Hi Ulrich, thanks for replying. My response will be a little longer here, but I think you've asked a really important question, so perhaps it deserves a good answer.
I have personal experience implementing sites with Stuts2 and MyFaces. I did look at Click, and I have some experience with Sling, in the CQ-Jackrabbit world. Frankly, it would have been significantly easier to use existing technology than to implement yet-another-framework. However, I felt that the needs I had identified were best served with something new. Two of the needs I had identified where simplicity and maintainability. That's not to say that the existing frameworks are not simple or not maintainable; they just were not so in the terms that I needed. The sites I was working with were primarily UI code (JS, HTML5, CSS3), and needed minimal server-side support. That is, they didn't need all the generality and flexibility of a general purpose framework; they simply needed to be able to post a form, consume a JSON request and produce a JSON response, etc. The target audience is not J2EE programmers; it's UI programmers, or "browser programmers". An primary concern was the need for extreme simplicity. I wanted to be able to turn over applications to junior programmers, interns, or even UI programmers with minimal J2EE skills, to maintain. So, for example a Pragmatach app uses maven exclusively to build and produces a war file, rather than needing to run inside the context of server software (Play) or build with sbt. It's not hard to find juniors who can use maven, and deploy a war file. Additionally, I did heavily borrow ideas from other frameworks; the use of annotations from JSF and Struts2 for example. I wanted to be able to fully specify a controller in a single .java file, so rather than a "routes file" (Grails, Play), so I chose to specify routes as controller annotations and then sort them programmatically to produce a routes file. Once applications go live, they have to be operated. I felt that this was a missing aspect; there was a tendency to just build out applications and then hand them over to deploy and run, without providing in-application support to operate them in the real world. For that reason I included the Pragmatach runtime console, and built-in JMX support. I also wrote a statsd plugin; that technology is becoming quite popular. Finally, with respect to Click and Wicket, which are both very cool, I felt that they were trying to be full-featured frameworks which addressed both client-side and server-side needs. I was specifically trying to stay away from the client side, in order to impose no constraints whatever on the target audience; client-side programmers. I tried to provide strong server-side JSON and XML support, knowing that client-side programmers prefer to communicate with servers via Ajax & JSON, but I didn't want to go as far as binding data to widgets, for example. On Tue, May 7, 2013 at 3:14 AM, Ulrich Stärk <u...@apache.org> wrote: > The ASF is home to a number of web frameworks (e.g. Struts, Wicket, > Tapestry, Sling, MyFaces, Click, > ...). Have you thought of contributing to one of those instead of creating > a completely new > framework? What was the outcome? Are there cross-pollination > possibilities? If your goals match with > the goals of one of these projects, it could be easy to find a champion > for you from one of these > communities. > > Uli > > On 07.05.2013 04:01, Tom Everett wrote: > > Thank-you. I am looking for both champion and mentors, for this project: > > > > > http://blog.khubla.com/wp-content/uploads/2013/05/pragmatach-incubator-proposal.pdf > > > > I hope that a pdf document is acceptable. The web site is here > > > > http://www.pragmatach.com > > > > If you're interested, there is a web-based presentation here: > > > > http://pragmatach.com/slideshow/index.html > > > > If there is any more information I can supply that might be of interest > > please let me know. > > > > > > > > > > > > On Mon, May 6, 2013 at 7:29 PM, Suresh Marru <sma...@apache.org> wrote: > > > >> Yes this is the right list for soliciting champion and mentors. > >> > >> Suresh > >> > >> On May 6, 2013, at 7:04 PM, Tom Everett <t...@khubla.com> wrote: > >> > >>> Good afternoon. > >>> > >>> I have a project which I am hoping to submit to the incubator, and I > >>> understand from the documentation that it is recommended to identify a > >>> champion prior to the incubator application. Is this mailing list an > >>> appropriate place to solicit for such a person? > >>> > >>> > >>> > >>> -- > >>> A better world shall emerge based on faith and understanding - Douglas > >>> MacArthur > >> > >> > >> --------------------------------------------------------------------- > >> To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org > >> For additional commands, e-mail: general-h...@incubator.apache.org > >> > >> > > > > > -- A better world shall emerge based on faith and understanding - Douglas MacArthur
