Joe Pluta wrote:A forum is a web application, but not a business application. The primary focus is the front end, it's more about presentation of textual data, and the business rules are almost trivial... simple enough to be written in languages like Perl and PHP. Yeah, they'd be sizzle-y, I suppose, but in the end they wouldn't prove much.
I suspect it's because the EGL Cafe isn't really a business application - it's a forum. EGL is designed to write business applications. Sort of the same reason midrange.com doesn't run on a midrange...
A forum is web application to access a database of messages (obviously this is greatly over simplified). I _think_ the software they are using (written by Jive Software, same company that writes the open source jabber server I use) uses MySQL as it's back end database.
And, FWIW, if I had the time and inclination ... I suspect I could write a list processor and archiving system that ran on an i.You could but I bet you won't. It's not worth the ROI. Similarly, it's not worth the time for the EGL team to re-implement something that's been done to death.
midrange.com runs on Linux because there I had the hardware to run it on and software was available to do exactly what I need.Exactly. Right tool for the right job. Forums are commodity software, nothing is to be learned by re-inventing that particular wheel. Besides, unless they did one that's better than everything out there, there is a certain segment of the community that would spend all their time saying things like "it doesn't have feature X that phpbb has".
My point is: It would be a great proof of concept (and validation of utility) if the EGL Cafe were written in EGL.I have to disagree. Having EGL run a forum would appeal to you and I, but not to people who actually but enterprise software. I'd wager that a forum ranks very low on the list of applications most C-level executives want their programmers to work on. Personally, I'd much rather that IBM do exactly what it is doing: spend their resources using EGL to integrate technologies and use those to build real applications. Even the simple scheduler that Chris and I wrote has more business logic in it than most forum software, and that's where I want IBM to focus.