The major issue is to become more open to other developers, projects and techniques.
One problem here is there are few other products that have enough interest in integration with MMBase - at least not enough to invest the time.Should a project show interest, then there will definitely be willingness to aid. Some Jakarta projects, such as Cocoon, are interesting enough to contemplate (we looked brtoiefly at Coccoon as an alternate template system).
Unfortunately mastering a new project, let alone bridging it to MMBase, takes quite a lot of time, which few people have lots of (else they would spend more time on, say, bugfizing or the optimization project).
The first thing is to accept the fact that people integrate with other products
I think there are very few people who think integrating is bad. What generally goes wrong though is that in the few cases where products are integrated, this is hardly ever shared with the community, either because of commercial reasons or because sharing requires investments people would rather not make (and I recently discovered that it is often not time or money, but the idea of having to enter discussion with - and risk the criticism of - other developers that is the big one here).
Helping people integrate with other products (answering questions) will have a positive effect on the community.
We generally do this, provided people ask questions or even simply inform us what they are doing, which with alarming frequency they don't.
The main issue here is that to get anything out of the community, you have to invest by taking initaitive and start communicating on the subjects you want to get done.
What I hear is people complainign that we are not doing the talking and thinking for them. Want versioning/workflow/sms-functionality? Start talking. Propose a project. Set up a framework and throw the idea on the devlist.
Do not wait until we have made some management paper (aka roadmap) on what we will do in the future, because, frankly, we are not very capable of doing this. In an OS project, you cannot plan anything as there are no set resources (time) available, and the community changes so demands may change too.
We can make a list fo what peopel want, but there is little us in it as we have already noticed that people don't listen anwyay.
Our 'roadmap' for last year was to run and finish the optimization project. This initiave has been posted various times, and I have been veruy explicit in the meetings this year (including the MMEvent), that there will be no 1.8 release until we finish the optimization project.
There have been almsot no offers for aid, with the exception of a small group of 'regulars' willing to do the performance sub-project.
One other thing I wonder about is not finding any samples or articles about integration with cocoon or struts. This is one of the major places to attract developers.
I looked very briefly in Cocoon and we had a talk with one of the project managers. It is likely posisble to combine these two projects, though I am unsure on the effectivity, as Cocoon relies very heavily on xsl transformations, which was exactly what slows down the editwizards in MMBase. I had the impression though that the people at Cocoon migth be willing to join forces. You would need a sample project for it to be worth the investment though.
People already do things with struts. There is little extra needed to 'integrate'. However, I've found struts to be overly complex, and mixing both techniques (MMBase and struts) in one applications risks making it totally unemanageable. I know, as I've been working several weeks on trying to get a struts-based site working.
Another thing is to look at and implement important java and other standards. This will also attract other developers.
The only problem here is backward compatibility. Switching to 'new' standards may break older applications. This is oftentimes somthing that you may want to risk, but which you still need to communicate and agree on.
For instance, we recently switched to jstl. We migth want to think about using the new JAXP (which I hear supports xpath). Perhaps we meant to shift to Java 1.5 and use it's nifty features.
We sdtill nee dfto agree on what and when we switch to a new technology or standard though. And then again we need to communicate.
If you want soemthing new or something different, you merely need to start communicating. In general, you will find little resistance against new and solid ideas. We will even give you a conciderable reign in implementing your ideas. But you have to start talking.
We can't read minds.
-- Pierre van Rooden Mediapark, C 107 tel. +31 (0)35 6772815 "Anything worth doing is worth overdoing."
_______________________________________________ Developers mailing list [email protected] http://lists.mmbase.org/mailman/listinfo/developers
