Home

News: Oracle for big businesses, open source for other businesses?

  1. I listened in on Oracle's web-cast on what their direction is going to be seeing that the Sun acquisition has been finalized. The main goal being that they will be collapsing Sun software technology into their current software stack where it makes sense, and to push conflicting Sun Software products such as Glassfish down to a role of being used for non-mission critical department apps. From "software to disk," their complete systems solutions will be fully integrated and fully tested to work as a complete software and hardware systems stack. They will also be seeking to collapse third party Sun support back into Oracle direct support for all their products (Basically no business or less business for previous Sun Partner companies.) Listening to the whole thing was very interesting and at the same time provided me with some of the answers I needed about what to expect from Oracle. On the one hand, when they kept repeating the theme of fully integrated open standards hardware and software components, I couldn't help thinking, "Frankenstein technology" rather than the modular approach that Sun had took. But they introduced customers who really needed and look forward to employing Oracle's fully integrated technology stack. These company representatives of really big Telco, engineering, and financial companies, are looking for a single point of support and accountability, and companies that can help them reduce system management complexity and overhead. The other aspect of their presentation was that they were not looking to play in the small to medium hardware (x86) and software technology space that companies like Microsoft play in. They will be looking to play in the big iron, high compute, mission critical space. So you really can't blame Oracle for being Oracle. They are that big and powerful because they have customers who buy what they are selling. With all the talk of complete systems integration and them looking to be the IBM of the 1960's, I couldn't help but wonder if this was consistent with the theme of Java, and Java EE. That is, open, portable, standards based, and modular. As a Java EE developer, I just didn't feel them. And now that they got MySQL, which they said that they'd roll up into an "open source" division of their company responsible for selling open source tech to customers who want that, it still didn't sit well with me. Basically, it seems that the technology landscape is going back the way of the days of the expensive mainframe and IBM COBOL technology days. I even found it funny when some of the former Sun VP's who were making their presentations were all wearing nice business suits - a far cry from the T-shirt and jeans days of Sun. Oracle is straight hard nose business. They don't play around. So that leaves me to now think that open source technologies, and other alternatives to Java, and Java EE now have a chance of really stepping up and expanding their place in the technology market to avoid being somehow constrained by the enveloping cloud of the big three technology giants - i.e. Oracle, Microsoft, and IBM. Maybe it's time for the programming language and software middle-ware transition to take place beyond the foundation laid by Java, just like like what Java did around the days of COM, DCOM, and other fragmented and proprietary middle ware server technologies that were being peddled by a host of vendors back in the old days. So maybe the Gartner report on light weight frameworks running on cloud platforms taking over from the perceived heavier Java EE, and .NET deployments may have a point. Basically, I see two camps emerging: One camp will be for the already established big fortune 100, 500, and 1000 companies going with the big iron, fully integrated, and pricey "Frankenstein" tech provided by SunOracle, while another group of technology savvy/capable organizations (e.g. Google, and Yahoo), going the way of stringing together best of breed open source solutions already out there. These solutions will be run on a grid of hundreds/thousands of commodity hardware or on some cloud platform.

    Threaded Messages (53)

  2. Small stacks are always more productive, because of their speed, focus and ability to adapt. My choices: 1. Wicket for UI 2. Spring for autowiring and small utilities 3. .... for document-oriented database Do you have any suggestions for the 3rd? Sometimes I'm a bit tired of re-building our home-grown solution....
  3. 3. .... for document-oriented database

    Do you have any suggestions for the 3rd? Sometimes I'm a bit tired of re-building our home-grown solution....
    I would suggest checking out the Java Content Repository API. While it hasn't been adopted to the level I was hoping, for example EMC Documentum doesn't support it at all; content repositories are an effective abstraction above databases that add valuable services like APIs for versioning of data; metadata; transactions; observation of changes in data; and import or export of data to XML in a standard way.
  4. Glassfish[ Go to top ]

    "The main goal being that they will be collapsing Sun software technology into their current software stack where it makes sense, and to push conflicting Sun Software products such as Glassfish down to a role of being used for non-mission critical department apps." I guess there goes the clustering and the general excellence that I've enjoyed with Glassfish over the past few years.
  5. Cool thing about Java[ Go to top ]

    The cool thing about Java is that you are not forced to use a single implementation. There are a lot of alternatives available and personally I find the difference between the different J2EE servers not that big from a developer point of view; since all/most implement the JEE specs. In most cases you want-to/can remain container independent, so it doesn't matter much which JEE server is used.
  6. IBM or Google should move in[ Go to top ]

    This would be a really good time for IBM or Google to move in with their own standardized open source J2EE stack based on JCP standards and existing reference implementations. Just like IBM did with Eclipse, they could change the game here. I think a lot people will feel more comfortable with IBM, and certainly Google, than with Oracle. Hard nosed business doesn't always mix so well with open source. This whole idea of just supporting the big iron makes me shiver. I mean, I'm just curious, for the average J2EE implementation, who has (ever) used Oracle software (excluding SUN obviously)?
  7. Re: IBM or Google should move in[ Go to top ]

    This would be a really good time for IBM or Google to move
    IBM? Have you worked with the bloatware they dish out under the "WebsFear" brand (and yes this includes v7). IBM like Oracle is much more committed to SCA than J2EE/JEE.
  8. Re: IBM or Google should move in[ Go to top ]

    This would be a really good time for IBM or Google to move

    IBM? Have you worked with the bloatware they dish out under the "WebsFear" brand (and yes this includes v7). IBM like Oracle is much more committed to SCA than J2EE/JEE.
    IBM has been committed to open source app servers through WebSphere Application Server Community Edition for quite some time. They also have a fair number of committers on Apache Geronimo, the basis for WAS CE. The Geronimo community recently published a roadmap indicating they're moving towards an EE6 implementation. Unless IBM embraced open source and was committed to Java EE, I find it hard to believe the Geronimo community would be heading in this direction. IBM has also continually updated and enhanced their "bread and butter" WebSphere App Server product, recently enhancing it to add new programming models such as CEA (or Communications Enabled Apps), which enables folks with Java skills to develop new apps enabling a form of user-to-user interaction that was more difficult to do before. So .. given all that, I can see why IBM is focused on SCA as well, as there's always been a need to make SOA more consumable. And SOA is a huge business for IBM. But I wouldn't say IBM is "much more committed to SCA than J2EE/JEE." IBM has enough of a customer base for their software that they can - and I bet will be - committed to both. Disclaimer: I admit that I work for IBM - my post here represents my opinions and may not accurately reflect IBM's opinions or positions.
  9. Re: IBM or Google should move in[ Go to top ]

    Folks,
    IBM has been committed to open source app servers through WebSphere Application Server Community Edition for quite some time.
    We are similarly committed to a solid, lightweight open source Java EE 6 Web Profile implementation with Resin and CanDI, aimed squarely at the mid-market, including clustering, load-balancing and management features. As part of our CDI portable extensions suite, we are creating integration with Struts 2, Wicket, iBATIS, Quartz, etc in addition to JSF 2, JPA 2, EJB 3.1 Lite @Schedule, etc. JBoss is creating a separate set of portable extensions as part of Seam 3: http://seamframework.org/Documentation/Seam3Modules. OpenWebBeans plans a suite of portable extensions for Apache APIs: http://openwebbeans.apache.org/1.0.0-SNAPSHOT/index.html. We definitely need community input as to what portable extension APIs developers find useful as well as early feedback on our Java EE 6/CDI implementation. Thanks for your support, Reza =============================== Resin EJB 3.1 Lite Container Developer Expert Group Member, Java EE 6 and EJB 3.1 Author, EJB 3 in Action
  10. Re: IBM or Google should move in[ Go to top ]

    No offense, but I see the CE version as a big waste of time for IBM. I think it would have been much more thoughtful to have OS'd the core of WAS way back when, and competed through non-OS parts that they wanted to charge for. I'm sure one reference customers that use it, but I haven't seen or heard of one myself (perhaps I'm in a cave). Personally, I'd love to be able to work in a place that forces (for lack of a better word) the use of big vendor products and have the same experience of shooting up GlassFish v3 in 1/2 second (actually 483ms) on my underpowered laptop instead of getting up to get a cup of coffee and stir in the sugar, and check email, and....then check to see if code is committed yet..back to email...etc.etc. I'd give CE more of a chance if it was actually the base to the pay for WAS product (which I don't think it is, but please correct me if I'm wrong). Jin
  11. Re: IBM or Google should move in[ Go to top ]

    No offense, but I see the CE version as a big waste of time for IBM. I think it would have been much more thoughtful to have OS'd the core of WAS way back when, and competed through non-OS parts that they wanted to charge for. I'm sure one reference customers that use it, but I haven't seen or heard of one myself (perhaps I'm in a cave).
    There are quite a number of customers that use it (as well as references). Can't state them here, but what I can tell ya is you can peruse the IBM Business Partner Directory and you'll notice hundreds of ISVs have solutions enabled on WAS CE .. I'd argue they wouldn't have solutions if there weren't customers to buy them :)

    Personally, I'd love to be able to work in a place that forces (for lack of a better word) the use of big vendor products and have the same experience of shooting up GlassFish v3 in 1/2 second (actually 483ms) on my underpowered laptop instead of getting up to get a cup of coffee and stir in the sugar, and check email, and....then check to see if code is committed yet..back to email...etc.etc.

    I'd give CE more of a chance if it was actually the base to the pay for WAS product (which I don't think it is, but please correct me if I'm wrong).

    Jin
    You're right, it's not the base for the WAS product .. we have made an App Advancement Tool available to help "advance" apps from CE to WAS (but for full disclosure, it's not perfect, for example any CE-specific features/APIs etc such as OpenEJB or ActiveMQ need to be manually moved). And I fully understand the pain you mention with regard to development unit testing .. and I tend to believe IBM understands the pain as well, and is listening. Therefore I wouldn't be surprised if in the near future there are some things that you see done to help with the situation. Watch the developerWorks zone for WAS , if you catch my drift. And also keep your eyes peeled on the Cloud offerings we're making available.
  12. I've been a firm Java EE follower for quite a while now, however, there have been quite a number of occasions that I found myself picking up bits and pieces of open source software to get a couple of things done that may not be elegantly covered in the Java EE stack. So while the projects started out as honest to goodness standard by the book Java EE projects, things like iBatis, Quartz, Struts, and Jakarata Commons, were being "sprinkled" here and there throughout the projects. Now as of late, I've been doing a lot of work with Wicket, including JPA, and EJB 3.x. So while I want to support the Java EE standard initiative, I recently began to ask myself, "if I keep using these open source techs, what's really left in the Java EE standard to keep hanging on to for me to continue deploying my apps to a standard Java EE app server?" The only answer I came up with was EJB 3.x, and maybe CDI. However, the takeover of Sun by Oracle, has now left me really wondering if the economics of the Java EE Application server will make sense for webscale projects going forward? Many developers will expend a lot of time on various threads on the web arguing about the functional/performance attributes of Websphere vs Weblogic vs Glassfish vs Tomcat vs whatever, but for me, I'm wondering if we as developers have thrown economics and common sense out the window. The reality is that these application servers don't come cheap, and unless you work at a company where money doesn't matter, breakout companies like a Google, or Facebook, and other great start-ups currently in the dark looking to bust out, surely couldn't have taken the initial road of saying, "lets build everything on Weblogic, Websphere, and Oracle." That just wouldn't make economic sense for companies looking to build web-scale sized projects. I mean, recently Facebook donated a whole US$100K to the Apache Software Foundation saying how they couldn't have done it without them - and I'm sure there are many more companies like that. So, if Facebook, Google, Yahoo, and others are doing open source, why not join the party I say? Heck, just two days ago I decided on really looking at Spring, and you know what?...I love it already, even though I'm just a couple a chapters through the Spring Recipes book. That's why I don't like to join the technology bashing game when it makes no real comparison on price, functionality, and performance. So after I'm done learning Spring, it looks as if I can finally cut the Java EE umbilical cord, and port the current Java EE app I'm working on over to a Wicket + Spring + Jetty/Tomcat + MySQL/PostgreSQL stack. If I'm going bananas here, I would love for someone to point out where I'm going wrong with this decision? And I'm not looking for people to say vague things like "Weblogic is faster than Jetty," or "Java EE is a standard and Spring and Wicket are not," type of arguments. As far as I see it, it is becoming clear to me that the label of Java EE being a standard is becoming harder to sell even though the latest Java EE 6 implementation looks functionally good. The only standard seems to really be, what people want, and I want good, affordable, well supported, well documented, portable, and interoperable components that work and scale well together. From where I sit, it seems that things like Apache Software Foundation is the real standard.
  13. - If we use Spring + Jetty/Tomcat then how will we achieve business object failover? - How hard to manage load balancing/cluster? - Will managing the application on this stack as easy as commercial apps servers? My experience is with Weblogic and Oracle apps server and lot of time I think how easy to move to open source stack that will able to answers most of the questions that enterprise application needs in its lifetime Thanks, Shabbir
  14. Bye Sun[ Go to top ]

    Just a curious thing we joke about around here...Oracle spelled backwards reads "el caro" which translated to spanish literally means "THE Expensive". Guess that speaks about it's business for itself.
  15. May be oracle can spin off a a community version of oracle DB itself or a oracle express at a cheaper cost? why not ?
  16. One thing Oracle has always been good at is, "killing products". Sometimes they kill good products, but more often they have managed more successful acquisitions and killed bad products. I would be happy to see Glassfish out of the way from JEE. It is an useless crap alive only to try out sample apps whenever new JEE spec comes out. Once you have other app servers compliant with new spec, no one touches Glassfish. Oracle will kill Netbeans eventually and follow Eclipse more strongly. Oracle and Weblogic had tried their best earlier with their own IDE and failed miserably. Netbeans is not a great product to bet on to compete against Eclipse. MySQL may find its place in open source world with no commitment from Oracle in few years and developed purely open source. Eclipse was started by IBM, but now it is on its own with many vendors sponsoring and supporting it. Hopefully MySQL follows same path. MySQL is used extensively by big companies and they will not let it die so easily. If MySQL cannot come out of Oracle control, you will see most companies planning exit strategy to move away to another opensource DB (may be Progresql..)
  17. It is an useless crap alive only to try out sample apps whenever new JEE spec comes out. Once you have other app servers compliant with new spec, no one touches Glassfish.
    Are you sure ? http://blogs.sun.com/stories/
  18. The issue is that before this, Sun was committed to making Glassfish an "enterprise" class and capable app server. It was designed, polished, and promoted to compete with the like the of Weblogic and WebSphere. How well it accomplished that is a matter for debate, but the intent was clear that this was the direction they wanted to go. The way it sounds now is that Glassfish will be downgraded to a "reference implementation". Any "cross BEA and Glassfish DNA" will likely ensuring that Weblogic can run Glassfish application with minimal changes, plus taking any technology out of the Glassfish that Weblogic needs (notably JEE 6 tech) and incorporating it in to Weblogic. I don't see any indication of motivation from Oracle to push tech down from Weblogic in to Glassfish, to promote or ensure that Glassfish is "enterprise class", or to even support it at that level. Rather they'll likely create as "painless" an upgrade path to move applications from GF to WLS, and upsell WLS to clients to solve "big problems" rather than improve GF to meet those needs. "Glassfish is for departmental servers, you need WLS for 'the enterprise'." This is basically a golden opportunity for RedHat/JBoss to step up, because it's primary OSS JEE app server competitor is going to have it's reigns pulled back stopped in its tracks. I also question if there will ever be a "JEE 7". I haven't seen anything from Oracle that suggests they'll be pushing for a new specification, particularly if they think they'll be on the hook for implementing the reference platform for it. Rather, they'll just "featurize" their existing products to distinguish their servers from their competitors. Basically, I fear that the app server market is going to fragment at this point, with the common core at JEE 6 (once others get up to speed). The bright side, is that if this happens, it happens at JEE 6 rather than 5, but the dark side is considering what nice improvements were made between JEE 5 and JEE 6, who knows what will be lost if there is no JEE 7.
  19. The issue is that before this, Sun was committed to making Glassfish an "enterprise" class and capable app server. It was designed, polished, and promoted to compete with the like the of Weblogic and WebSphere.

    How well it accomplished that is a matter for debate, but the intent was clear that this was the direction they wanted to go.

    The way it sounds now is that Glassfish will be downgraded to a "reference implementation". Any "cross BEA and Glassfish DNA" will likely ensuring that Weblogic can run Glassfish application with minimal changes, plus taking any technology out of the Glassfish that Weblogic needs (notably JEE 6 tech) and incorporating it in to Weblogic.

    I don't see any indication of motivation from Oracle to push tech down from Weblogic in to Glassfish, to promote or ensure that Glassfish is "enterprise class", or to even support it at that level.

    Rather they'll likely create as "painless" an upgrade path to move applications from GF to WLS, and upsell WLS to clients to solve "big problems" rather than improve GF to meet those needs. "Glassfish is for departmental servers, you need WLS for 'the enterprise'."

    This is basically a golden opportunity for RedHat/JBoss to step up, because it's primary OSS JEE app server competitor is going to have it's reigns pulled back stopped in its tracks.

    I also question if there will ever be a "JEE 7". I haven't seen anything from Oracle that suggests they'll be pushing for a new specification, particularly if they think they'll be on the hook for implementing the reference platform for it. Rather, they'll just "featurize" their existing products to distinguish their servers from their competitors.

    Basically, I fear that the app server market is going to fragment at this point, with the common core at JEE 6 (once others get up to speed).

    The bright side, is that if this happens, it happens at JEE 6 rather than 5, but the dark side is considering what nice improvements were made between JEE 5 and JEE 6, who knows what will be lost if there is no JEE 7.
    @Will Harthung I agree with everything you have said here, and yes, some of these speculations were also confirmed by Oracle as their future plans. The plus side to all of this I hope, is that the JRockit deterministic JVM GC capabilities are as good as they say in regards to their plans to mix JRockit JVM features with those of the Hotspot JVM to get one unified JVM having all the best features from both - i.e. Oracle JVM. I will concede though that Oracle like any other business in this situation has to somehow consume, digest, and produce a common technology platform set to support. It might not be business feasible for them to be supporting everything from Sun as is with their already existing and overlapping technologies, that according to them is already making good money for them - hence, somethings will have to be either put on the chopping block or assimilated into Oracle's technology portfolio over the short to medium term. I will miss Glassfish for sure. Peace.

  20. @Will Harthung
    I agree with everything you have said here, and yes, some of these speculations were also confirmed by Oracle as their future plans.

    The plus side to all of this I hope, is that the JRockit deterministic JVM GC capabilities are as good as they say in regards to their plans to mix JRockit JVM features with those of the Hotspot JVM to get one unified JVM having all the best features from both - i.e. Oracle JVM.
    Combining Hotspot / OpenJDK and JRockit would indeed yield a pretty impressive JVM. Anyone from Oracle have thoughts about how that would be done :
    GPL JRockit ? Fork and Relicense OpenJDK ? Something else ?
    Rich Sharples Red Hat

  21. @Will Harthung
    I agree with everything you have said here, and yes, some of these speculations were also confirmed by Oracle as their future plans.

    The plus side to all of this I hope, is that the JRockit deterministic JVM GC capabilities are as good as they say in regards to their plans to mix JRockit JVM features with those of the Hotspot JVM to get one unified JVM having all the best features from both - i.e. Oracle JVM.


    Combining Hotspot / OpenJDK and JRockit would indeed yield a pretty impressive JVM. Anyone from Oracle have thoughts about how that would be done :

    > GPL JRockit ?
    > Fork and Relicense OpenJDK ?
    > Something else ?

    Rich Sharples
    Red Hat
    Indeed. This is a pleasant exercise for the combined JRockit and HotSpot teams :-) If you have followed the public announcement you already know that our plan is to carry both forward and merge the technologies over the long term. The exact method and timeframe is not clear yet. These are early days and we need to get the development teams time to do the due diligience - the code bases are quite different and both have strong and weak points. Luckily, our new ex-Sun colleagues have experience in merging code bases from previous acquisitions. OpenJDK will remain the single open source Java and JVM implementation that Oracle contributes to. Open sourcing the current JRockit code base simply does not make sense. Better focus our efforts on the future unified offering which will be a merge of some kind between the two technologies. Another thing worth mentioning here is that some features in JRockit do not make sense as part of an open source JVM. One example is integration with the Fusion Middleware logging infrastructure - definitely not of interest outside of an Oracle context. The open source codebase should not be polluted by such proprietary extensions. I don't have an answer to how to accomplish this yet, but it's possible that the modularity work in JDK7 can help. I fully understand that the community is anxious to know the details of where this is heading but I must plead for patience. Our clear goal is to keep Java strong and vibrant for the benefit of Oracle, our customers, partners and the community (including our esteemed competitors). The role of steward of the Java community and provider of the reference implementation is new to us and we don't want to upset anyone by rocking the boat. Regards, Henrik Ståhl Director PM, JRockit Products speaking for the combined Oracle and Sun JVM teams

  22. @Will Harthung
    I agree with everything you have said here, and yes, some of these speculations were also confirmed by Oracle as their future plans.

    The plus side to all of this I hope, is that the JRockit deterministic JVM GC capabilities are as good as they say in regards to their plans to mix JRockit JVM features with those of the Hotspot JVM to get one unified JVM having all the best features from both - i.e. Oracle JVM.


    Combining Hotspot / OpenJDK and JRockit would indeed yield a pretty impressive JVM. Anyone from Oracle have thoughts about how that would be done :

    > GPL JRockit ?
    > Fork and Relicense OpenJDK ?
    > Something else ?

    Rich Sharples
    Red Hat


    Indeed. This is a pleasant exercise for the combined JRockit and HotSpot teams :-)

    If you have followed the public announcement you already know that our plan is to carry both forward and merge the technologies over the long term. The exact method and timeframe is not clear yet. These are early days and we need to get the development teams time to do the due diligience - the code bases are quite different and both have strong and weak points. Luckily, our new ex-Sun colleagues have experience in merging code bases from previous acquisitions.

    OpenJDK will remain the single open source Java and JVM implementation that Oracle contributes to. Open sourcing the current JRockit code base simply does not make sense. Better focus our efforts on the future unified offering which will be a merge of some kind between the two technologies.

    Another thing worth mentioning here is that some features in JRockit do not make sense as part of an open source JVM. One example is integration with the Fusion Middleware logging infrastructure - definitely not of interest outside of an Oracle context. The open source codebase should not be polluted by such proprietary extensions. I don't have an answer to how to accomplish this yet, but it's possible that the modularity work in JDK7 can help.

    I fully understand that the community is anxious to know the details of where this is heading but I must plead for patience. Our clear goal is to keep Java strong and vibrant for the benefit of Oracle, our customers, partners and the community (including our esteemed competitors). The role of steward of the Java community and provider of the reference implementation is new to us and we don't want to upset anyone by rocking the boat.

    Regards,

    Henrik Ståhl
    Director PM, JRockit Products
    speaking for the combined Oracle and Sun JVM teams
    Henrik, thanks for responding - it's good of you to keep the dialogue open. I think that more than anything will keep the anxiety at bay. I wasn't really expecting a full answer this soon - I fully understand the size and complexity of the integration. Regards, Rich
  23. You are in a position to solidify the JVM as the platform of choice, by combining the best of both worlds from HotSpot and JRockit, for developers for the next 20+ years. I can see people get all sentimental about Sun, but I firmly believe Oracle can deliver both, solid technology /and/ profits, whereas Sun could obviously do only one. Godspeed!
  24. Will/Douglas, Hopefully, you guys know by now that I don't say nonsense that I don't believe in (at the end of the day, the truth is that I really have no a reason to). Here is what I can say without overstepping my personal boundaries and jeopardizing hard-earned relationships:
    The way it sounds now is that Glassfish will be downgraded to a "reference implementation".
    I think you are reading too much into this and it is understandable if you feel strongly about GlassFish. The fact at the moment is that nothing has changed or likely to change in the near future. GlassFish simply has too much traction for the pretty smart Oracle folks to do something stupid here and Oracle can easily support two product lines that don't really compete directly. I think you guys should give Cameron's post a close look in this regard and respect the fact that he probably can't overstep his personal boundaries either. In the highly unlikely scenario that Oracle does really do something stupid, folks like us or JBoss are no dummies and can easily fill any gap in the market.
    I also question if there will ever be a "JEE 7".
    I can pretty much guarantee you are dead wrong on this one. Oracle has every intention of continuing Java SE, Java EE and Java ME. No matter how pessimistic of a mood you guys are in at the moment, the simple fact is that Oracle just made a gigantic investment in Java and they would maximize those investments, not do things to reduce their ROI. The only thing that endangers Java is willy-nilly making these kind of statements that really don't have much basis in fact :-). That's as far as I can go on this. What you choose to believe after that is really up to you. After all, you can't make someone believe there is no bogey man hiding in the closet if you are dead-set in believing that there is :-). Maybe the Oracle guys will think it is worth it to go a step further and keep the line of communication wide open and minimize rumors/speculation the best they can considering the size of the company and the long chain-of-command. Hope it helps guys, Reza
  25. The fact at the moment is that nothing has changed or likely to change in the near future. GlassFish simply has too much traction for the pretty smart Oracle folks to do something stupid here and Oracle can easily support two product lines that don't really compete directly.
    You may well be better informed with your place in the industry, I am but a casual outside observer. I've always considered GF a competitor to WLS. Arguably it was more competition to JBoss, that's valid, but it certainly competed with WLS on the low end, and strived to make a dent on the high end. There was even marketing materials and benchmarks contrasting GF with WLS in terms of a "value" solution. From the SPEC marks, WLS appears to be the fastest, per instance, on really high end hardware, but it also seemed that GF compared favorably. If you were stuck on a small amount of large machines, WLS performed very well. But if you were willing (and able, given application design) to throw cores at the problem, GF scaled pretty good. While GF may not have been able to take WLS on the top and, the bulk of the app server market isn't at that level (at least in terms of number of instances/licenses, perhaps the top tier is there in terms of revenue). So, I have always considered GF a competitor to WLS, as it offered the licensing value and options of JBoss (i.e. free if you want, cheaper if you want support), with the ease of use, and polish of WLS (arguably GF is easier to use, but I haven't used WLS in some time) with "performance enough" for many applications. That said, it is hard to imagine Oracle putting the time and money into making GF even more enterprise level, to be even more competitive at the top end. I'm not doubting Oracles vision or commitment, I just see GF, for the most part, redundant in their line up. Make migration from GF to WLS as painless as possible (native support for sun-web.xml, along with an appropriate class loader model is a good start, letting many GF apps simply deploy) and slowly phase it out or leveraged solely as a reference platform. It is reassuring that you feel they are committed to the JEE as an industry standard, and to its improvement.
  26. Folks, In case you missed it, here is some more info on the GlassFish issue: http://blogs.sun.com/arungupta/entry/glassfish_strategy_by_oracle_sun. Now, this is from an Oracle guy, but I am sure the GlassFish community leaders will speak out as soon as they can too. You simply have to have some patience/faith and give them some time considering they are on the wrong side of the command-chain at the moment (but are in a good position nonetheless exactly because of the strong GlassFish community and the revenue stream from GlassFish licensing). Cheers, Reza
  27. Will,
    That said, it is hard to imagine Oracle putting the time and money into making GF even more enterprise level, to be even more competitive at the top end.
    This is indeed fact and has been fact for as long as the acquisition has been on the table. If this is what you are wishing for, I agree that is not going to happen now (the same, BTW, is true for MySQL vs. Oracle DB). That doesn't mean GlassFish is not valuable as a community outreach/R&D tool for Oracle that has decent licensing revenue on its own, meets mid-market needs and will likely result in up-sell customers. All these factors mean that it will be a bit more than "just a reference implementation" - meaning likely continuing to have clustering, load-balancing, management, etc. If you sincerely mean GlassFish well, selling Oracle short is the wrong thing to be doing at the moment, especially if it is based on shallow stereotypes/ignorance. Oracle is exactly the right company for taking the helm of Java at this point. From where I stand, their hard-nosed, no-nonsense approach to things is a lot better than Sun's constant struggle to find business chops/make the tough decisions and they are far more progressive and open than some other large companies that would have been likely candidates for an acquisition of this magnitude. From a personal standpoint, their track record at the JCP makes them great partners for some of the changes we need to be making to Java EE in the next coming years and I look forward to working with them for that reason alone if not anything else. They have certainly helped make decisions that are pretty community oriented, much more so than some of the people/organizations that you might believe to be community oriented. As far as stuff that's generally visible, Cameron mentioned some of those. Hope it helps, Reza
  28. Folks, Here is even more info on GlassFish on the updated Oracle application server site: http://www.oracle.com/us/products/middleware/application-server/index.htm. Some quotes from the landing page:
    Sun offerings including the Sun GlassFish Enterprise Server, Sun Java System Web Server, Sun Java System Web Proxy Server, and Java HotSpot Virtual Machine complement and extend the existing product line-up to increase the breadth and depth of Oracle's foundational middleware software.
    GlassFish continues as the Java EE reference implementation
    GlassFish Server included in every Oracle WebLogic Server offering
    Sun Java System Web Server part of new Oracle Web Tier offering
    Hope this helps with some of your (genuine and understandable) concerns around this. Cheers, Reza
  29. Something that our marketing folks did not put on the page but is clearly stated in Hasan Rizvi's broadcast around 8:25 in voice and slideware is the additional statement of: "GlassFish also available as a standalone offering" - that is to be read that as independent of the commercial packaging with WebLogic Server. It is also, perhaps more important, to remind everyone that the GlassFish community/implementation remains open source - there is no change in the GlassFish project per the various commitments you can see on the Oracle sites. The software remains in open source, downloadable and usable as pure open source software independent of commercial licensing. Folks like Eduardo Pelegri-Liopart and team still are the voice and conscience that will drive Oracle's plans around GlassFish and work with the community to shape the GlassFish implementation. Eduardo's summary post today gives related links: http://blogs.sun.com/theaquarium/entry/summary_of_post_oracle_changes Clearly as Oracle we made a very large commitment to WebLogic Server underpinning Fusion Applications and Fusion Middleware and executed on that over the last 18 months with the delivery of Fusion Middleware 11g. Adding GlassFish to the portfolio enables our customers to take advantage of capabilities in GlassFish as they see fit for Java applications. Frankly we have a fair number of customers who use both and that is a good thing. Interestingly, with GlassFish being the only production quality Java EE 6 implementation on the market is a great example where the open source community using GlassFish and anyone else including Oracle customers can take advantage of production quality software that is standards compliant with the latest innovations *now* (for example the Web profile being an important one for many folks to try out and get their heads around in practice). Further as we articulated at Open World this year we also are very much pursuing Java EE 6 on WebLogic Server as well and have further incentive to do so given Oracle's increased role in the Java community. I look forward to IBM and JBoss stepping up to their formal roadmaps around Java EE 6 to so that we continue to have a strong competitive ecosystem around Java EE (as JBoss did a week or so ago). We want to go head to head with other servers - you don't have to be a rocket scientist to see where the servers differ and are stronger/weaker than one another and having choice is a good thing. Lastly, at Oracle, as hardhearted as folks may think we are, there is an equal commitment during acquisitions to preserve customer's existing investments in software and either carry forward that commitment or provide a clear transition path. Sometimes those decisions can be painful but on the positive side we tend to be pretty clear about them and provide the roadmap with pretty reasonable support timelines for those transitions. As Reza notes from the oracle.com pages and Hasan's broadcast that carry forward commitment is pretty public in the case of GlassFish Server and Sun Java System Web Server - the product (current and old versions) continue and existing customers will be supported, it remains available standalone as well as in new packages that provide the option to use WebLogic. We are very excited to have GlassFish on board at Oracle. Areas like GlassFish proper and key constituent pieces like Metro, Grizzly, Jersey etc are state of the art; there is huge incentive to shepherd them correctly. You will have to judge us on how we execute over the next few months while we realign roadmaps and delivery schedules but I hope we can surprise folks with some good turns we can do as a combined company. Mike. For the record, as I am not as prolific as Cameron and other Oracle folks, my role: Mike Lehmann Product Management WebLogic Server and Java Infrastructure (now including GlassFish)
  30. Glassfish[ Go to top ]

    I've always considered GF a competitor to WLS. Arguably it was more competition to JBoss, that's valid, but it certainly competed with WLS on the low end, and strived to make a dent on the high end. There was even marketing materials and benchmarks contrasting GF with WLS in terms of a "value" solution.

    From the SPEC marks, WLS appears to be the fastest, per instance, on really high end hardware, but it also seemed that GF compared favorably. If you were stuck on a small amount of large machines, WLS performed very well.
    Keeping in mind that the following is my opinion only (as I do not work directly on WLS or Glassfish): Glassfish is an extremely effective JBoss competitor in the market, judging both by download statistics and market penetration, so I would guess (and hope) that Oracle will find a way to continue to grow it. Peace, Cameron Purdy | Oracle Coherence http://coherence.oracle.com
  31. Folks,
    One thing Oracle has always been good at is, "killing products".
    While I'm certainly not in the business of defending Sun/Oracle (they should be doing a much better job at that themselves with their deep pockets and not simply assume some high-level web presentations are sufficient in addressing people's real concerns :-)), I see this kind of statements as way too speculative/sentimental/reactionary and is likely the wrong predictions by all signs that I can see and have known about for quite some time now. In fact, it seems to me that Oracle is committed to supporting NetBeans, GlassFish and MySQL without sacrificing their "bread-and-butter" large enterprise market. I certainly have nothing but respect for the folks that I know personally at Oracle as outstanding, progressive, grounded and open-minded engineering professionals. For reference, here are a few articles that I feel more closely reflects what Oracle/Sun actually said: http://java.dzone.com/articles/oracles-roadmap-sun http://java.dzone.com/articles/new-dawn-java http://java.dzone.com/articles/sun-has-not-set-oracle-makes http://java.dzone.com/news/javaone-lives Hope it helps, Reza =============================== Resin EJB 3.1 Lite Container Developer Expert Group Member, Java EE 6 and EJB 3.1 Author, EJB 3 in Action
  32. Glassfish in production systems[ Go to top ]

    "It is an useless crap alive only to try out sample apps whenever new JEE spec comes out" It's absolutely not true that Glassfish isn't and can't be used for anything other than playtime apps. We use it for a high volume telephone vote system capable of handling 32K votes per second and it works damned well. We built our system around 64 bit Linux, Glassfish, Wicket, MySQL and J2EE 1.6 (Jrockit 64 bit) and it goes like a (J)rocket. http://www.clickvision.co.uk/ if you're interested...
  33. "It is an useless crap alive only to try out sample apps whenever new JEE spec comes out"

    It's absolutely not true that Glassfish isn't and can't be used for anything other than playtime apps. We use it for a high volume telephone vote system capable of handling 32K votes per second and it works damned well. We built our system around 64 bit Linux, Glassfish, Wicket, MySQL and J2EE 1.6 (Jrockit 64 bit) and it goes like a (J)rocket.

    http://www.clickvision.co.uk/ if you're interested...

    Just to add to this, we've used the system to run the Comres opinion polling for the British election leadership debates for the past 3 weeks (including tonight's, which is on at the moment). Glassfish has so far performed very well and managed, on commodity hardware, a callout to several thousand of the public, the management and appropriate weighting of the results and the publication of the collated results for the ITN News. All this within 2 minutes. Not bad for useless crap.

  34. I'll have to chime in with the others and heartily reject your notion that Glassfish is crap. It was solid at version 1.0 and has been getting better with each release. Much easier to get up and running than JBoss, much more natural to configure than Weblogic, a really nice admin GUI, a REALLY nice admin CLI, native integration with Apache HTTP, and in version 3.0, services on demand. I too, will miss it, and am really sorry to see it go now that the 3.0 version is out, which is so lightweight. My guess, and I'm pretty much 100% on this, is that you've never actually even used the product. There's no way you could come to the "crap" conclusion if you did. And as to Reza's defense of Oracle, basically, "Surely Oracle wouldn't be stupid enough to kill Glassfish" (paraphrased), all I can say is, really? Have you paid attention to Oracle? Sound technological decisions are not their forte. Decisions based on how much money they can generate for sales guys and overpriced support contracts are far more in their wheelhouse. The whole Oracle/Sun buyout is really like a nightmare scenario for me. I can't imagine a company less qualified or motivated to make good decisions based on the needs of developers or customers.
  35. I'll have to chime in with the others and heartily reject your notion that Glassfish is crap. It was solid at version 1.0 and has been getting better with each release. Much easier to get up and running than JBoss, much more natural to configure than Weblogic, a really nice admin GUI, a REALLY nice admin CLI, native integration with Apache HTTP, and in version 3.0, services on demand. I too, will miss it, and am really sorry to see it go now that the 3.0 version is out, which is so lightweight. My guess, and I'm pretty much 100% on this, is that you've never actually even used the product. There's no way you could come to the "crap" conclusion if you did.

    And as to Reza's defense of Oracle, basically, "Surely Oracle wouldn't be stupid enough to kill Glassfish" (paraphrased), all I can say is, really? Have you paid attention to Oracle? Sound technological decisions are not their forte. Decisions based on how much money they can generate for sales guys and overpriced support contracts are far more in their wheelhouse.

    The whole Oracle/Sun buyout is really like a nightmare scenario for me. I can't imagine a company less qualified or motivated to make good decisions based on the needs of developers or customers.
    Agree - just as in any good public company, revenue will drive product decisions. GlassFish is at risk currently and Oracle's statements around aligning GlassFish with WLS stack gives clues on what direction they are taking. The good news is that for majority of web applications, users can and will easily migrate to Apache Tomcat.
  36. Hop on a cloud instead[ Go to top ]

    So Oracle is going to offer lots of "integrated" bells and whistles on top of its database as loss leaders. The database model with value-add might work better than Sun's hardware model with value-add. But at least Sun had style. I hope we see start-ups and small (MySQL-type) companies move more towards Amazon/Google clouds for hosted app/data needs. Oracle is a great database but I just don't see their tight "cathedral" integration story winning anyone over. Especially with open-source and app/data hosting becoming even more accessible.
  37. cloud hopping?[ Go to top ]

    Andrew - After a year of not being able to join in any conversation on this topic (while the deal was pending), it's nice to be able to respond.
    So Oracle is going to offer lots of "integrated" bells and whistles on top of its database as loss leaders. The database model with value-add might work better than Sun's hardware model with value-add. But at least Sun had style.
    Loss leaders? Honestly, there are a lot of things you can complain about with Oracle, but "loss" is not one of those. Oracle has a very good track record at finding ways to make business, including with software that has an open source license. To quote from a recent email that I received on this topic:
    Open Source is just one of the licensing and distribution models Oracle uses with some of its software products to make those products available to customers. Oracle has a history of using the licensing and distribution model that makes sense for a given product and market. Even prior to the Sun acquisition, Oracle had a long track record in Open Source products in terms of distributing, evolving and adding value to Open Source products while also leading and contributing back to Open Source projects. Examples of this include Linux, Virtualization (Xen), Databases (BerkeleyDB, InnoDB), Developer Tools (Eclipse), and Middleware (EclipseLink, TopLink in GlassFish, Apache).
    You said:
    I hope we see start-ups and small (MySQL-type) companies move more towards Amazon/Google clouds for hosted app/data needs. Oracle is a great database but I just don't see their tight "cathedral" integration story winning anyone over. Especially with open-source and app/data hosting becoming even more accessible.
    Sure, why not? And chances are very good that they'll be using some technology products and/or open source packages from (or contributed to by) Oracle to do so. Oracle's in a great position to show real technology leadership, and while that includes leadership in traditional areas (databases, applications, integration and now hardware from Sun), it goes well beyond that. Peace, Cameron Purdy | Oracle Coherence
  38. Re: cloud hopping?[ Go to top ]

    Cameron,
    After a year of not being able to join in any conversation on this topic (while the deal was pending), it's nice to be able to respond.
    It is fantastic to see you reaching out to the community once again! It is a breath of fresh air that Oracle is accelerating the pace of development for Java SE, Java EE and Java ME and showing progressive technological leadership in the same way that you guys have done for Java EE 5+, EJB 3+, JPA 1+ and CDI. The commitment to greater openness/participation in the JCP as well as putting your financial muscle behind Java is much appreciated (beyond the somewhat understandable nay-saying/fears/doubts given the scale of change). Personally, I hope for much better days ahead for Java with a strong Oracle instead of a weakened Sun behind it. Cheers, Reza
  39. JavaFX[ Go to top ]

    Does JavaFX have a future at Oracle? After all, Oracle has its own ADF Faces Components.
  40. Re: JavaFX[ Go to top ]

    Ralf,
    Does JavaFX have a future at Oracle?
    Take a look at the announcements/analysis on JavaLobby that focuses on the concrete parts of the Oracle/Sun road-map. Oracle plans to invest heavily on JavaFX/RIA, not to mention concrete plans for the JCP, Java SE, Java ME, Java EE, modularity, JavaFX, NetBeans, GlassFish, Kenai, OpenOffice, MySQL, JavaOne, SPARC and Solaris. I am sure you can always find negative things to say if that's what you are looking for, but I personally think the road-map is overwhelmingly positive. The only thing I am unsure of is the decision to continue down the SPARC/Solaris path. These two product lines were some of the major sources of chronic weakness for Sun despite being critical to their early success (along with Java of course). Like James Gosling, I really like the idea of Java stewardship being in the hands of a progressive (and aggressive) software company at heart rather than a hardware engineering shop. The other regret of course is that us mid-market guys I am sure will have to work harder to earn our keep now, but then I'd rather be earning my supper fair and square against a capable contender I respect any day... Cheers, Reza
  41. RE: JavaFX[ Go to top ]

    Does JavaFX have a future at Oracle? After all, Oracle has its own ADF Faces Components.
    Today some of ADF Faces components actually generate Flash UI today (for example the graph components which can also generate simple PNG) - we could create ADF Faces components that spit out JavaFX UI if it would make sense. Just one possibility with the combined Oracle/Sun stack
  42. Back to Middle age[ Go to top ]

    Hello Douglas, Your paper is so right. we come to the same conclusion. The vertical stack with Oracle,Oracle and Oracle from the disk to the application send me back to the OS/360 Cobol that rule the 60's and 70's. Every one worked hard to convince our customers, partners, managers and colleagues that open standard approach is credible and beneficial. Sun provided a support to this movement (Thanks to Jonathan Schwarzt). After Wednesday Oracle's web cast, we understand that this movement will not a priority for BIG RED. I hope we are wrong and we will no go back the middle age http://www.pymma.com/eng/People/Blog-Paul-Perez-Chief-Architect/Back-to-the-middle-age2 Paul Perez
  43. Re: Back to Middle age[ Go to top ]

    Hello Douglas,
    Your paper is so right. we come to the same conclusion. The vertical stack with Oracle,Oracle and Oracle from the disk to the application send me back to the OS/360 Cobol that rule the 60's and 70's.
    Every one worked hard to convince our customers, partners, managers and colleagues that open standard approach is credible and beneficial. Sun provided a support to this movement (Thanks to Jonathan Schwarzt). After Wednesday Oracle's web cast, we understand that this movement will not a priority for BIG RED.
    I hope we are wrong and we will no go back the middle age

    http://www.pymma.com/eng/People/Blog-Paul-Perez-Chief-Architect/Back-to-the-middle-age2

    Paul Perez
    Hi Paul. Happy you got a chance to sit in on that webcast too. It was quite long as you mentioned. Many things were said, but you and I seemed have picked out the key/salient points apart from the individual announcements made about the myriad of technologies Oracle has to meld together and/sieve through. I agree with Reza that a perceived weakness in Oracle's game plan may be to continue with the Sparc processor line, and by extension, big iron servers. That was how Sun got into trouble in the first place. They didn't see that the market had shifted to x86/x64 based commodity/mid range machines until it was too late. Dell and HP were killing them. We have our work cut out for us - i.e. - Those who have been preaching to customers and clients, good low cost, quality, commodity solutions to tackle their big expensive problems. If Oracle wants to sell big iron systems that's ok. I just hope they don't go evangelizing to the whole world that big iron, expensive is more reliable, and better performing than everything else. God knows they carry a lot of clout in the technology market and with no other big vendors like Sun to preach otherwise, things can get sticky quickly. Maybe Reza and Caucho can help out here.:)) I remember in the webcast there was a CEO from an IT company in West Africa complaining to Larry about Larry allowing Oracle to go directly to his big Telco customer that he was already providing support for, which would mean killing his business. Larry simply said to him not to worry. To paraphrase Larry, "If Oracle goes to his customer and they confirm that they were already receiving quality service from his IT company, then Oracle would leave them alone, otherwise, they Oracle are going in to take business where no real support value is being provided." :) To me, that's in your face business brawn/brute force.:) What was also funny to me yet serious, was when the same CEO of the West African IT company mentioned how a couple years ago Larry was over there somewhere in Africa preaching at some technology conference about how Unbreakable Linux, running on commodity based servers, in clusters was the way to go to run Oracle DBs - "now," the CEO said, "you, Larry, are now talking about Big Iron systems being the future... what's up here?" to paraphrase the CEO. :) From where I sit, anybody who owns, operates, or works in a company that provides, Java EE based app servers, or platforms, should be very concerned about these kinds of developments. At least they should plan for the worst case. :) Peace.
  44. I appreciate the concerns expressed in many of the posts in this thread. Some of these concerns are based in interpretations of segments of the webcast, some are based on market analysis and some on expectations of Oracle's behavior. The concerns about lack of future investment on JavaEE 7 seem the easiest to address. *All* of Oracle's middleware is based on JavaEE, Oracle has a history of strong support for the JCP, and there are several places in the webcasts where Jeet and Thomas/Hasan talk about a desire to invest on the JCP. The discussion about GlassFish / WebLogic is trickier to address effectively. I know - from having talked with "Oracle classic" folks in the last few days - that some of the phrases used in the Webcasts are being interpreted in a way different to how they were intended. For example Oracle said "GlassFish remains the RI" and they just meant that, but some observers interpret that sentence as code for "we will remove production features" because *many* years ago Sun had separate RI and AppServer groups and two AppServer offerings. Now, Sun stopped doing that because it was too expensive to keep two code bases... and Oracle is more cost conscious than Sun ever was, but I surely understand why the phrase raises eyebrows. Another example of tricky phrasing is "Departmental"; on that, I've been told that that specific nuance was discussed up to the last minute. Another example is the slide shown at 7'40' http://bit.ly/9hsHKb One could interpret it as saying "left column is GlassFish - for development; while right column is WebLogic - for production". Actually, that's how I read the slide it the first time! But that is not what it says, nor what it was intended - what it says is that the combination of GF+WLS addresses the needs of development and deployment. Now, at this point, I can just have a plea for not rushing to (mis)interpret what was said at CiC. You can read what Mike Lehman (WebLogic PM) says in his post in this thread for some other comments, but I think only a RoadMap for GlassFish will address people's concerns. Paraphrasing Miguel de Cervantes: "the proof of the pudding is in the roadmap" - and Oracle has not provided one (yet). Which I think should be interpreted by pro-GlassFish folks as a "good sign" because it shows that the "Oracle classic" folks have waited 9 months to be able to work with the "GlassFish classic" group to put together a combined roadmap, rather than just dictating one. So, my suggestion to GlassFish fans is to please give us some time to put together that roadmap, and then use it to make your own decisions. Thanks for your support of GlassFish. - eduard/o
  45. Thanks for feedback Eduardo. I was busy trying to pay some bills so I'm kinda late responding to all this.
    The concerns about lack of future investment on JavaEE 7 seem the easiest to address. *All* of Oracle's middleware is based on JavaEE, Oracle has a history of strong support for the JCP, and there are several places in the webcasts where Jeet and Thomas/Hasan talk about a desire to invest on the JCP.
    I wouldn't want you and Reza to lose too much sleep over the previous Java EE 7 statement. Don't read too much into it, as from where I sit, I didn't take it at its literal meaning but rather that developers anticipate a significant slow down in the release of subsequent Java EE versions much like the way most developers see how long an IBM takes to get their Java EE app servers implemented to the latest Java EE specs. Glad that you responded to it though, to further shed light on this perceived grey area. Now, in regards to you and the other Oracle folks on this thread trying to do damage control after the last Oracle webcast, I fear that you guys may have go beyond this little thread as I've seen the same negative developer sentiments appearing on other tech news websites apart from TSS.
    I know - from having talked with "Oracle classic" folks in the last few days - that some of the phrases used in the Webcasts are being interpreted in a way different to how they were intended.
    You can't blame the viewing/listening audience for how they interpret things that were explicitly said by Oracle senior executives (i.e. Glassfish to be RI App Server, and Glassfish for Departmental apps). I don't like to feel as if my intelligence is being insulted. If Oracle folks said some things in a webcast that didn't go over well with some people listening on, don't come and say that you didn't mean ...., you meant .... It's not like they were talking to lay people. They were addressing a technology centric audience, hence, you should be clear in what you try to communicate. Saying something is to be used for departmental apps is not a statement to make off the cuff without backlash if you didn't mean it that way, especially when you went further to say that Weblogic will continue to be strategic app server for enterprise deployments. Two contrasting statements here. I'm sorry for being a stickler for language and exposition. :)
    Now, at this point, I can just have a plea for not rushing to (mis)interpret what was said at CiC...
    Even you seemed to have been doing a little deciphering yourself. I can only try to ask Oracle folks to try and get the communication right the first time please. Most Java techie folks I suspect are INTP/INTJ personality types. This Sun/Oracle merger is a very significant event in the technology space - hence the lines of communication from Oracle needs to be frequent, and clear. You don't need someone like a Reza who doesn't work for Oracle trying to do damage control for Oracle. We now have a situation where a camp of Sun techie fans are being pushed into a situation to now join the Oracle camp of which they were originally not apart. You're going to have a clash of cultures, and expectations. To ensure some balance here, I've always agreed with the view that Java needs a savvy and strong business steward with deep pockets behind them, and I think that Oracle has that going for them. It helps to portray the image to big business of Java being upscale, and gated community like - much like a .Net rather than say a PHP :)) Peace.
  46. I don't like to feel as if my intelligence is being insulted. If Oracle folks said some things in a webcast that didn't go over well with some people listening on
    It was not my intent to dismiss anybody's concerns but rather to provide some additional context to what was said in the presentations. Remember, all this was happening pre-CiC and the Sun/GlassFish team had no opportunity to review any of the presentation. We could not say: "hey, the GF community will likely interpret *that* statement in *this* way" As I said, I think the only way to address the (valid) questions about the future of GlassFish at Oracle is through actions like delivering a roadmap. We are working on these but won't happen overnight.
  47. Douglas, I really don't think there is any reason to expect delays on Java EE 7. On the contrary, I believe it will be a faster and even more effective release than Java EE 6 :-). I do agree some of the other media outlets/blogs seemed to have interpreted things the same way you (guys) did. Now that we all hopefully have a better understanding of things, perhaps we should do our part in helping ensure bad things don't happen to a platform all of our livelihoods depend on (at least avoidably) :-)? As such, I see it as my privilege to speak out on anything that I see as being incorrect, there really is no burden here. The good folks at Oracle/Sun certainly deserve that much and so much more... For one thing, I am in a position to have better insight than most folks, so I would be remiss if I did not use it where it is needed... Cheers, Reza
  48. Netbeans future[ Go to top ]

    As a long-time Netbeans user, I'm most curious about that IDE's future. It's gained a lot of traction but is/will probably never be as popular as Eclipse. Netbeans is a great IDE. JDeveloper is also a really good IDE. So, Oracle - aside from supporting Eclipse in many ways- has 2 really good, though not heavily used (compared to Eclipse), IDEs which are free. I could be wrong, but I can't really see Netbeans as having a great future with Oracle. I'd love to see them spin it off.
  49. Re: Netbeans future[ Go to top ]

    Michael, I too think that NetBeans is very valuable to the community. With all due respect to the sensibilities of the JDeveloper team, I think given the current deployment numbers and traction of the open source development model, I think the right choice for Oracle would be to give more emphasis to NetBeans if it really came a time to make a decision either way. For one thing, it would show an unmistakable commitment to the community and openness. Just my two cents, Reza
  50. RE: Netbeans future[ Go to top ]

    I think we made it clear that we are going to keep NetBeans - read this: http://www.oracle.com/technology/tech/java/htdocs/javatoolsfaq.html and watch this: http://oracle.com.edgesuite.net/ivt/4000/8104/9236/12627/lobby_external_flash_clean_480x360/default.htm One key distinction between JDeveloper and NetBeans/Eclipse is that JDeveloper is more than just a Java IDE - it is also our SOA development environment, and WebCenter/Portal development environment - this is in addition to the DB/XML/Web Services capabilities. It is also the development environment for Oracle ADF - so overall you can say that JDeveloper is the development tool for the Oracle Fusion Middleware, while NetBeans/Eclipse and JDeveloper are possible development tools for the Java parts - you are free to choose the one that works for you, and we'll make sure it works the best with the Oracle Java EE platform.
  51. Re: RE: Netbeans future[ Go to top ]

    Shay, I did look at this presentation and frankly it was much better than some of the other ones because it came cross as technically focused, rather than managerial/marketing focused :-). Again, no offense, I realize at some level it depends on unique personality of the individual in any given role as to how something comes across rather what the reality is... Cheers, Reza
  52. Re: RE: Netbeans future[ Go to top ]

    Shay, Unless Oracle plans on adding the “fusion middleware” capabilities to Netbeans and Eclipse we run the chance of having to run at least two IDE’s? If we use JDeveloper for everything we would be able to do the "java parts" and the "fusion middleware". Is this assessment correct? -Serge-
  53. RE: Re: RE: Netbeans future[ Go to top ]

    If we use JDeveloper for everything we would be able to do the "java parts" and the "fusion middleware". Is this assessment correct?

    -Serge-
    If you'll use JDeveloper you'll be able to have one IDE that does both the Java EE and the rest of the stuff.
  54. Re: RE: Netbeans future[ Go to top ]

    well - if you have more than 1 portal products, having more than 1 IDE, i guess is acceptable :)