What, you say, not another framework!?! The world doesn't need more of those!

Well, I respond: I have been not-writing-my-own-framework for years now. Frankly, I deserve a break from not writing frameworks.

But mostly I like Webware. I think it has a lot of good aspects. So I wrote something like it, but newer. Webware's basic model is good:

• Class hierarchy (the SitePage often used in Webware) is a nice way to share code and build application-specific mini-frameworks.
• The simple one-file-per-screen approach is easy to use, and easy to disect from the outside.
• It emphasizes UI first, and abstract data modeling second.

It's not the only framework like this. But anyway, I still like it. Still, there's lots of specific things that annoy me. It's an old framework, from the Python 1.5 days. Python has changed a lot since then. But Webware is still using getter and setter functions, and it has lots of cruft, and internally it has more hierarchy and abstraction than it needs -- most of it not used for anything. It's also missing a few things I find really important -- things I added with Component, but that would be better in the core.

The result is that I have often found myself having to apologize for crufiness in Webware. Or explaining the history of its decisions (most of which were made well before I started using it). I sometimes find myself typing more than I want (like self.request().environ().get('HTTP_HOST')).

So... in comes Wareweb, wherein I try to distill the parts I like in Webware, and avoid the parts I don't like or don't think are important. I'm actually still using Webware for all my work projects (I'm now running it under Paste, but Webware all the same), and this is more experimental than anything (with only one real app). Which is why it has no real web page, and not much documentation (but I think it doesn't need much documentation). But I like it, it's satisfying.

So, if anyone is curious about what a framework written by me looks like, there you go.