Joseph Ottinger

Every month we're told again and again how Java is on its way out. A multibillion-dollar company tells us that, while hiring other large companies to say the same thing. One sad group of souls says it's because of Java's licensing, or the lack of features available in other languages or frameworks, and another wails that Java's too flexible, while another set says that Java's too slow.

It ain't so.

The oddest thing about all this, to me, is that Java is one of the best languages around, and all the yammering about it is an indication of its health. You don't eulogize for very long; you only discuss healing while the patient's still alive, while there's hope. Microsoft doesn't bother advertising how much better Windows is than XENIX or OS/9 or OS/2 - those are battles it's won. Instead, Microsoft reserves its fire for Linux and, dare I say it, Java.

There are a lot of obstacles in front of Java developers today. Most of them are philosophical in nature, and can be fixed by a little bit of logical thought and education; I find that in my own sphere of expertise, developers tend to be unaware of many critical aspects of J2EE programming, mistaking some of the component specifications as being representative of the entire J2EE spectrum (which I'm guilty of myself!), or simply riding on the assumptions offered to them by other similarly inexperienced coders. Some of the obstacles are simply based on outdated information (such as the "Java is slow" myth); others are based on advertising. Some of the issues are bullet-point related (such as the "Java needs templates" discussions), and even more are based on simple defeatism.

All this is fine, really. I don't mind that these issues and others even exist, because we're aware of them. Sun Tzu, in "The Art of War," said that you should know your enemy; these are our enemies. There are those who pick on this column and various Web sites for being negative when Java doesn't need it - they want positive analyses and happy success stories only; I think these Pollyannas need to wake up and smell the Java! You can't solve a problem you don't know about, and while I try to be constructive in nature, even those who are less constructive serve a very valuable purpose: they show you things that need to be improved. Even if the things they say are wrong, their data had to come from somewhere - perhaps that's an opportunity to improve the documentation or education process.

In fact, I'd go so far as to say that I distrust a purely glowing review. I've been really happy with a few products, to the point where I'm happy to agree that their warts are very minor - Borland's Optimizeit ServerTrace is one of these products. That doesn't mean the warts aren't there or can't be improved, and I offer feedback in an attempt to make every product everything it can be.

Java's no exception. In my last editorial, I asked where the components were (Vol. 8, issue 9), and I'm glad to say there were a number of vendors and programmers offering answers. I've been willing to critique Sun's management of Java along with many others, and here's the thing that shows Java's health: those critiques have been acknowledged and answered!

Sometimes the answer isn't what we want to hear, but the fact is that little of this truly falls on deaf ears.

Java's alright, and unless things drastically change, the continued growth of Java will go on, unstemmed. It's still fun, still useful, still worthwhile - and will be for a long, long time.