Dmitri Colebatch wrote: > > > Ahh yes... and here's the real cruncher. OK, I'm more than happy for > someone to say "framework X is more appropriate for my requirements than > EJB", but saying "thats too hard to set up, and its cheaper to roll my own" > is hard to see being entirely accurate. Writing your own framework (or > worse still not using a framework) means you will have more lines of code to > maintain, and hence more bugs to fix. Sure, EJB containers have bugs, but > at least its not your responsibility to fix them. >
I have worked with companies that are willing to roll out immediate bug fixes to their products when it is obviously apparent that it is broken. Those are few and far between. I have also worked with companies that not only refuse to admit that their code does not have a bug, they will refuse to look at anything that demonstrates it. And sometimes this extends to how they interpret a standard (their interpretation is correct and all of the other vendors interpreted it wrong.) So in that case they will not fix it, even when it can be demonstrated. So if it doesn't work, and I need to work, then I am going to be coding a solution anyways. So I end up writing the code that I would have in the first place. And I lose the time spent on the first implementation. I realize that in general the "not invented here" view is not a great one. However, J2ee and EJBs are not simple systems to implement. So they will have bugs. Especially with constantly evolving standards and reworked systems. Contrast that with JDBC where I have never used any feature that was not in JDBC 1.0. (And I was personnally told that they considered JDBC "done" with version 3.0 which was finalized something like 2 years ago.) =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff EJB-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
