> But perhaps now is the time to discuss this, before we end up > with something not everyone is happy with. Perhaps a 'core2' > project, devoted to deciding these kind of thins, should be > initialized now?
I think it is a good to discuss it. It is a good idea to have a core design before we speculate about the implementation. I have worked with mmbase now for a year. I like the concept of mmbase and how it handles the disadvantages of the concept. I have been in the core a couple of times this year and saw a lot of duct tape to keep it together. When I had to fix something I was really scared to pull some tape off and put it somewhere else. At finalist, I am one of the mmbase "experts", but when someone asks me something core related I usually answer: "Well, it should do this, but I am not sure it does." It is probably very easy to make a good core design. We have a system that works so we know what it should do. We know which issues there are in the core at the moment. The components where the core has to interact with are better than a year ago. There are probably a dozen ideas how to build a new one. The only thing we have to keep in mind is that old websites will still run on the new core. That will be a challenge, but not impossible. The bridge interfaces and taglib are doing a good job in hiding the core. The scanpages and other stuff will give some headaches. Another challenge is to minimize the changes in the parts where custom implementations were possible (builders, modules, etc.). Nico ---------------------------------------------------------------------- Experience is what you get when you don't get what you wanted in the first place.
