Slashdot Mirror
Oracle Outlines Plans for Sun Products, Casts Doubt on NetBeans
An anonymous reader writes to tell us that a recent FAQ released by Oracle outlines the plans for many of Sun's popular products like GlassFish, MySQL, and NetBeans. Many are worried at some of the possible avenues the decisions outlined could lead to, especially with respect to NetBeans. "What should have happened, Oracle should not have missed a beat and should have announced work on Oracle plugins for NetBeans and active Oracle support of NetBeans. This type of announcement would have brought a large and some-what skeptical NetBeans community much closer to Oracle. It would have been a big win for Oracle. NetBeans will continue to grow either way - but Oracle has missed a big chance to really change perceptions and at the same time move their tools to another level. What JDeveloper lacks is buzz, a wealth of community developed plugins, a wealth of support for other languages and a very, very large community. And of course it does not offer a platform in the NetBeans and Eclipse sense of the word. This is a huge missed opportunity for Oracle."
I wasn't aware anyone seriously used it. I used it for school and I've been on Eclipse since I started doing real projects.
Whale
http://www.infoq.com/news/2009/11/oracle-sun-palns
I don't know what palns are or why Oracle/Sun thinks they are important, but ...
Unfortunately for Netbeans zealots, it has never caught up with Eclipse. It may have surpassed it temporarily for certain apps (think Grails support - but look at STS 2.2.0). It's also not as good as IntelliJ IDEA (previously, always non-free).
Yes, both Netbeans and Eclipse are also RCP platforms, but how many real Netbeans platform apps are there? (The Nokia one on the web site is vapourware - yes it shows a real customer RAN - without their permission, I should add! - but it's never been a product delivered to customers). Real Eclipse RCP apps do exist (XMind, Lotus Smartsuite...). Realistically, they both over good RCP platforms (one pure Java, one SWT) but Oracle won't really care about that.
As for JDeveloper - well it's a typical Oracle product - if you're in an Oracle house, it's pretty good, but no, it's not got a large userbase or community supporting it.
Oracle should let Netbeans drift off into open source land. Perhaps it'll thrive? I don't know. JDeveloper's functionality should be ported to Eclipse (along with SQL Developer, while we're at it).
Oracle are great at giving you tools once you've signed up for the ride, and why not rebase your products on the best? Which in my opinion is Eclipse.
Bus error in your favour. Collect 200kB
If Oracle is not interested, other people/companies will carry on the development. In general Sun customers should be applauding the foresight of the company to make pretty much every peace of their hardware and software Open Source and compare their situation to that of Peoplesoft or Siebel customers. Even if everything Sun is killed off tomorrow, it would still be possible to manufacture Sparc-based servers running Solaris and with applications developed using Java and Netbeans.
...if you want to interact with Oracle products. I tried really hard to use it, even using it as both a Java IDE and a PL/SQL IDE and, while yes, it does work, I found it too slow and clunky to just "bang out some code" when you need to write up a throwaway program really really fast.
But, like I said, if you want total interaction with your database or app server (assuming that app server is oc4j), then I suppose, if you have to use only a single tool, I guess, well, shrug, I guess it's better than nothing...I guess.
MySQL matters. NetBeans, not so much. Most of the web runs on MySQL. There aren't that many good open-source alternatives. (Oracle owns BerkeleyDB, too.) PostgreSQL is about it, and because that's Berkeley-licensed code, not GPL, it can be forked and the open version abandoned.
Oracle has to dump something. I'm surprised they kept the SPARC line alive. It just doesn't seem to be necessary any more, and it was a money drain for Sun.
TFA is quoting Gartner. When is the last time Gartner got something right? It's full of weasel words. Lots of "If ..."
Read what Oracle wrote. They're not abandoning NetBeans.
Fuck Gartner. Fuck them in the heart.
Back in October, JetBrains announced that they were making Idea 'Community Edition' open source, covered by the Apache 2.0 license.
Caution: Do not stare into laser with remaining eye.
It looks like someone didn't bother to check the NetBeans platform application showcase where there are so many platform apps that they had to categorize them:
http://platform.netbeans.org/screenshots.html
I hate to belabor the obvious here, but Oracle is not terribly concerned with what developers think about them. There are two reasons companies buy Oracle licenses: they either absolutely have to have them, or someone much further up the chain than the developers -- at least in most companies -- thinks that they do. From the altitude in the org chart where those decisions are made, there's no difference between us and the janitors.
Proud member of the Weirdo-American community.
My personal opinion is that Oracle is very dedicated to the entire Eclipse ecosystem as well as to JDeveloper. It's about choice. There is an entire free download product that is continually being enhanced called the Oracle Enterprise Pack for Eclipse (Oh-Pee is how we say it within Oracle). In fact I believe it was one of the first, if not the first commercial IDE to support the latest Eclipse 3.5 Galileo. http://www.oracle.com/technology/products/enterprise-pack-for-eclipse/index.html OEPE is targeted for Java and JEE developers and is mostly about supporting the Java standards. Additionally, the majority of the TopLink code was donated as the EclipseLink project and is currently the JPA reference implementation. Just take a look at the presence has at the next Eclipse conference and I think you will see that Oracle is committed to Eclipse. http://www.eclipse.org/eclipselink/
When you get into the "upper-stack" components like SOA Suite for integration and WebCenter Suite for enterprise portal development, and Oracle's Application Development Framework (ADF) that Oracle strongly recommends JDeveloper. Those products have been based on JDeveloper for a long time and the user-experience developing for those products is extremely smooth because Oracle can influence everything about the IDE. If you want to do Java and JEE development in JDeveloper, you can do that too. It's your choice.
IMO both Netbeans and Eclipse are a waste of time. What is the point of an IDE that can't even get basic text editing UI right?
If you ask me all the IDEs are getting bloated to the point where I think in 2-3 years you're going to need a 64 Bit environment with 8GB or more of RAM just to develop. (Where I am we find Weblogic development using Eclipse is getting slow on 32bit machines with 2GB RAM is getting...difficult) There is some attempt to address this with each IDE by making the platform extensible, and component/plugin based. Unfortunately to do even basic things you end up finding yourself stuck needing a long list of plugins, so whilst this is theoretically a fantastic move, in practice you're still left with a bloated environment. If you need support for multiple languages or environments you find the plugin architecture is no help at all for preventing bloat. (It's still great for adding features).
These posts express my own personal views, not those of my employer
Does that mean Sun will now stop reneging on their promise to open source the new Java plugin? http://blogs.sun.com/darcy/entry/openjdk_and_the_new_plugin
Do you know what the high end Sparc machines can do?
I am sure that the terminology does not even exist in Intel-AMD processors, because they simply can't scale in the same way. You would have to look perhaps at IBM or HP.
Certainly an SPARC desktop will be soon a thing of the past, but in the high end arena SPARC can't be touched.
IANAL but write like a drunk one.
Early versions of NetBeans really, really sucked. It has come a long away. But I think it was too long getting there and Eclipse and other IDEs got mindshare and market share. Right now there are virtually zero Oracle customers who are going to defect to someone else because they don't have NetBeans support. There are some that want Eclipse very badly (mostly those that are not pure Oracle shops but use WebSphere too). So it's a no-brainer for them. There's no payback to them supporting 3 IDEs (NetBeans and Eclipse and JDeveloper, until/unless they converge).
But, you do realize that something that is BSD licensed can also be forked into a GPL version? Simply keep the BSD license notifications and ALL future additions to the fork are licensed GPL/LGPL, GLPv3 etc. You now have a GPL fork. So, if someone (who by the way?) abandons the BSD licensed version and begins solely working on a proprietary fork, the rest of the community can simply take the last BSD licensed version, create a GPL fork, and lock-out future proprietary forks and prevent the proprietary company from using the now GPL contributions to the GPL fork in their proprietary product.
Over-the-top Response Guy! Giving "Over-the-Top Responses" since 1970.
I agree with everything you said except the 'speed' part.
Eclipse in my experience is consistently faster than Netbeans (at least for small projects, any big project I do in Netbeans).
Yeah, NetBeans is not used all that much. Eclipse is used a lot more in the real world where consultants get paid more if they spend longer doing basic project/environment setup and waiting for the UI to come back.
Of course for anyone who wants a quick and feature rich IDE, you can't go past newer versions of NetBeans. I am thankful that more recently I have been in a position to say "I am using NetBeans, even if all you idiots settled on Eclipse" and it has worked out quite well.
Of course, Eclipse will drop off in popularity, developers will look at critisism and improve it and in another 5 years Eclipse will be the better IDE with smaller market share.
JDeveloper should be dropped, IMO. It doesn't offer any real advantage over anything else out there, and suffers in comparison to NetBeans and Eclipse for most tasks. Oracle could be missing a great opportunity here...
I don't therefore I'm not.
Seems zembly is the first victim...
http://zembly.com/static/suspend/index.html
Oracle is internally, as I understand it, an eclipse shop. They always have been. They've got Jbuilder for some specific applications, but for the most part it's eclipse all the way. Sun has tried a number of times to change this, but it's never worked, Oracle just doesn't like NetBeans much. When you combine that with the fact that it's not a hugely popular product(it's a lot more limited and a lot less powerful than eclipse) it's not really a huge shocker they're not going to pour resources into it. They're already working heavily with the eclipse project(including donating a rather large chunk of source code to them) and of course JBuilder as well. Maintaining a third IDE which they don't believe in and don't seem to particularly like and which isn't tremendously popular wouldn't really be terribly sensible of them.
On the plus side for all you NetBeans fans out there, it's all open source and it's all written in Java so you can all get together and maintain it yourself(if you're using NetBeans you're almost certainly a Java developer so there's no excuse). If it's not worth it to you, learn to use Eclipse like everyone else. It's a bitch to set up, but it's incredibly powerful.