This post originated from an RSS feed registered with Java Buzz by Weiqi Gao. Original Post: NetBeans vs. Eclipse---Why Not Both? Feed Title: Weiqi Gao's Weblog Feed URL: http://www.weiqigao.com/blog/rss.xml Feed Description: Sharing My Experience... Latest Java Buzz Posts Latest Java Buzz Posts by Weiqi Gao Latest Posts From Weiqi Gao's Weblog

The Eclipse vs. NetBeans flame war is heating up again, mostly because Sun has began a switcher blitz that included a fair amount of Eclipse bashing.

This can be looked upon from several different angles.

On the stages of winning scale:

First they ignore you,
then they laugh at you,
then they fight you, <-- you are here, Eclipse
then you win.

I'm a very happy IntelliJ IDEA user. But, there are things that IDEA doesn't do, like supporting C++ projects. I recently installed Eclipse (and started using it productively) at work just to take advantage of the CDT and the AnyEdit plugin. Thanks go to Dan Troesser for doing the CDT talk at the OCI internal C++ lunch, where we have Buffalo Wild Wings instead of pizza.

Eric Burke took another look at NetBeans:

I do plan to spend some more time with NetBeans, primarily to use the profiler. I may try to load it at work and run the profiler on a really huge project. That would be wonderful, but I'll probably continue my day-to-day work in IDEA.

You see, with IDEs, especially open source ones, it doesn't have to be either-or. You can have both installed and choose to use whichever one that will do the job at hand best. (I use both vi and Emacs. I use Emacs to edit my ChangeLog in its changelog-mode and vi the edit my README where I use the !!date command liberally.)

Obviously both the Eclipse and the NetBeans camp want to have more users. My general feeling is that there are way more Eclipse users than NetBeans users. The reason for this may be historical, but the implications are not. It's good that NetBeans 4.1 will be a lot better earlier versions, if you believe the NetBeans developers, and I do, but the fact remains that NetBeans 3, up to 3.6, sucked. NetBeans people need to realize that it is not enough to simply say "Hey, 4.1 is a lot better than 3.6. Come back!" Why should I believe you?

(By the way, I had the same experience with Ruby. The first Ruby book (the PickAxe) was so bad it left a bad taste in my mouth. And I'm not going back to it just because the authors say the second edition is much better.)

They need to become credible, as in delivering updates that address user issues version after version in short time intervals, even with a minority market share. They also need to stop bashing Eclipse because doing so gets them nowhere with current Eclipse users.

One thing that the NetBeans people have been pounding at against the Eclipse people is SWT. "Swing is right and SWT is wrong," they say. To which I say they are right, most of the time, most of the time when Java was the language of choice for the product anyway. In this fight, NetBeans happens to be on the majority side. By that I mean, more Java GUI programmers are Swing programmers than SWT programmers. It will probably remain so for the next few years.

At the end of the day, it is not the GUI toolkit choice that decides whether an application is successful. It's the application itself. One could argue that the most successful GUI applications are written in neither Swing (LimeWire?) nor SWT (Azureus?), not even in Java.

So, Java people, let's write a killer GUI application that everyone loves to use and the developers love to maintain. I don't care what toolkit it is written in.

The value of SWT, as I see it, is its usefulness in situations where Swing is not available, such as in GCJ+Classpath. Not all Java programmers are going to buy into this platform though. Another role that SWT is playing is to drive Swing development in a direction that increases platform integration. This is similar in nature to the role C# played in bringing autoboxing and the for each loop into Java.

So what does all these leave us? Well, we are entering the world of twos. Just like the "vi vs. Emacs," and "GNOME vs. KDE" wars, we now have "Swing vs. SWT" and "Eclipse vs. NetBeans" wars.

And I say, "let them compete." If IBM wants to pour millions of dollars into SWT, go ahead. If Sun wants to pour millions of dollars into Swing, great!

We'll end up with something better whichever camp we are in.