Response: Is Java a Sinking Ship?
May 29th, 2005I noticed that my blog entry on the future of Java drew a response from in Limbo. I would like to see more thoughtful discussion on the topic. Below is my comment back to his blog entry.
I am sure Java will be around a while, but the question between Java and whatever MS was pushing with their cute visual interfaces came down to a single argument. If you build it in Java it will take longer to build but will be more robust in the long run. And if you build it in the MS technology it will be up and runny quickly but will not hold up well over time. Essentially MS solutions were seen as a one-time use application which you replaced instead of upgraded.
That argument is not as true now that VB 6 is becoming unsupported and the .NET solutions are getting more and more robust. It is also less true because the tools for Java are much better now, such as Eclipse or IDEA. It is clear both sides have worked to minimize their weaker points.
With the Java shops I worked with I saw a sort of culture growing. People calling themselves architects would build a monument of infrastructure to the lords of Java. They did it in a way to show how clever they could be. And these infrastructure projects spawned many web and desktop frameworks (Struts, SWT, etc) which competed with the standard solutions in the JDK. It clearly fragmented the community.
My weariness with Java stems from going down one path over and over which ends up being a dead end. I chose Java with JSP and JSLT and eventually got burned with it being dead ended by JSF. And the same would be true with Swing vs SWT.
Having 10 projects which have the same goal simply dilutes the talent in the Java pool while the .NET community focuses on a more clear path set forth by MS. I personally want MS to experience some more fierce competition so they cannot abuse us (the comsumers) so badly. If it does not come from the Sun camp, maybe it will come from other places. IBM seem to be putting a lot of weight behind PHP, MySQL and Linux while Ruby on Rails is draining a lot of brain power from the Java community.
All it seems to me is that MS has logically continued their .NET framework to the 2.0 release with many productive improvements while the competing technologies continue to fall victim to infighting. Perl was once the leader in the web application arena and has suffered from infighting. Is it the leader now? Will the Java community learn from mistakes made with Perl?
May 30th, 2005 at 7:23 pm
I don't see where you backup any of your arguments.
"Ruby on Rails is draining a lot of brain power from the Java community", really?
Sun predicts that the Java developer pool will grow from 3 million developers, now to 10 million developers over the next couple of years.
"IBM seems to be putting a lot of weight behind PHP"
IBM is pretty much a Java consulting firm. They have put out products like Websphere, Eclipse, JikesRVM, Jikes. They may like PHP, but Java is their mainstay.
I will agree with the .NET comments, Java is basically competing with .NET, but if you want to develop around Microsoft, that is fine. If you want to develop around something else, you use Java.
I don't think the Java community is fragmenting. It is just growing, so you have different technologies being leveraged by different sets of groups.
My only problem with the Java Framework Cheerleaders, there seems to be a lot of "My framework is better than yours". If you use Struts, JSF, Spring; Fine. But, I just hate to tell a company, "Struts Sucks", you need to use Spring, now. This is expensive because you have to rewrite your existing software over something that may or may not provide benefits.
May 31st, 2005 at 3:03 am
Also, there are stuff like Trails going on. Is it draining a lot of brainpower from Ruby on Rails?
May 31st, 2005 at 10:40 am
I do see the Java brain drain due to Ruby on Rails. One great Java developer, Bruce Tate, has moved over to Ruby on Rails and recruited many others from the Java community. He speaks at the No Fluff, Just Stuff symposiums which was just Java a couple of years ago and now it includes .NET and other technologies.
As for IBM, they are coming to grips with the new relationship Sun and Microsoft are developing. They realize they will not be able to co-opt Java like they had planned and are now going to head to PHP and work to be leader in that area.
And for whatever Sun projects for growth, I cannot speak to that with any amount of optimism. If Java developers become so commonplace then the average salary of a Java developer will be comparable to your average accountant. Sun is also inflating those numbers. With the way software development is evolving it will not be necessary to have all of those developers. The tools are allowing individuals to do more work and get more done. The current need will shrink simply because 1 person in 5 years will be able to do the job of 5 people. I never believe those growth numbers.
And for Trails, that may be worth looking into for some people. But essentially I want Sun to choose a web framework and build around it. There were portlets but there seems to be no energy behind it. And I have no idea which web frameworks are compatible with it. In the .NET world it is very clear what you can do with all of the components. And you can make good money producing a few components and selling them.