Discussions

News: IBM withdrawing Websphere 5.1 as a Software Product

  1. WebSphere World has reminded users that IBM is withdrawing WebSphere 5.1 as a sold product in late February 2007, with technical support ending likely in February 2008. This was announced back in June 2006; potential users should be looking at WebSphere 6 instead by now, and existing users should consider upgrading. Again, this does not mean support for WebSphere 5.1 is going away yet! While it's something to expect in the future (repeat: probably February 2008), it's not a critical issue yet. All this reminder means is that the writing is on the wall for Websphere 5.1: it's no longer being marketed.

    Threaded Messages (28)

  2. What is the bid deal about it.[ Go to top ]

    Who really cares about it in the era of open source...
  3. Re: What is the bid deal about it.[ Go to top ]

    Who really cares about it in the era of open source...
    Perhaps the many large companies that use WebSphere? Peace, Cameron Purdy Tangosol Coherence: The Java Data Grid
  4. Re: What is the bid deal about it.[ Go to top ]

    Who really cares about it in the era of open source...
    Some unimportant companies with really big money who will pay a lot just for taking some risks of their back. Apart from that the annoucement can be safely ignored. :-) http://www.enterpriseware.co.uk
  5. Re: What is the big deal about it.[ Go to top ]

    Some unimportant companies with really big money who will pay a lot just for taking some risks of their back.
    Apart from that the annoucement can be safely ignored. :-)

    http://www.enterpriseware.co.uk
    And it's this kind of attitude that's the reason why open source doesn't gain more traction in enterprises.
    Open source for open source's sake doesn't get an enterprise anything. Instead of banging the "open source! open source!" drum, the open source community should be talking about enterprise ready features such as clustering/failover, high availability, etc.
  6. Re: What is the big deal about it.[ Go to top ]

    Just for a clarification: I was being ironic. http://www.enterpriseware.co.uk
  7. IBM is a big dustbin[ Go to top ]

    With Every new release, they create new adapater to provide compatibility to the old release...so they create a big dust bin....some day this product would be busted with all dust. People should move to websphere community edition that is newly written and built on geronimo....thats an ideal choice i feel... I suggest people to go for glassfish or geronimo or weblogic...
  8. Re: IBM is a big dustbin[ Go to top ]

    I heard that unlike Websphere 4 and 5.0, Websphere 6 was quite a good J2EE/JEE platform. Can you please develop your point of view? (article, comparison, stories of personal past projects...) (I understand your concern about the "big dustbin", but you don't have to use this part of the platform if you begin a new project - this is about your suggestion to use glassfish, geronimo or weblogic rather than Websphere 6) Cyril
  9. Re: IBM is a big dustbin[ Go to top ]

    I heard that unlike Websphere 4 and 5.0, Websphere 6 was quite a good J2EE/JEE platform.
    I kind of have to disagree with your comment on 4 and 5: These websphere versions were quite horrible as development platforms (5 was better than 4 though).. But.. In production, they are rock solid! And that is worth quite a lot to a lot of organizations. I've used other app servers that where really nice to develop on, but fell over every once in a while because they where just not robust enough, so I'd say Websphere, even in the past has a lot going for it. (PS: I'm not an IBM employee, but I have had years of exposure to Websphere, which may have tinted my view).
  10. Re: IBM is a big dustbin[ Go to top ]

    I'd have to disagree with you. I've used WAS in production since version 3.5 and we currently have 5.1 in production now with a migration plan to 6.x. I've never been happy with either the development or production experience with WAS, but alas, it is forced down our throats in such a way that we end up paying in multiple ways. I would even suggest a strategy that some take, which is to dev against tomcat and deploy to WAS (if you've stripped out the ejb stuff). There's nothing like not having to have a laptop tricked out with 2 gigs just to run a rational branded outdated version of eclipse and then the WTE on top of that. So far I've been through 3 major global companies and 3 major versions of WAS and its been like in law you can't get away from.
  11. Re: IBM is a big dustbin[ Go to top ]

    I'd have to disagree with you. I've used WAS in production since version 3.5 and we currently have 5.1 in production now with a migration plan to 6.x. I've never been happy with either the development or production experience with WAS, but alas, it is forced down our throats in such a way that we end up paying in multiple ways. I would even suggest a strategy that some take, which is to dev against tomcat and deploy to WAS (if you've stripped out the ejb stuff).
    +1, even if you have to be very careful because Tomcat is much more configurable and flexible than WS (think at the classloader structure of Tomcat and compare it with that of WS 5), so expect a certain amount of work for the first deployment the first time.
    There's nothing like not having to have a laptop tricked out with 2 gigs just to run a rational branded outdated version of eclipse and then the WTE on top of that. So far I've been through 3 major global companies and 3 major versions of WAS and its been like in law you can't get away from.
    From my experience none USING WS was happy there either. Guido
  12. Re: IBM is a big dustbin[ Go to top ]

    I would even suggest a strategy that some take, which is to dev against tomcat and deploy to WAS (if you've stripped out the ejb stuff). There's nothing like not having to have a laptop tricked out with 2 gigs just to run a rational branded outdated version of eclipse and then the WTE on top of that.
    Are you kidding? Have used both MyEclipse/Tomcat and WSAD in the past, I have found the development experiences are very similar (though WSAD was on Eclipse 2). The WTE embedded in WSAD is actually quicker than the MyEclipse/Tomcat combination in a single Web user (i.e. you, the developer) scenario, as well as quicker to start/stop/restart. What a pity Rational fucked up WSAD6 (aka RAD6)!
  13. Re: IBM is a big dustbin[ Go to top ]

    Also, in WSAD5 when you publish a Web project (not a EAR project) to a server, you can choose to publish to Tomcat or WTE4 or WTE5. I havn't tried Tomcat in WSAD myself, but shouldn't be much different from WTE.
  14. Re: IBM is a big dustbin[ Go to top ]

    no I'm not kidding ;-). Anything based off of eclipse 3.x is going to be faster than the eclipse 2.x that WSAD was based on. I can get full debugging trace to the console and jboss booted with embedded tomcat in 1/4 the time to boot the WTE in either WSAD or RAD/RSA. After all, the WTE for both 5.1 and 6.x is just a non-federated instance of WAS. That the experience is similar speaks to how the pricing of MyEclipse is 0.5% the pricing of RSA, and of course eclipse + WTP is ZERO.
  15. Re: IBM is a big dustbin[ Go to top ]

    I heard that unlike Websphere 4 and 5.0, Websphere 6 was quite a good J2EE/JEE platform.
    I kind of have to disagree with your comment on 4 and 5: These websphere versions were quite horrible as development platforms (5 was better than 4 though).. But.. In production, they are rock solid! And that is worth quite a lot to a lot of organizations.
    I find this differentiation pretty important as well: production platforms and development platforms are very different affairs. Production J2EE servers are often treated as system software, just like databases and operating systems are, which means that they're upgraded as conservatively as the latter. Just like Oracle 9i is still pretty common, so is WebSphere 5... simply because they're trusted production platforms, integrated with systems monitoring etc. Makes perfect sense from a system admin perspective. FWIW, this is the main reason why Spring 2.0 still supports JDK 1.3 and 1.4, and why even Spring 2.1 and later 2.x versions will still support JDK 1.4 (for years to come). We often find WebSphere users developing with Spring on Tomcat or Jetty and then deploying to WebSphere for integration tests (and of course for production), with WAS 5.1 and 6.0 being the common target platforms at present. A similarly common production platform is WebLogic Server 8.1. (All of the latter being based on JDK 1.4...) Juergen (Spring Team)
  16. Re: IBM is a big dustbin[ Go to top ]

    I heard that unlike Websphere 4 and 5.0, Websphere 6 was quite a good J2EE/JEE platform.


    I kind of have to disagree with your comment on 4 and 5:
    These websphere versions were quite horrible as development platforms (5 was better than 4 though).. But.. In production, they are rock solid! And that is worth quite a lot to a lot of organizations.
    They are not that horrible as development platforms if you have used WSAD.
  17. Re: IBM is a big dustbin[ Go to top ]

    Not to be contentious, but WSAD? WSAD was based on the eclipse 2.x core. RAD/RSA 6 on the older version of eclipse. We are currently staying off of RSA 7 due to the memory footprint/install space profile. I guess what I am trying to say is that for some of us who live in the big corporate world with governance boards ramming this stuff down our throats, its an uphill battle to be productive. I would say that for the money that is paid, "not that bad" is not good enough. The dev experience of using MyEclipse to deploy/debug to WAS is much better than in the rational products IMO. I personally wish that somehow vendors would finally get it and get leaner in their products. Unfortunately, the only way I think this will ever happen is if they start losing revenue off of it, and for the risk averse, its much easier to blame a vendor. Here is an example of how some of this works: a company has a hardware/software agreement with a big blue company. The developers are restricted to ordering a certain z-unintelligent semi-pro workstation that uses rdimms (not the cheap stuff), and the cost is over $6k for something you can get at walmart or compusa for $1200 and a copy of world of warcraft. You can only trick this lame puppy out 1 gig RAM, 2 if your lucky. You strip down the os to the bare bones, boot up their branded IDE, do some coding, watch the memory profile go over 600 megs, start debugging against the app server which adds another 250+ megs of ram, crack an egg on top of the case and warm your happy feet to the tantric sounds of the hard drive platters spinning around. You are then told that you have to use it, b/c the hardware credits are then used in a rebate type of form against more hardware or software, but then the software requires more hardware, and then hardware they sell you gets a sunset schedule 3 months after they are delivered to your data center. But then you go into the bureaucratic maze and your told that its too difficult to fight b/c they are so entrenched and too far gone and noone gets fired for selecting them. And you do what any one else would do, you purchase your own laptop, go portable, stop using their IDE, stop dev'ing against their behemoth app server, and strip everything vendor specific out of your apps until you can make the switch and decide that even if your business unit is charged with it, its not worth the loss of productivity. And then the sales force comes in with the field engineers pitching the next version of something with a carrot that is more than a quarter away, you look at your manager, then shake your head ;-) I'm not trying to pick on anyone in particular, perhaps I am just venting. However, I just wish that the same kinds of market economics that drive some open source projects to the front while the rest get delegated to the dust bin of sourceforge would work its way into the corporate world where there is real productivity gains to be had, or, that vendors would take their core product open source instead of a token straw man so that it can be improved by everyone without having to use one or more tools + google to figure out if anyone else has had the same problems. I think I am even annoying myself at this point, but I say let's spread the word and become more vocal about it. If you are in a similar situation, let your voice be heard! Talk to someone, provide your managers with real metrics, cost figures, POC, anything.
  18. Re: IBM is a big dustbin[ Go to top ]

    I like to echo my similar sentiment here. WSAD and RAD are decent products since they are all based on great Eclipse platform. The installation of RAD takes 4 to 5 CDs. On top of that, if you need to have it work with the WAS Portal Server, you need to spend a couple of hours to download many additional patches and go through very convoluted installation procedure. You will spend half of a day or a day just to get your environment set up. That was the experience I had when I was with JP Morgan Chase. But if you are a big IBM shop, you don't have much choice. Eric Lu www.chainforge.net
  19. Re: IBM is a big dustbin[ Go to top ]

    I haven't used RAD yet but heard RAD6 is much fatter than WSAD5 and terrible to use (RAD7 only came out last Dec). My experience with WSAD5 is far more pleasant than with a Weblogic shop. Think about it, set up iPlanet, set up Weblogic, set up Weblogic security, config Weblogic, set up the Ant build process on dev machine, edit code in Eclipse, run Ant build, create a script to start and stop Weblogic. Perhaps MyEclipse provides the same support to Weblogic as WSAD to WebSphere.
  20. Slightly off topic but does anyone know when IBM are going to add JEE 5 support to WAS. J2EE 1.4 is feeling rather long in the tooth now and its frustrating not being able to use the new model.
  21. Slightly off topic but does anyone know when IBM are going to add JEE 5 support to WAS. J2EE 1.4 is feeling rather long in the tooth now and its frustrating not being able to use the new model.
    Apparently WAS 7 will support JEE 5.
  22. Apparently WAS 7 will support JEE 5.
    Hope so – any official or rumoured date for WAS7 yet?
  23. There is a feature pack for JEE 5 on WAS 6.1, but JEE 5 won't be fully supported until 7. https://www14.software.ibm.com/iwm/web/cc/earlyprograms/websphere/wsvwas61/
  24. My mistake -- that is a WS pack. Apparently no JEE5 yet.
  25. My mistake -- that is a WS pack. Apparently no JEE5 yet.
    WAS6.2 already runs on Java5. You don't need JEE5 if you don't need JSF1.2 or EJB3.
  26. There is an alfa pack[ Go to top ]

    There is an early preliminary implementation (EJB3 Feature Pack alpha ) on https://www14.software.ibm.com/iwm/web/cc/earlyprograms/websphere/was61ejb3/. Not all features (annotations/deployment decscriptor) are implemented yet and OpenJPA is the JPA Persistence Provider.
  27. J2EE 5 on websphere[ Go to top ]

    http://www.theserverside.com/news/thread.tss?thread_id=41796 I'd just like to voice my appreciation to all the websphere FUDists on this forum, who scare other developers from working with the platform. I'm getting loads of well paid work doing WAS development and infrastructure work, there is a real shortage of people who know the platform. So, thanks very much.
  28. Re: J2EE 5 on websphere[ Go to top ]

    You're very welcome ;-). FWIW, it would be much better for shop's like ours to have someone with deep java and distributed knowledge as opposed to something like WAS per se. I would imagine that there are lots of opportunities for WAS experience in the corporate world, at the same time, I would argue that perhaps the reason why those opportunities exist there is b/c of the amount of effort it takes to develop/deploy using the patterns and technology platforms that are used, as opposed to being more on the pragmatic side. I would also argue that it isn't so much a shortage of knowing the platform as it is the platform.
  29. No one should consider Websphere 6[ Go to top ]

    Having used Websphere 6 for about a year, I can only recommend it to my competitors, as it is absolutely impossible to have any sort of progress with it.