There's no consensus yet on what, precisely, we'll be working on, but I know I've a personal bias towards MiddleKit enhancements, DAVKit enhancements, and bringing UserKit out of its alpha-ness.

Feel free to contact me (Tripp Lilley, tripp+pycon-webware-sprint@perspex.com) directly about the sprint topics, or, better yet, post to webware-devel.

Note that, by design, sprints are "not intended for newbies". That doesn't mean newbies aren't welcome, it just means that our focus is going to be on enhancing the code, not teaching the code.

Now, with that said, I'm rusty on the code I expect that I'll personally spend at least some portion of my sprint time fleshing out and diagramming Webware's architecture, bringing up to date some of the earlier work by Chuck, Ian, and others.

I expect this exercise would be helpful to those who haven't yet peered deeply inside Webware's source, but would like to. So, uh, newbies are, in fact, welcome to this particular sprint

-- TrippLilley

I'm don't really use MiddleKit (hence SqlObject), so that's not really of interest to me.

DAVKit would be interesting, but I feel like a lot of its development should focus around testing and maturing (and also handling error conditions). I'm not sure if that's good sprint material.

UserKit certainly is good sprint material. Well... I personally would want to rethink the whole thing from the ground up -- I find the design a little baroque, considering it has few features and yet is still hard to understand. It uses inheritance where mere composition is called for, among other architectural complexities. But there's a bunch of interesting and useful stuff to be done with user management.

There's also CMS-like issues. All websites have static content as well as dynamic -- Webware deals well with the dynamic stuff, but it could stand to do better with the static stuff. This is Zope territory, obviously we'd do it more simply. UserKit would be a great way to start with this. The two could be worked on at the same time.

Another related item to UserKit and CMS is a better security structure. Right now it's do-it-yourself just like user management.

Lastly, we could actually develop an application. I think it would be really nice to have a Webware application of general appeal, both as an example, and just maybe as a killer app (shoot high!) -- writing an application can be very satisfying. And we could do a little infrastructure work where the application calls for it -- I can imagine a True Webware App being a way to codify some of the ideoms that are currently just ideoms, but deserve to be more. Things like application layout, class layout, application distribution, as well as all the previous issues (user management, security, CMS).

-- IanBicking