Lack of free web based java apps – a conversation at JavaLobby

Some interesting posts in this discussion on the lack of free Java web apps.

Some point to the lack of web hosts that provide Java services. I think Kattare, the host I am using, is great and I recommend them, but this is correct – there must be many more then there is today. Sun would do Java a world of good if it evangelized to web hosting providers and made it easy for them to provide basic services.

Some mention a difference in approach between the Perl/PHP/Python folks and the Java folks. Supposedly Java developers get overconcerned with archetecture and forget the real task is to get the job done. There’s some truth to that. I’ve run into many developers like this. I’ve been accused on occasion πŸ™‚

Eventually it’s going to be projects like Roller and miniblog that will change people’s attitudes.

7 thoughts on “Lack of free web based java apps – a conversation at JavaLobby

  1. Damn, I can’t decide if those last two sentences are a slam against Roller or not! I know one of Dave’s goals was to build a showcase app, and that he wanted to demonstrate some best practices. As is always the case, there are places I would have done it differently: not better necessarily, just different. That doesn’t mean I think Dave did it “wrong”. And we are actively looking at ways to improve the design and implementation (including a switchable persistence layer).

  2. Not a slam πŸ™‚ A compliment! I like what Dave’s doing with Roller. I don’t think it’s overengineered. I’ll think that when the EJBs start flying! And I think Russel Beattie’s approach is great too πŸ™‚

    The proof is in the pudding. Do people like the end product? And Roller and miniblog pass that test.

    We developers can argue archetecture till the end of the world. It’s in our nature. But it’s that sense of perfection I think some of us see in Java that drives a few of us batty! We may start to chase our tails instead of the real end goal – to put out a product that satisfies users.

    It’s kinda like songwriting. You can songwrite to impress other songwriters – using technique to win people over – or you can songwrite to impress listeners – and kick some ass.

    I should add – the greatest achievements are those that score a balance between both πŸ™‚

    I am a hypocrite when it comes to this. I sweat archetecture way, way too much. It can keep me from being more productive. I gotta change my evil ways…

    But on the flip side of that, software I’ve written is still being used by my former employeers and there is alot of satisfaction in that.

  3. Well, while we’re confessing, I have to say that by-and-large I throw “perfection” out the window and go for KISS and “make it work.” I save perfection for refactorings (most of which never come). This has had the detrimental effect of making refactorings extremely difficult – I’ve been left to stew in the juices of my own making more than once (uh, mixing metaphors?).
    I’m trying to reform (somewhat) and at least put more consideration into common patterns and best practices *before* diving into the code πŸ˜‰

  4. I think Java folks have forgotten about the first part of Perl’s slogan: “Make the easy things easy, and the hard things possible.” mod_perl folks are in the same boat, though they at least dangle a carrot in front of service providers by making CGI scripts scream.

    That seems to be the way in for the three P’s (Perl, PHP, and Python) — let people try it out under CGI, then when they’re hooked, show them how it can be brought in as a module for high performance and tweakability. Although it may sound like heresy, is there any way to try out server-side Java under more pedestrian CGI support?

  5. Yeah, well, whatever. The point is that ASP is free with IIS, and PHP is drop-dead easy to build into Apache or can be run as a CGI if you’re feeling skittish. Java and JSP need that kind of gentle installation curve to be acceptable to service providers.

Comments are closed.