Hi Josh,
I'm starting migrating some projects to maven and personally I think
that maven is the way to go because of three things (the 3 selling
arguments I've used at work ;-) :
1. we can locate a project or a project module (an artifact in maven
terminology) in space and time, i.e. we specify a maven coordinate (
repository url, groupId, artifactId, version) to what we are producing
and as a consequence we're able to reference it in other projects.
Previously to maven copying jars around was a really nightmare.
2. The POM (Project Object Model). We specify how a project looks like
and let the plugins do the work. The nice thing about it is that the
plugins are artifacts themselves and can be therefore easily located,
downloaded and used.
3. Standard directory layout. Previously to maven we had a hard time
trying to find a mutual consent on how the project structure should look
like. The problem was that different groups work on different projects
and we were not sure which layout would best fit our needs. Having a
proposed layout from maven was at least a very good starting point.
Cheers
Alex
Josh Long wrote:
Hi everyone,
I'm trying to make a case for Maven and I'm going to need to provide a
better reason than "it's better than a kick in the face!" to managment. I'm
at a bank, so everything is subject to heavy scrutiny. Essentially, our
project hardly compiles and is so untestable that upon check out no less
than 5 files need to change to get it to the point where we can compile it
(once we've created a working project in either IntelliJ or Eclipse, which
has hitherto never been too successful. Some of us just ant deploy every
change instead of iteratively deploying using our IDE's weblogic facilities.
So basically, I know I could solve the file issues with Maven profiles. I
know that using the Maven site plugin and reports like PMD and JUnit and so
on I could provide great "dashboard" like functionality into our
application. I know that I can solve the broken project descriptors, too.
All with Maven, but strictly speaking these are technically still "process"
enhancements, which come down as a liability.
In terms of shear "resource" hours, I should imagine 10 hours or so to get
our two projects moved over and acheive parity with our current Ant script
and even perhaps to solve all the Eclipse/IntelliJ nonsense and get decent,
default mvn site generation, and to change our existing production support
script which is Ant to interface with the ant script Maven will generate for
us. Basically, it won't take a lot. But that's not enough.
How do I make this case in the face of so hostile a mentality? Have you ever
had to make the case? Any insight on how to move forward would be
appreciated.
Thanks,
Joshua Long
Sun Certified Java Programmer
http://www.joshlong.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
