It's been cool, lately, to debate the various parts of Struts, WebWork and other web MVC frameworks.
I've recently played with Struts and it's not exactly making me happy. I'm a big fan of using regular expressions to take synthetic URLs and remap them to actual URLs, changing parts of the path to query parameters. Struts simply does not play friendly with that concept. It must work in terms of static URL patterns instead of Regexps. Since it's ignorant of URL rewriting, I cannot submit forms to URLs that I know are valid (or will be after rewriting), and thus Struts forces me into the world of Ugly URLs.
So, I've looked at WebWork. I finally understand the concept of WebWork, since it's exactly the same concept that Jelly uses. A WebWork Action is like a Jelly Tag. Something pulls your tag, instantiates it, calls some settors, and then executes it. Possibly then wash, rinse and repeat.
So, why am I not using WebWork? It's not mavenized. Too many dependencies that don't currently exist in the ibiblio maven repository. So, what would get me using WebWork? If someone would either Mavenize WebWork, or at least point me to a tarball that I could use to easily get all WebWork-based jars into the ibiblio repository with the correct structure.
Bottom line: if it's hard to use in a mavenized project, I won't use it.
Replying to myself because I can't find the 'comment' button on Mike's blog.
No, no, no. I wasn't saying that I will only use mavenized projects. But, if the projects want me to use them, it'd help a lot if they'd get their jars and ensure that their dependencies are in the maven repository.
Having your jar in the maven repo doesn't mean you have to use maven.
But, like CPAN, folks are going to be more like to use your stuff if it's versioned, labelled and available in a well-known location.
mavenized or not mavenized i use webwork for all my projects and i find it particularly interesting and superb and to me much better than struts.