Dougma (dŭg·mə) n.

Well the taxes are done, the gas leak has been plugged, the water heater is fixed, ‘Green Eggs and Ham‘ and ‘Curious George goes to School‘ have been read. With that my emergency laundry list is reduced to… well… laundry. Time to get back to all those things I have had to put on hold. My work todo list is rather stable if overly full and my home todo list is growing like the crab grass I need to deal with. Following in Brett Cannon and Jack Diederich’s footsteps (not that I feel my efforts come close to theirs in scope or impact) here is my Python todo list. It is not exhaustive and I know I will not get to everything. There are a number of outside forces which will be influencing what gets done when, and what things I will have to give up on. I don’t expect anyone to care enough to tell me what I should be working on (well besides my wife), but I welcome any and all input. The list isn’t in any particular order except maybe anxiety.

• PyCon 2007 Media This is going to take up the majority of my time for the next few months. I have mailed off the hard drive to be filled with all the raw video and started the video templates. I need to revive my old python video encoding toolchain from my anime fansub days (has it really been 4 years?) I still don’t have a DVD duplication house selected yet. I am waiting to see the quality of the raw video before deciding on exactly what DVD’s will be produced from the material. I do have some friends on the hook to help me out at a home studio off the union clock, so the ISO masters should not be a problem. I also have a graphic designer lined up to do the liner art for almost free (Hint: Laundry is on my emergency laundry list). I might need more help with this, but I will not know until I start ripping into the data.
• Learn Roundup The PyCon-Tech work desperately needs a bug track system. Other groups are looking to use the code, so this is going from a low short term priority to high. The roundup folks are are willing to host a vanilla instance for free, which is fantastic. I started looking into this and made 3 pages of notes and questions about the software. This was about the time the python tracker group was ramping up plans to convert python-dev over to roundup. Needless to say I was not going to bother anyone with my questions, which I can answer myself by reading the documentation and learning the system myself. I am quite familiar with Trac, and the basics all seem to be the same, and this is not exactly a large project.
• PyCon Sponsorship Application Integration Lots of big words that mean I need to take the very clean and and elegant code of Steve Holden and stuff in into the somewhat organic and undocumented PyCon-Tech code. The sponsorship application is wxPython, db module based with a postgresql backend. The goal is to have a django web interface into the data and have both working against the same network database. I already did a proof of concept which worked well, but I need to go back and flesh out all the views and web forms needed. My hope is to have have a nice system for people wanting to be sponsors to sign up online; this is the one area lacking in the existing application. I don’t see a way to improve on any of the stuff Steve did, and I plan on replicating the xwPython interface as closely as I can in web form. Some of it will not be as good.
• Python Jobs Board I took part in the sprint at PyCon and after three days we had all of the existing functionality implemented, plus a web admin interface, google sitemaps, rss and atom feeds, auto expiration of 31day old jobs, a workflow and more. I wish I could take credit for this, but I played mostly an advisory role while working on PyCon-Tech stuff. The team did a fantastic job zoning out my incoherent ramblings about Django and hammering out 90% of a very complex site in only a few days. Peter Kropf is running this project but has been very buisy on other things and the effort has been put on hold. There is a private svn/trac site he is hosting and a timeline for a demo. I really want to help out on this and I think there is enough overlap between this and other projects that I can make that happen.
• Satchmo This is a storefront and shopping cart framework written in Django. I need to get my wifes buisness site up and running and I want it to be in Django. I have tried many, many, many other packages, and there is nothing out there for the type of merchandise she sells in the way she wants to sell it. It is one of the rare occasions where I feel I can do better than other packaged solutions and that it is worth my effort to do so (but I am not stupid enough to start from a blank slate).
• OLPC Sadly I fear I will not get to this one. Most of the other items have people waiting on them and a local development environment in place. Not sense I started working on speech recognition have I been this excited about a project. Not sense starting on speech recognition have I felt so overwhelmed by the complexities and immense problems which need to be solved. This is my reward for getting other work done.
• importlib/Python 3.0 It seems I might have over estimated my abilities to help here. This one is unique in that it is directly impacts work. We do some nasty hacks in our custom version of Python at Py_Initialize() time to bootstrap things and get imports to work the way we want. We have our own print function and handle python print output. The features being implemented in Python 3.0 will help us remove the majority of our special purpose code. The new annotation system in particular will help out in a major way. Due to details I wont bore people with, I need not worry about NDA and copyright issues as we are talking about are very different implementations. Still, this is a good two years out for us and I have many other projects to work on.
• bitvector/frozenbitvector Over two years ago I wrote a python module bitvector.py under the Python2.0 license. It solved many problems I had for doing compute intensive graph theory. At the time sets were implemented in pure python and not yet part of the standard library. The bitvector.py code still operates two orders of magnitude faster than builtin C sets for large amounts of data. I would like to get this implementation done in C as well as python.
• PyCon-Tech 2008 Lots to do here. Social networking, registration, schedule design, bio’s, etc., etc., etc. Moving to Django 1.0pre with newforms, django-values, content-types, and hopefully a new app Ardien demoed at PyCon. Package up SiteFeatures and the ReStructuredText extensions. Documentation. Testing. Getting help, getting help, getting help.
  

Looking over the list, one thing concerns me. It is very Django heavy. I am not a web developer or web frameworks person. I got into Django by accident. My main python work has been C/C++ embedding and extension work and doing crazy things like adding strong type safety and automated C++ integration. The break was nice, but I would like to get back to that.