DISQUS

Thinking inside a bigger box: Drinking from the Java firehose: A manager's primer to Java projects

  • Øyvind Wergeland · 2 years ago
    A good historical perspective, Johannes, but I lack suggestions for improvement! It's really not sufficient to say that .Net is not any better. Some may even decide .Net is better, since it's completely controlled by one company.
  • Johannes Brodwall · 2 years ago
    Hi, Øyvind. Good point. Maybe suggestions for improvement could be a good subject for a new article. Maybe you could even write it. :-)
  • Geir Hedemark · 2 years ago
    Good article, Johannes. Keep up the good work.

    I halfway agree with Øyvind. The article would be "complete" in a totally bogus aesthetic sense if it also tried to give pointers to what is happening today. There sure is enough to choose from: AOP, testing moving from unit testing to subsystem-style tests (?), the JCP dying, "developer" roles starting to become less technology oriented and more people-interaction oriented in some ways (and not in others, since there is such a bewildering amount of technologies and frameworks to keep up with) ...

    Write it myself? Hum, I already have three projects that aren't moving.
  • Johannes Brodwall · 2 years ago
    Alright, I hear you. I'll add a follow-up of this article to my backlog. But you guys better read it!
  • Kukenspeil · 2 years ago
    J2EE was horrible from the start. It's a great example of what happens all too often in software development. Good solutions exist, someone ignores all the hard lessons learned making it and goes ahead and makes something new that is full of problems.

    Look at CORBA - don't laugh yet - especially if you don't have any real experience with 2.x and POA. Sure, it had several problems, but look how J2EE basically took all that was bad from it, removed all the powerful extensibility, and everyone thought it was something brand new and "Incredible! Look how I can call remote methods!" Woo hoo. And then came SOAP, a horribly broken standard - SOAP had serious technical shortcomings, but, as a market strategy, it was a masterstroke. Again, people with no real understand of distributed systems working in areas they have no business doing.

    It happens all the time. Something works. Then someone re-invents it without looking at the previous work and ends up with something that really could have been great, but turned out to be garbage in the end (and lots of disastorous projects based on it).