibernate forum.
    Mats,
    I answered you already, not well enough it seems. The container EM is already implemented and part of JBoss EJB3, the hibernate core functionality is compatible with the EJB3 spec, and the standalone EM will be available as soon as the spec will deliver a proposal. Believe me, wrapping the Hibernate API into the EM ones is really not the hardest part :-)
  7. OK OK, I'll be patient[ Go to top ]

    I was disappointed to see the EM missing, since it is half of EJB 3.0 entity beans (annotations being the other half). It seemed like such a simple thing to do, but I'll give you till next release to give us one. OK?
  8. OK OK, I'll be patient[ Go to top ]

    Mats,

    The JBoss guys have a point, here, that we are still hammering out the details in the group about the final API's of the EntityManager outside the Container. Once this is completely settled (and I think we are close) then you will be able to use the EntityManager in a standardized way outside the Container.

    In our Oracle EJB 3.0 preview (http://www.oracle.com/technology/ejb3) we allow getting and using the EntityManager outide the Container in a test environment, so you can try that out if you like. This was released before the standard had been settled, though, and we just wanted to let people try some of these things out to see how easy it was to use.

    -Mike
  9. POJOs ok with EJB30 and Hibernate?[ Go to top ]

    Hi all,
    While we are all waiting for EJB30/JDO, Hibernate and TopLink to shake out, I wanted to know if all these technologies will work with a POJO based business domain model?

    When I say "will work" I mean will any of these persistence technologies require me to code or package my POJOs a certain way to take advantage of their features. I believe that there is no longer a need to make my POJOs implement certain interfaces. The callback methods (if any) are outside of my business domain model.And meta-tags annotations are the only changes that I may have to make to my POJOs. Is that correct?

    I have taken care to make my business domain POJOs persistence agnostic. So the EntityManager/PersistenceManager/Session is all that knows about persistence.

    Is that the right approach?

    TIA,
    Pankaj
  10. Its exciting to see the real implementations released in February 2006 for EJB3 persistence layer. Development with WebLogic 9 and EJB3 looks very promising.
    Definitely hibernate and ejb2.1 came much before. Most of the j2ee ecommerce/ products and frameworks based on portal,rule engines, content and workflow solutions still use ejb2.1. EJB3 is still considered as a discussion point for technical architecture. While JBOSS already has EJB3 preview release last year, IBM Websphere v6 onwards may release EJB3 container soon so that developers can develop components with IBM Rational Application developer.

    Its exciting to avoid creation of home and remote interfaces with EJB3 code. EntityManager does everything and sets a great persistence development principle based on JSR220. Writing POJOS in EJB3 is quiet tricky. But you can write them. Tools help for annotations (embedding details in code) So not to spend too much time on maintainability issues. I am sure other peristence engines like Toplink will also support it more.Hibernate components and entity beans would definitely defer since their life cycle is different. When EntityManager and ejb3-persistence with JDK1.5 is around, its exciting to use jdk1.5 features in development and make use of EntityManager when you dont have home interfaces. I may write more comments in detail.

    Good Luck
    Maneesh Innani
    Senior Technical Architect
  11. I would.
    There is a radical change between EJB 2 and 3 thanks to the existance of Hibernate.
  12. There is a radical change between EJB 2 and 3 thanks to the existance of Hibernate.

    There is a LOT of difference between EJB2 and 3. But to say that it is because of Hibernate alone is misrepresentation. Those who say that have never looked at TopLink.. ever.

    I've been using TopLink since 1999. When I first looked at Hibernate in 2003, it looked a lot like TopLink to me, except perhaps Unit Of Work.
    However it was Hibernate that exposed the masses to Object Relational Mapping by being Open Source.

    Cheers.
  13. EJB, Hibernate and TopLink[ Go to top ]

    There is a radical change between EJB 2 and 3 thanks to the existance of Hibernate.
    There is a LOT of difference between EJB2 and 3. But to say that it is because of Hibernate alone is misrepresentation. Those who say that have never looked at TopLink.. ever.I've been using TopLink since 1999. When I first looked at Hibernate in 2003, it looked a lot like TopLink to me, except perhaps Unit Of Work. However it was Hibernate that exposed the masses to Object Relational Mapping by being Open Source.Cheers.

    Well, by being free and a bit more straightforward. I've heard of Toplink for years. To me, it was synonymous with ORM, but I never had the opportunity to use it. It cost too much.

    After being purchased by Oracle, I worked on a project using Oracle9iAS and, of course, you get a Toplink license with that.

    We were a bit disappointed in Toplink and after reading some early(circa 2003) Hibernate vs. JDO threads, we decided to give Hibernate a try. While Hibernate focused exclusive on POJOs(no CMP like Toplink) it seemed a bit more straightforward to use and be bit less invasive.

    Struts succeeded, IMO, because it made harder things easier and easy things were only slightly harder so it was easy to do the right thing, like no scriptlet code.

    Toplink seemed to make easy things harder while Hibernate seemed to make easy things only slightly harder. So again, instead of using say raw JDBC, it was easy to use Hibernate.

    Also, the support system for Hibernate(forums,books,etc) seem more vibrant.

    Is Hibernate perfect? Of course not. But when you add the open source and free and it has documentation at least as good as Toplink it is hard to resist.

    Both still suffer from good examples. I think that the Hibernate book could use another 50-100 pages of examples. For example, the id generator can be native, sequence, identity, hi-lo, etc...

    Ignoring what the database supports, why would you use say sequence on Oracle over native? Or some better examples of querying on sets of objects using HQL.
  14. I would.There is a radical change between EJB 2 and 3 thanks to the existance of Hibernate.
    ...and TopLink, and JDO...that's the whole point. Come on, one product does not a comprehensive specification make.

    Don't you remember infamous thread 771 so many years ago on TSS? That preceded Hibernate by a long time. I would say that was the beginning of the change, not Hibernate.

    --matthew
  15. representations[ Go to top ]

    I would like to add one level of clarification here. (This is not in response to Mike's blog, but to some comments here, and especially to some comments on the blog.)

    It seems that after a year of attacks on EJB 3 from various dark corners of the industry, from various little groups who were threated by the existence of a strong, mass market, J2EE persistence specification, we are now in a situation where everyone and their budgie is trying to claim credit for the new EJB spec. That's Just Great, it shows how successful this effort has been. I expect this process to continue, and I'm very happy about it. Broad support for the spec is Exactly what we are trying to achieve here.

    However, some of us were there a year ago, and others weren't. (Those of us who *were* have had to endure all kinds of hurtful, personal attacks upon our character and motivations. For some of us this has not been so easy to endure.)

    And, since some people are now trying to minimize the role of myself and JBoss in crafting the EJB 3.0 persistence spec, and since Mike has started a discussion about this, I would like to point out that some of the most critical features (and broad approach) of the EJB 3.0 persistence specification were proposed to the JSR-220 group by me personally, before any other ORM vendor was involved, before even Mike joined the group.

    Certainly, the EJB 3.0 specification is a collaborative effort that has involved very many well-intentioned people, all who have made deep contributions to the specification. And certainly, if Mike would have been a member of the expert group at the time, I have no doubt that he would have proposed similar things to what I proposed. And most certainly, we have all relied regularly upon Mike's wealth of experience and good sense. And most certainly, there are several things in the spec that are more TopLink-y than Hibernate-y. (Though TopLink and Hibernate are more similar than they are different.) And certainly, the spec would not have been able to navigate the political minefied we have just crossed were it not for the vocal support of Oracle.

    But I would like to claim some credit for having been involved in *initiating* this. And I hope no-one wants to be so mean-spirited as to try and take that away from me. (Of course, most credit for initiation goes to Linda.) I'd also like the Hibernate project and my team to take some credit for having *popularized* and validated this technology as a mass-market solution. The success of Hibernate was instrumental in demonstrating to Sun's J2EE group that ORM was a mainstream technology.




    Meanwhile....

    Cameron Purdy (who just loves spreading canards about JBoss) claims in comments on Mike's blog that:
    The marketing fluff from JBoss has been that EJB3 _is_ Hibernate and vice versa, and that message was crafted well over a year ago and has been pushed ever since.

    This is a plain Lie. This is most certainly NOT our message and has never been our message and Cameron cannot point to anywhere we have said this online. This lie originally came from a misrepresentation by David Jordan of a presentation given by Marc to the Raleigh JUG. David Jordan was not actually present at the meeting (!), and simply picked up a stack of Marc's slides without hearing the words to go along with them. Cameron is just repeating stuff he's heard third hand via that oh-so-reliable of sources.

    If it matters....

    The point Marc was making was that - at that early stage - we had heard a lot of people asking "Why do I need EJB3?? EJB is big and slow and bloated and ugly! Why not just use Hibernate??". His answer was "Well, this *is* just Hibernate, in essence!". Sure, knowing what we know now (ie. that people like to spread lies), I'm sure he would have better framed this as: "Well, Hibernate is POJO ORM, and this *is* just POJO ORM, just like Hibernate". But, in the context of a JUG, we expect sympathetic, well-intentioned listeners, not people like David Jordan (or people like Cameron), who want to spin and quote out of context.

    I have heard this line over and over again ever since. The number of people who have tried to pick up on this one single line and try to use it to damage the entire EJB 3.0 effort is breathtaking, and demostrates just how selfish and shortsighted some of these people are, IMO.

    So, ever since we realized how political all this stuff was going to be, we have been much more careful about how we express ourselves. I've limited my posting on TSS. I usually don't give away copies of slides. I re-read everything I write 3 times, looking for possible misrepresentations. Very difficult for me, since I am an extremely direct person. :(

    Don't mean to sound bitter, but I need to say this stuff for the record.
  16. representations[ Go to top ]

    Just in case pople took my blog that way, my blog entry was not intended to take any credit away from Gavin for his role in the spec. His personal efforts have been more than substantial and I completely agree with all of the first part of his comments that related to the spec. It happened just as he suggested, and the very fact that he was only on the EJB group a month or two before I was and was able to steer it in the POJO direction (under Linda's direction, of course) is a testament to how much Gavin is able to accomplish in such a short amount of time. If credit is being asked for then heaps of it certainly should be directed at Gavin and Hibernate for both setting stage and being one of the leading roles.

    The second part is getting a little too public and personal for me and maybe could be taken offline..

    -Mike

    -Mike
  17. representations[ Go to top ]

    oops. Double signature.
    (There really is only one of me.)
  18. representations[ Go to top ]

    Gavin,

    I don't really care either way since the Open Source Lab mostly chooses to use Object/Relational Bridge for J2EE projects. So we'll likely choose neither EJB3 nor Hibernate.

    What really annoys me is your use of almost childlike logic to call Cameron a liar:

    1) This is a plain Lie.
    2) [Marc's] answer was "Well, this *is* just Hibernate, in essence!".

    You call Cameron a liar and then quote Marc in a way that show's that Cameron's interpretation was justified. Maybe liar is a little...personal. At worst, and you state so below, he's guilty of a contextual mis-interpretation.

    You say it was taken out of context. Ok, mistake... But the whole community has consistently seen a lot of marketroid hype from JBoss and company. And all of the loud-mouth marketing speak alienates people more (both Open Source community and organizations). The BEA/IBM-killer stuff, the turfing, the now-imfamous "suck my..." comment, and just a lot of cruft like that.

    I've talked to a lot of people who've moved _off_ of JBoss lately in commercial or enterprise use not just for technical reasons, but because they're tired of the noise.

    Now, I personally think it's great that JBoss is a successful company. It's one example of how Open Source really can be an economic win. But JBoss is beginning to come across like the organ-grinder company of the Open Source world. Let's hear less talk, and more tech. You guys surely ought to be able to succeed on the merits of your services and technology.

    Jason McKerr
  19. as[ Go to top ]

    Jason, it is a plain lie that this is our marketing message. One misinterpreted comment before a small audience at one JUG, made does not make a marketing message. JBoss has a marketing department and they determine what is our message, not Cameron or David Jordan.

    I don't know what's difficult to understand about that.
  20. as[ Go to top ]

    whatever "as" means, but Gavin you have to understand that JBoss will experience "no rest for the wicked" first hand for what they've done in the past.
  21. "no rest for the wicked"[ Go to top ]

    whatever "as" means, but Gavin you have to understand that JBoss will experience "no rest for the wicked" first hand for what they've done in the past.

    Please name a "wicked" act that I or anyone else on my team has performed. Can't think of anything? OK, then speak to the hand.

    In reality, JBoss is the focus of so much attention because JBoss is so *successful*, both technically and commercially. (And because a lot of people are very, very jealous of Marc Fleury.) If JBoss stopped releasing good software, stopped making money, stopped growing, the hataz would disappear overnight. You simply wouldn't hear about JBoss anymore.
  22. "no rest for the wicked"[ Go to top ]

    Gavin, that hurts. "Speak to the hand" ?!

    Anyways you seem to be in dire need of some memory improvement pills.
    Your "what? who?" act is so ridiculous I can't be bothered to search the TSS threads.
    Just go back in your corner now, and maybe no one will kick you where it hurts, show you what you're made of (and by you I mean JBoss and not Gavin King - the always frustated JBoss employee).

    Oh and another thing. Since you're so hugely successful and all popular and dating the jocks and so this years' prom queen. I've yet to see an successful business' employee brag about how good his company does and try to turn all the shit that it's done into allegations of envy (I mean, Cheesus, are you for real?). Oh wait, I actually have. Microsoft wasn't it ?
  23. Oh boy, this is fun...[ Go to top ]

    but shouldn't all this anger/negative energy be targeted at Microsoft? But on a more serious note, it's kinda sad for me to see such smart people going at it like this.
    -Duc
  24. The Gangs of Java[ Go to top ]

    We are like local factions of a gang. Sometimes we fight amongst eachother, but when a rival gang (M$) attacks we bind back up. I guess M$ being behind on almost every one of its products, leaves us a little time to fight with eachother ;)

    I'm personally glad to see the likes of Hibernate and Toplink leading the new EJB3 spec. I'm also glad that one is open source and one is a database vendor. Not that I'm saying they should be the only ones with input, and I'm also not saying that there are not other good products/companies out there.

    The community will build and improve on EJB3 as time goes on, but I think this is a great start.
  25. "no rest for the wicked"[ Go to top ]

    I think that by "wicked" he means the unprecedented level of FUD and outright lying the JBoss/Hibernate team has done over the years, as has been proven outright in the past. Also, successfully subverting the EJB spec team to convince them to adopt a crappy and limited persistent model when a far superior and standard one already existed is the kind of thing that ruffles the feathers of serious developers and people who care about the future of the Java Enterprise market.

    It's pretty funny that you dismiss any criticism as being "because a lot of people are very, very jealous of Marc Fleury". It's really because you are a bunch of liars, and you give "open-source" a bad name through your sleazy activities.
  26. "no rest for the wicked"[ Go to top ]

    Again, name a single lie spoken by me or anyone else on the Hibernate team.

    The fact that no-one can is very, very revealing.

    (I really should know better by now than to get upset by the hataz.)
  27. "no rest for the wicked"[ Go to top ]

    Again, name a single lie spoken by me or anyone else on the Hibernate team.

    Emphasis mine, Freudian slip yours.
  28. bunch of liars?[ Go to top ]

    Rob and Radu-Adrian

    your style is going overboard IMO. This thread is in danger of becoming a disgrace to out trade (or species. Not the first one, I admit). I dont think you have a case against the people mentioned that even closely justifies this tone.

    I have used Hibernate once now, and in hindsight am still amazed that something like it can be free (beer AND freedom). I was not even harassed by JBoss salespeople that would keep asking whether I liked the download.

    They have a lot of credit with me for that.
    Christian
  29. bunch of liars?[ Go to top ]

    Christian,

    I think you don't know what you're talking about, frankly. If I'll have a spare half hour tomorrow I may want to put together say 5 links to the best sum-ups of the events related to JBoss' practices. However, being part of the "trade" you might want to educate yourself on the subject. Hint: google +jboss +astroturfing .

    In regards to the "i've used hibernate once and it's amazing it's free/credit blah" I agree. Give credit where it's due. The exchange of replies was not about that though.

    I'm still waiting on a clarifying comment from Gavin on my previous, very short post.
  30. take it easy[ Go to top ]

    Radu,

    I have been following most of what has been public on this site, and other related ones as well over the past few years. I dont think "astroturfing" deserves the death penalty or eternal damnation (assuming its a valid accusation). I dont have complaints about anything that is being said nowadays by the people in question.
    Also, AFAIR Gavin & Hibernate were not part of JBoss at that time

    chris
  31. take it easy[ Go to top ]

    Is Christian Seil an alias?

    See, that's how much astroturfing has corrupted these fora.
  32. take it easy[ Go to top ]

    I dont think "astroturfing" deserves the death penalty or eternal damnation (assuming its a valid accusation)

    This has got to be one of the saddest comments ever.
    Christian, this forum has been manipulated and poisoned for months by a schizoid organisation. And you're now questioning the very fact that it ever happened. What else never really happened ? Right, the Holocaust isn't it.
    Again, pay attention: I've never claimed Gavin or Hibernate were part of it, I've never questioned the quality of the Hibernate product.
  33. bunch of liars?[ Go to top ]

    Christian,I think you don't know what you're talking about, frankly. If I'll have a spare half hour tomorrow I may want to put together say 5 links to the best sum-ups of the events related to JBoss' practices. However, being part of the "trade" you might want to educate yourself on the subject. Hint: google +jboss +astroturfing .

    I don't understand this dig up the past and throw the baby out with the bathwater line of thinking. If think you'll find that in the natural progression of life, everyone has had moments, both public and private, that we wish we could take back either what was said or what was done.

    Life has a way of naturally weeding out those who aren't genuine in their approaches, and the good folks at JBoss aren't exempt from this.

    One of the good things that is being done by JBoss Inc., as well as others, is focusing the J2EE industry on ways to improve the J2EE standard. In some cases, they've done this by actually backing up the ideas with real implementations, ie. Hibernate. Some times they are still works in progress.

    Nevertheless, things are generally moving in the right direction, and I think we should all focus our energy on helping that process continue and spend less of our time casting dispersions around.
  34. bunch of liars?[ Go to top ]

    I don't understand this dig up the past and throw the baby out with the bathwater line of thinking.

    It's been caused by Gavin's attempt to save face for the entire JBoss organisation. His claims of innocence in this thread simply could not have gone without a proper reminder.
    That's all. It does not reflect in any way upon the Hibernate product or whatever - I'm not throwing any babies out really.
  35. bunch of liars?[ Go to top ]

    I don't understand this dig up the past and throw the baby out with the bathwater line of thinking.
    It's been caused by Gavin's attempt to save face for the entire JBoss organisation. His claims of innocence in this thread simply could not have gone without a proper reminder.That's all. It does not reflect in any way upon the Hibernate product or whatever - I'm not throwing any babies out really.

    I did not get the impression from reading Gavin's posts in this thread that he is trying to "save face for the entire JBoss organization". He was just trying to clarify *his* level of contribution to the EJB3 spec.

    I do get the impression from reading your posts that you are implying guilt by association and hence my comment about throwing the baby out with the bath water. Your comments could also be interpreted as reflecting negatively upon the Hibernate product since you are basically attacking the person who was very much so responsible for its existence.
  36. representations[ Go to top ]

    Gavin,

    Your kind of post might go some way towards clarifying things for those who are heavily invested in the politics of JDO/Hibernate/EJB3, but for the community in general, it doesn't help you.

    Sumit.
  37. representations[ Go to top ]

    But JBoss is beginning to come across like the organ-grinder company of the Open Source world. Let's hear less talk, and more tech.

    Absolutely. Remember Enhydra all those years ago, when it was the darling "BEA killer" open-source J2EE solution? And remember how it imploded so spectacularily because its own inept attempt to screw over the open-source community?

    History repeats itself yet again.
  38. representations[ Go to top ]

    But JBoss is beginning to come across like the organ-grinder company of the Open Source world. Let's hear less talk, and more tech.Absolutely. Remember Enhydra all those years ago, when it was the darling "BEA killer" open-source J2EE solution? And remember how it imploded so spectacularily because its own inept attempt to screw over the open-source community?History repeats itself yet again.

    And how exactly have we screwed over the open-source community other than plowing all of our profits into dev and documentation efforts of our own projects and others like Tomcat, mod_jk, JacORB, Eclipse J2EE tools, Hibernate, jBPM, etc...

    I think you read too many blogs.

    Bill
  39. representations[ Go to top ]

    But the whole community has consistently seen a lot of marketroid hype from JBoss and company.

    Since when have we not delivered on something we've "hyped"? Sure, it may have taken 3-6 months to meet a marketing vision, but we've delivered on everything we've promised. And how is this different than any other company or open source project? Seriously? How is this Versant announcement any different for example?

    We work hard, we are evangelists, we love our company, we believe in what we're doing, and we're extremely successful.

    Bill

    VOTE FOR JBOSS
    "...and all your wildest dreams will come true"
  40. representations[ Go to top ]

    Bill

    VOTE FOR JBOSS
    "...and all your wildest dreams will come true"

    I'm voting for JBOSS, and if this turns out to be more marketing deception, I'm going to be very angry.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Cluster your POJOs!

    p.s. it's a joke .. no need to start the name calling ..
  41. representations[ Go to top ]

    Meanwhile....Cameron Purdy (who just loves spreading canards about JBoss) claims in comments on Mike's blog that The marketing fluff from JBoss has been that EJB3 _is_ Hibernate and vice versa, and that message was crafted well over a year ago and has been pushed ever since.

    This is that truly bothers me. IMHO, the erudite & esoteric few (includes Galvin and Cameron and kinds) should *not* be browbeating out each other over things that would typically deemed as "office politics". This doesnt help! The direction of J2EE, as we see it, is being influenced by these few folks and there should be NO infighting. This doesnt foster confidence with decision makers who sink in top dollars for them J2EE products.

    What should happen in true sense if collaboration. May be Galvin could incorporate Coherence as part of the spec :) and an official channel for Tangosol to make revenues!!!
  42. representations[ Go to top ]

    ...(Those of us who *were* have had to endure all kinds of hurtful, personal attacks upon our character and motivations. For some of us this has not been so easy to endure.)

    Unfortunately most people in technology these days lack respect for other people. Just read /. and imagine working with those people!

    Another unfortunate thing is that people like myself, who see Hibernate and EJB3 (along with Spring :) as the best thing to happen to J2EE, don't post.

    So that leaves only the negative posters who if you watch them *NEVER* have anything good to say about anything. In short they're the miserable techie we all dread to work with.

    Just rise above it Galvin.
  43. Re: representations[ Go to top ]

    Gavin, I understand Cameron pissed you off, but wouldn't a simple request for specifics backing up his claims have been more effective than an STFU?
  44. Re: representations[ Go to top ]

    Gavin, I understand Cameron pissed you off, but wouldn't a simple request for specifics backing up his claims have been more effective than an STFU?

    You're right. I overreacted. It pushed a button, because I've been trying to be *so* careful not to misrepresent our involvement in and contributions to EJB3 - and I believe that this goes equally for everyone at JBoss - and yet I've been hearing this same line over and over for the past year (always from people hostile to EJB3) and I was very upset to hear it repeated by Cameron of all people.

    I should have dealt with it offline.

    If you realised how much of our time, energy and heart the EJB group are putting into trying to create a great spec for the community, you would understand why I get upset when people say such hurtful and unfair stuff about us. Seriously, I'm just dumbstruck some of the kind of stuff that people in this thread have said about us.

    I'm done with this thread now.
  45. Re: representations[ Go to top ]

    Stand tall. Thousands of people are getting value from your work. I'm constantly finding things in Hibernate that makes my job and thus my life easier.

    When I can sit at home and type this because Hibernate saves me effort, who cares what the naysayers think.
  46. Don't hate the playa...hate the game![ Go to top ]

    Agreed...if it wasn't for the progressive thinking of folks like Gavin, Rod, Craig, [insert your favorite thought leader here], who knows where the state of J2EE would be by know. These folks contributions have made my and many others' lives much easier and their efforts should be recognized.

    If Top Link was open source, maybe we would all be singing their praises and (inadvertently) projecting EJB3 as Top Link. But the fact of the matter it is not and so credit is well-deserved for those who have made ORM technologies **accessible** to the masses. This is not an easy feat in the fickle world of Java. As peers, you should be glad we are finally moving EJB towards a sensible direction.

    That said, anyone with half a brain and with some knowledge of the history of ORM tools can make the realization that Top Link (et al) are definite players and that their impact on ORM has been a positive, albeit to a smaller user community.

    Ultimately, I wish all the personal attacks and infighting would subside and lets embrace what Linda has set out to achieve with EJB 3.0. Now everyone gather around the campfire and sing with me, "Kumbaya my lord, kumbaya..." (sorry, if this seems like an American/Canadian thing...or maybe it isn't).

    -Lou

    p.s. Does anyone remember those MTV Celebrity Death Match series with claymation?? Somebody has got to make up one between Gavin and Cameron...I wonder what the finishing move would be?? ;)
  47. The Game: Vendor Screw the Customer[ Go to top ]

    Agreed...if it wasn't for the progressive thinking of folks like Gavin, Rod, Craig, [insert your favorite thought leader here], who knows where the state of J2EE would be by know.

    Maybe we'd be using the non-broken parts EJB2.1 with JDO/Hibernate/Toplink as Persistence and be producing quality applications - wait a minute, that's what we're doing already.

    EJB (for me) is transaction control, messaging & security in middleware. Morphing it into an O/R Mapper and adding Annotations to streamline development is simply packing features that are available in existing components that can be integrated in existing EJB containers without any problems.

    Putting all the eggs in one basket will make for nice consulting revenues and sell lots of books for those who have 'advanced' the spec.

    For some, progress means throwing out the old, working, accepted specifications every few years. For those who have to invest in technology/platforms, it's a bit of a headache and I can understand when people see this as vendor-driven and not in the interest of the consumers.
  48. Re: representations[ Go to top ]

    Gavin and others.

    Hibernate is way better than Toplink because it's free.
    Tomcat is free as well.
    EJB3 will be free (AND STANDARD I HOPE - all implementations to spec)
    JBoss is free.

    To all involved many thanks. A great job. Hopefully a job that is soon complete, and you will leave the specs alone and stop churning this problem/solution with yet another persistence solution.

    I will choose Tomcat with EJB3 and native SQL behind DAO. Nicely layered so the bis objects know nothing of ORM/DB's. I'll even have custom sql optimised for reporting when needed.

    I really really hope that everyone involved in EJB3 has a huge party and then stops working on it entirely.

    Until the JDO guys rock the boat because they have ego's too.

    Jonathan
  49. Re: representations[ Go to top ]

    Hibernate is way better than Toplink because it's free.

    You win the prize, Jonathan. Of all of the things that I have read on this thread this is by far the most memorable. I hope, for the sake of your company's future, that no software procurement decisions are left to you.

    PS You forgot to mention how HSQL is better than Oracle.
  50. Re: representations[ Go to top ]

    I wanted to reply to this yesterday, but I didn't feel like I could address anything substantive without fanning the flames, which were (at the time) doing quite well by themselves.
    I've been hearing this same line over and over for the past year (always from people hostile to EJB3) and I was very upset to hear it repeated by Cameron of all people.

    It is my own opinion (as opposed to being an echo) and it is not an opinion that is hostile to EJB3 nor to Hibernate.

    There was, IMHO, a market positioning effort starting at (or before) last year's TSS Symposium to make the term EJB3 (particularly w.r.t. EJB3 persistence) synonomous with JBoss and Hibernate. I sat through several presentations, including Marc's, and was absolutely convinced of that being the goal. (One could argue that I could be "absolutely convinced of that being the goal" without it actually being a goal. That's why we call it an opinion. Note that I allude to no "unnamed sources" .. just my own presence in those presentations.)

    I perceived it to be a marketing maneuver, and a shrewd one at that, but I did not approve of it. Hibernate is a very successful project on its own merits. JBoss and EJB are also both already successful. I didn't see any good that could come from the positioning, particularly since standards and spec are about cooperation, and implementations are about competition. Frankly, I'm quite surprised that my opinion on the topic has garnered any attention at all (other than the fact that I called it "marketing fluff"), considering that it's a marketing dream to get a product associated so closely with a concept. Look at Redhat = Linux, Windows(tm) = windows, SQL Server(tm) = SQL server, etc.

    My concern was that the attempt to position an implementation and the standard as being the same thing would only cause resentment. And while it may be entirely innocent, claims like "JBoss Inc. is pleased to announce the Preview 3 release of EJB 3.0" and "I'm proud to announce the EJB 3.0 Preview Release of the early draft of the EJB 3.0 specification" are guaranteed to piss off anyone else who contributes to the EJB3 work and/or has a product that implements the EJB specs. (Study those exact quotes very carefully, and tell me in all honesty that they don't say what they're saying. Or, for a slightly simpler exercise, prove that 1=0.)

    At any rate, my comment was solely about marketing, not about the spec or anything technical. I am not on the EJB3 EG, so I cannot speak to any of those aspects of the discussion. From what Gavin has told me, he has poured himself into the EJB3 effort, and I believe him. I regret that this conversation has detracted from the technical work being done, and from the efforts of the EG.
    Seriously, I'm just dumbstruck some of the kind of stuff that people in this thread have said about us.

    I'm disappointed by the tone of some of the comments. People can disagree (and even disagree strongly) and still be respectful.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Cluster your POJOs!
  51. what is your problem?[ Go to top ]

    And while it may be entirely innocent, claims like "JBoss Inc. is pleased to announce the Preview 3 release of EJB 3.0" and "I'm proud to announce the EJB 3.0 Preview Release of the early draft of the EJB 3.0 specification" are guaranteed to piss off anyone else who contributes to the EJB3 work and/or has a product that implements the EJB specs. (Study those exact quotes very carefully, and tell me in all honesty that they don't say what they're saying. Or, for a slightly simpler exercise, prove that 1=0.)

    Really Cameron, are you really that full of it? I really shouldn't be surprised though. It is knit-pick "analysis" like this that really shows your true colors. For the record, I drafted these announcements in about 30 seconds at 2 am after working my ass off to get out those releases. It's posts like this from you that further solidifies my opinion that you have an agenda against us. I don't think it is a personal agenda. Maybe you do this because you find that smearing us creates attention for yourself and for your company. Or maybe its because JBoss Cache is hurting your bottom line. Or maybe you're trying to look cool for your blog friends. I don't know...

    But thanks though for bringing this up. I'll be more careful next time when I craft an announcement so others have a harder time of twisting my words for their own agenda.

    Bill
  52. what is your problem?[ Go to top ]

    For the record, I drafted these announcements in about 30 seconds at 2 am after working my ass off to get out those releases.
    So you are basically saying you shouldn't have posted them in the first place, thus making Cameron's point.

    Why are you accusing him of twisting your words since you are admitting yourself it was a mistake to post them?

    Bottom line:

    - Never post something at 2am
    - Have somebody else review all PR announcements

    --
    Cedric
  53. what is your problem?[ Go to top ]

    Bill, you're an idiot. Shut up and let the adults talk.

    (Credit.)
  54. what is your problem?[ Go to top ]

    Bill, you're an idiot


    Idiot is a compliment in this neck of the woods. So I guess I should say thank you.
  55. what is YOUR problem?[ Go to top ]

    For the record, I drafted these announcements in about 30 seconds at 2 am after working my ass off to get out those releases. It's posts like this from you that further solidifies my opinion that you have an agenda against us. I don't think it is a personal agenda. Maybe you do this because you find that smearing us creates attention for yourself and for your company. Or maybe its because JBoss Cache is hurting your bottom line. Or maybe you're trying to look cool for your blog friends. I don't know...But thanks though for bringing this up. I'll be more careful next time when I craft an announcement so others have a harder time of twisting my words for their own agenda.Bill

    I find it amazing that the company who feels they are "professional" open source (heck one might say the coined the term?) can


    1) Post releases they spent a whopping 30 seconds drafting.
    2) Claim that all their marketing messages come from the marketing department, then say they release PR's with 30 seconds of work and no review
    3) Open the door by asking for proof then getting personal when someone tries to respond with some and
    4) Take on this kind of simply childish banter

    You may be upset at Cameron's comments but his posts have be professional, to the point, and reasonably unemotional. In short they have been "professional".

    I cannot believe the tone of all the JBOSS "executive" posters. Professional isn't close to the words I would choose.

    Dave Wolf
    Cynergy Systems
  56. what is YOUR problem?[ Go to top ]

    You may be upset at Cameron's comments but his posts have be professional, to the point, and reasonably unemotional. In short they have been "professional".I cannot believe the tone of all the JBOSS "executive" posters. Professional isn't close to the words I would choose.
    Since this behaviour is totally consistent with past behaviour (even though it is, obviously, not internally consistent), why be amazed? Isn't this exactly what they are counting on: that everyone have guppy memories and after a round in the proverbial waterglass (let's say, a week or so in forum-time) everyone will be "shocked" and "amazed" by any "new" behaviour, even though the exact same pattern was witnessed not long ago.

    If anything, I am amazed by the amazement.

    There is no point in trying to answer Gavins semi-rhethorical-questions-with-obvious-answers, and no point in pointing out Bills constantly hurt ego, at least not for the sake of having a reasonable discussion. You might as well be talking to a wall, and even that might be more useful because then it will at least be painfully clear that you're wasting time.
  57. Re: representations[ Go to top ]

    And while it may be entirely innocent, claims like "JBoss Inc. is pleased to announce the Preview 3 release of EJB 3.0" and "I'm proud to announce the EJB 3.0 Preview Release of the early draft of the EJB 3.0 specification" are guaranteed to piss off anyone else who contributes to the EJB3 work and/or has a product that implements the EJB specs. (Study those exact quotes very carefully, and tell me in all honesty that they don't say what they're saying. Or, for a slightly simpler exercise, prove that 1=0.)

    You sound the same way my wife does sometimes...trying to real malice into what could seemingly be an innocent comment.

    I think most people read the headlines, and then read the meat of the blog, and realize that Bill is talking about JBoss's implementation of the current EJB 3.0 spec. Couple that with the fact that he posts the exact same articles to JBoss's blog site, you can see exactly what is intent is.
  58. *damn!*[ Go to top ]

    You sound the same way my wife does sometimes.

    Damn -- this has been a pretty heated exchange, but I'd say that this takes the cake for the most brutal comment. /me hopes that your wife doesn't read tss!

    -Patrick

    --
    Patrick Linskey
    http://solarmetric.com
  59. Re: representations[ Go to top ]

    Does your wife read tss?
  60. Ladies and gentlemen.

    I have not posted on TSS before, so if I'm not following the correct ettiquette (i.e. there does not appear to be enough hate/offense in my message), then I apologise in advance. I have been reading the discussion threads that appear on TSS and related sites (JRoller, etc.) and a couple of things really stand out.

    Firstly, the emotion.
    I'd like to say that this is good. You cannot dedicate yourself to something without a bit of passion. It sustains your motivation to excel, to deliver product that the world cannot live without. Wonderful stuff, please keep it going since we (the rest of the world) NEED your output. In many cases, we depend upon the products you create and champion.

    Secondly, the talent.
    Man, there are some clever people here. I look at what you lot produce and think "these guys are really clued in". No buttering up is needed, you *know* what you can do. I'm just standing on the sidelines.

    Thirdly, the lack of politeness.
    I see these talented people using their clever minds to skewer one another, sometimes on an hourly basis. They just don't appear to get along, even when they are saying almost the same thing. They *leap* to show the small difference between opinions that makes them feel superior.

    Most of you would qualify for the title of "Software Architect" and you really know how to do that. Architects are united by two things:

    1) they each really want to do it right.
    2) they each know better than everyone else how it should be done.

    These are not bad things, ok? Step back and see that Architects provide leadership to the rest of the software world. Sometimes you are facing a room full of hostile business people and you have to get the budget out of them to do it right. sometimes you have to corral a group of developers and get them to move in *one* direction. Leadership is *required*. It's part of the Job Spec.

    But leadership requires more skills than just shouting. You have to persuade. You have to understand, empathise, discuss and sometimes, ignore.

    If you want to get rid of someone in an organisation, you fire them, right? That's not leadership, that's management. If you have to discipline a worker, that's management. When you are fighting colleagues for political gain, that's management. When you use authority to control someone's approach to their work, that's management again.

    Question - why are you guys trying to manage each other? With so many independent approaches and so many innovative thinkers, do you feel that a show of strength will work?

    Watching you as a group is like watching a championship soccer team (football, for other parts of the world :) stealing the ball off one another and trying to do tricks in the middle of the pitch. It may be fun, but it's not the game we are here to play.

    I see some of you reaching out for recognition. In fact most of you do this and your passion for "rightness" is worn openly. This can be good and bad. You can be *too* eager for that recognition and trip over the rest of us on the way. :) You are already recognised as Technical leaders - look at the input you have on the way that the rest of us work!

    Please don't take this the wrong way, but as someone who hires people like yourselves into projects, I'd have to say that there are members of this community that I would not engage professionally. Certainly not because you are bad at your work, oh no! You obviously have excellent technical skills. It's because I doubt your ability to engage with those who differ with you. Those who don't have your grasp of the big picture.

    Those very people, in fact, who need your leadership.

    I'd like to point out your similarity as a group to the scientific community. They also scramble for recognition, but they are much more polite about it. Draw your own parallels, there.

    I've got more to say, but this is turning into a rant and I really don't like that thought. Especially here and now.

    Let me finish by being all touchy-feely and asking you to please think about your strengths and weaknesses as a group. Play to those and before posting that incendiary bomb wrapped in a "reply-to", think about whether it actually contributes to your reputation, your work and the longer-term Java/J2EE group goals.

    Give it a shot. Be polite.

       Ivan McCann
  61. Ladies and gentlemen.I have not posted on TSS before.

    same here.
    Give it a shot. Be polite.   Ivan McCann

    Thank you Ivan. Some times I find my self reading these type of threads and it's sort of passing a road accident. You don't want to look but you can't help it.

    Maybe I'll start ignoring the serverside again :-(
  62. Are there really people out there that believe the EJB3.0 will throw naway everything that works well and implement only the part that doesn't work (Entity Beans) - get real!

    Sounds like SOMEONE needs to get Hibernate back on top of the headlines since Versant/JDO is going Open Source and attracting attention.
  63. As we progressed and began specifying the persistence layer it was obvious that we had similar features at the 80-90% functionality level that EJB is trying to achieve

    At a theoretical level and conceptual level. Agreed.

    But in terms of TDD , Easy of use , IDE intigration , books and RAMP up time for a developer all the persistance technologies you mentioned vary a lot.
  64. ORM or Appservers[ Go to top ]

    I am not really sure what this thread is about? EJB3.0 ala JSR220 is a spec which implies there will be many implementations.

    This technology (ORM - POJO - JSR220 - etc ) is destined to become a commodity item that everyone will be using instead of the JDBC that developers have had to slug things out with for years now. It is high time the Java community has come to address this issue.

    Seems to me that the ORM discussions these days are all wrapped up into the political wrangling of the application server companies trying to grab their markets. Is this really about Hibernate and Toplink as EJB or is it about JBoss and Oracle making sure they can get their fair share of the application server market?

    Well, Versant has stated our position. This is an important standard emerging and we want to make sure it happens in a public way. Of course, we want to make a little money on the services around this too...families to feed you know ;-)

    So, JBoss\Hibernate & Oracle\Toplink wrangling for the billions in the application server markets ....who cares lets just get the spec done so people can start using it. I think the Eclipse JSR220-ORM project is a great place to get the community building an implementation of this specification. It will GIVE a core runtime to EVERYONE who is not using an application server and it will generate meta data that can work with EVERYONE's application server runtimes.

    Go get it and get involved...

    http://www.versant.com/opensource/download/en-us

    Otherwise, kick back an relax, it is coming to Eclipse.

    Robert Greene
    Versant Corporation
  65. ORM or Appservers[ Go to top ]

    I am not really sure what this thread is about?

    Maybe you're a little, uh, "Greene" then... LOL
    It will GIVE a core runtime to EVERYONE who is not using an application server and it will generate meta data that can work with EVERYONE's application server runtimes.

    Uh, Hibernate and TopLink have always run in any app server, IDE and toolset. TopLink for a decade now. This continues in the EJB3.0 preview too. What's new? What's unique about what you're proposing? It just smacks of a desparate "me too".

     - Don
  66. EJB3 is not Hibernate[ Go to top ]

    Without getting into the politics of EJB3, Hibernate could have achieved the same status (in Java ORM) as Struts has achieved in the Web presentation layer (that is somthing, JSF, even as a standard, will probably never be able to achieve).
  67. Respect goes a long way[ Go to top ]

    For those about to rock <fire> we salute you!


    Seriously, I agree with one of the other posters about showing respect to our peers when it comes to discussing technology.

    I work at one of the larget global financial audit companies in the world and we use Hibernate in all of our J2EE apps.

    I also own a copy of "Hibernate in Action"

    Gavin - I understand your passion for the work you have performed, which is outstanding - but I was shocked by the way you handled yourself on this thread.....I have a lot of respect for you (even though I have never met you), so please do not lower yourself to the bottom feeders trying to attack you.


    Best Regards,
    Tom Pridham
    Technologist
    Tom@CoastalSoftwareSolutions.com
  68. Because if Sun and Toplink were the driving force behind
    ejb3.0 rather than Hibernate and Spring I wouldn't touch
    it with a bargepole ......
  69. Lucky you told me in time ....[ Go to top ]

    Because if Sun and Toplink were the driving force behind ejb3.0 rather than Hibernate and Spring I wouldn't touch it with a bargepole ......

    Bryan, you seem to have gotten lost somewhere in this thread. Sun and TopLink and Hibernate are driving forces behind EJB 3.0. Apart from the personal stuff this is the whole point. Spring is not represented on the expert group.

    Anyway, the expectation is you and others will choose what your bargepole touches based upon the specification itself and not what influenced it. I hope that the fact that you now know these things won't affect how you plan on poling your barge.
  70. Lucky you told me in time ....[ Go to top ]

    Because if Sun and Toplink were the driving force behind ejb3.0 rather than Hibernate and Spring I wouldn't touch it with a bargepole ......
    Bryan, you seem to have gotten lost somewhere in this thread. Sun and TopLink and Hibernate are driving forces behind EJB 3.0. Apart from the personal stuff this is the whole point. Spring is not represented on the expert group. Anyway, the expectation is you and others will choose what your bargepole touches based upon the specification itself and not what influenced it. I hope that the fact that you now know these things won't affect how you plan on poling your barge.

    I'm more talking about the way that it took spring and hibernate to popularise pojo's and inversion of control before these other guys take notice , co-opt it and then act as though it was their idea from the start.

    I have read the specification and I have to say I like it. I'm going to be using EJB3 for development as soon as possible ( on JBoss ).

    When I talk about the "driving force behind the specification" I'm not refering to whoever gets their names on the credits list, I'm taking about the people who Sun have decided "if you can't beat them ... join them".

    --b
  71. Lucky you told me in time ....[ Go to top ]

    I'm more talking about the way that it took spring and hibernate to popularise pojo's and inversion of control before these other guys take notice , co-opt it and then act as though it was their idea from the start.

    Spring and Hibernate have certainly helped popularise these technologies, but these ideas have been around for a very long time. For example, TopLink celebrated 10 years of providing Plain Old Object Persistence last year (I did not say poJo, as they did Smalltalk too). The ideas of Inversion of Control can probably be traced back to patterns used by Smalltalk and C++ developers in the early 90s, or perhaps earlier.
  72. Lucky you told me in time ....[ Go to top ]

    I'm more talking about the way that it took spring and hibernate to popularise pojo's and inversion of control before these other guys take notice , co-opt it and then act as though it was their idea from the start.
    Spring and Hibernate have certainly helped popularise these technologies, but these ideas have been around for a very long time. For example, TopLink celebrated 10 years of providing Plain Old Object Persistence last year (I did not say poJo, as they did Smalltalk too). The ideas of Inversion of Control can probably be traced back to patterns used by Smalltalk and C++ developers in the early 90s, or perhaps earlier.

    Well Steve, I agree 100% with you here. Also it wasn't Sun or Oracle who invented these techniques.

    However Toplink may have been providing these techniques for the last 10 years or not ( I've never used their product because I found it too much of a pain to setup when I looked at it a couple of years ago ).

    What for me really differenciates Hibernate/EJB3.0 entity
    beans is the use of dynamic bytecode generation to remove the need to implement 101 interfaces and also the fact
    that you can test the persistance subsystem without having to do it within a container with it's associated slow compile/generated descriptors/deploy ( slowest part )/test/repeat cycle...
    to run/test it in the container.
  73. Lucky you told me in time ....[ Go to top ]

    Well Steve, I agree 100% with you here. Also it wasn't Sun or Oracle who invented these techniques.

    Indeed.
    However Toplink may have been providing these techniques for the last 10 years or not ( I've never used their product because I found it too much of a pain to setup when I looked at it a couple of years ago ).
    What for me really differenciates Hibernate/EJB3.0 entity beans is the use of dynamic bytecode generation to remove the need to implement 101 interfaces and also the fact that you can test the persistance subsystem without having to do it within a container with it's associated slow compile/generated descriptors/deploy ( slowest part )/test/repeat cycle...to run/test it in the container.

    Even that is still not a recent thing. I was using this kind of rapid development and testing of plain object persistence nearly 10 years ago with Gemstone in Smalltalk. (And, of course, JDO now provides the same freedom as well). There seems to be an (understandable) tendency to assume that all features of popular and high-quality products like Hibernate and Spring are innovative and new, whereas many of the ideas have been around for ages. This is not to say that the developers of these products have necessarily copied any existing products; principles like IoC and plain object persistence are so useful that they tend to be often re-invented. It can be hard to trace the first occurrence.
  74. Lucky you told me in time ....[ Go to top ]

    Well Steve, I agree 100% with you here. Also it wasn't Sun or Oracle who invented these techniques.
    Indeed.
    However Toplink may have been providing these techniques for the last 10 years or not ( I've never used their product because I found it too much of a pain to setup when I looked at it a couple of years ago ).What for me really differenciates Hibernate/EJB3.0 entity beans is the use of dynamic bytecode generation to remove the need to implement 101 interfaces and also the fact that you can test the persistance subsystem without having to do it within a container with it's associated slow compile/generated descriptors/deploy ( slowest part )/test/repeat cycle...to run/test it in the container.
    Even that is still not a recent thing. I was using this kind of rapid development and testing of plain object persistence nearly 10 years ago with Gemstone in Smalltalk. (And, of course, JDO now provides the same freedom as well). There seems to be an (understandable) tendency to assume that all features of popular and high-quality products like Hibernate and Spring are innovative and new, whereas many of the ideas have been around for ages. This is not to say that the developers of these products have necessarily copied any existing products; principles like IoC and plain object persistence are so useful that they tend to be often re-invented. It can be hard to trace the first occurrence.

    Ok sensai, I bow before your superiour knowledge. But let me
    ask another question. For what they are .. do springframework and hibernate rock or what ?
  75. Lucky you told me in time ....[ Go to top ]

    Ok sensai, I bow before your superiour knowledge. But let meask another question. For what they are .. do springframework and hibernate rock or what ?

    I'm sure Hibernate does, but I'm not a user - I have more persistence requirements than just relational. However, I think Spring is one of the neatest products I have ever used for development, and I have been developing for decades. It is rare to find something so simple to use and yet so powerful.
  76. EJB3 is a fashion[ Go to top ]

    What I dislike in EJB3 is a fake "POJO/POJI" concepts this spec try to push.
    EJB3 annotated POJO is not a POJO in sense of Spring Framework or other IoC-based containers. Your EJB3 code becomes dependent on EJB3 interfaces (I mean annotations), this is almost the same as exposing dependencies on interfaces directly. In general EJB3 POJOs classes will require annotation specific interfaces to be on classpath - this violates the main principle of POJO.

    Best Regards,
    Theodore Kupolov
  77. EJB3 is a fashion[ Go to top ]

    There is no need to use annotations in any way, the old "XML Configuration" way will still be available, also in EJB3
  78. EJB3 is a fashion[ Go to top ]

    There is no need to use annotations in any way, the old "XML Configuration" way will still be available, also in EJB3

    Here is a link to a simple EJB 3.0 session bean sample that uses XML deployment descriptor with dependency injection, call back listeners, etc

    HTML

    Zip

    regards
    Debu Panda
    http://radio.weblogs.com/0135826/
  79. EJB3 is not Hibernate[ Go to top ]

    I had the impression that Hibernate sort of is EJB3. I do not know where that impression came from, who said it, or where it was said, but somehow that message got through to me in that form.

    Most of the time when I hear about EJB3, it comes from JBoss people. It is certainly the overall impression that EJB3 is sold with the success of Hibernate - at least by JBoss people. Therefore it is not a surprise that the mental connection is made between Hibernate and EJB3. If there is some other message that JBoss has tried to cultivate, it has grown out of their control a long time ago.
  80. EJB3 is not Hibernate[ Go to top ]

    I had the impression that Hibernate sort of is EJB3. I do not know where that impression came from, who said it, or where it was said, but somehow that message got through to me in that form.Most of the time when I hear about EJB3, it comes from JBoss people. It is certainly the overall impression that EJB3 is sold with the success of Hibernate - at least by JBoss people. Therefore it is not a surprise that the mental connection is made between Hibernate and EJB3. If there is some other message that JBoss has tried to cultivate, it has grown out of their control a long time ago.

    It's true, then only reason I gave ejb3 a second thought is
    that I was looking for a sun endorsed implementation of the
    hibernate "standard".
  81. can any body tell me how to implement busiess process in hibernate without using the session bean. Current i have created a class which calls presistent layer. but i think there should be some other solution in hibernate for that.