Hi to all, > Pivot is cool, but it will never touch Flex and WPF. There's simply too much > money there and momentum to catch up. I agree completely with this, and personally I hope this in the future could happen, but also if not it wouldn't be a problem, we are here at Apache not for money, so in my opinion we have to (slowly) grow with the quality of Pivot (for the most part done by Greg and Todd), as many other Apache products have done over years.
Our main problem is the we are a small team that needs to grow. And of course also our base of users. For example, next month probably I'll have to speak of Pivot at a local Java User Group (a mail on this will follow), this could be another little channel. For the SWT port, personally I don't love too much them for many reasons (mainly the lack of deploy as Applets, the difficult skinnability, poor performances on some platforms, etc). So I'd prefer to NOT go to SWT, OR at least if necessary, SPLIT the wtk part (and the default skin for what it's possible, or better implement a new default skin for SWT) so one implementation for AWT (as currently) and maybe another for SWT ... and in the future maybe another for Java3D/OpenGL/other ... On the editors support I'd prefer another approach, as seen in many products: have standard tools that run outside development environment as standalone applications, so no problem supporting only a specific IDEs (and no integration by default, ok, but someone could do it later). As edge technology I'm also interested in trying to make Pivot run with Scala applications, and maybe also rewriting some parts of it directly in Scala (maybe could be deployed also with their tool, sbaz, currently they have something for Swing), but that's another story, for my mid/long term experiments (someone could be interested in this, who said). This could help having some other interest in Pivot from other users/developers ... And last, we need some killer-features, for real-world applications: for example I had little comments also from other developers on some ideas (probably not the best, Ok) ... do you remember more usage of annotations at compile time to save users from runtime checks. More important, we are focused to data-centric applications, so why not some tool to rapid prototype of CRUD applications with Pivot, already discussed (but in my opinion only a little, and for this I don't like the idea to depend on another framework like Griffon) ... etc. Another idea could be to improve Pivot way to load wtk files, but generated server-side (like JSP pages) and work like a browser drawing the GUI from those files. What do you think ? Bye, Sandro
