Author archives: Mike

I've been tracking my corporate books using GNUCash for about 21 months now. At the start it was quite rough (a number of nasty glitches that have since been fixed), but it's served me reasonably well.

In the end the problem isn't the software, it's that the accountants charge the same to read/review the books ...

Spent the whole day testing and cleaning up tutorial code for PyCon. Ran smack into a big bug in OLPCGames; a thread encapsulation issue that would cause core dumps every once in a while as DBUS tried to run in the Pygame thread. Basically a network (DBUS) operation was being called synchronously... you can't do ...

Interesting lecture, touches briefly on the idea of designers (intenders) attempting to set up a conflict in the mind in order to have the user consider the ideas they have not yet been exposed to. General points here and there of interest as well. (Yes, I listen to these lectures as I work).

Spent a few hours this afternoon on trying to make Productive more fun to play. Biggest change was making the "takeover" process much faster (factor of 8), so that you have to react to a takeover attempt within 10s to avoid losing your units. Also implemented a resource cap on each unit so that you ...

As those who have signed up for PyCon may already know, I'll be presenting a tutorial on developing Activities for the Sugar platform on the tutorial day before the conference.

There are also a number of OLPC talks scheduled, by Dr. Brown, Charles and Kirby. I believe we're also going to have a ...

Given that gcc can be a bit heavy, it would be neat if we had a compile farm somewhere made out of Qemu images with overlays that people working on XOs could shell into in order to compile various software. I use this approach on my workstation when I want to do something that's messy ...

Developed a product a few years ago using Twisted, which got left on a shelf due to unavoidable complications at the business level. Client with a license to alter the code asked me to make some changes to it, so I sat down today to rebuild the development environment...

Twisted and Nevow have moved along, ...

Got the AStar implementation working again. Was a couple of stupid little bugs in the code that creates the topology from the map. AStar is a graph-based implementation (which is nice, because Productive uses a graph-based world definition so that at some point we can have very complex map types), which traces through the connections ...

Since Productive is not considered an official task, it's been stalled for a while. Took a few hours today to try to un-stall it. Mostly trying to optimize the AStar algorithm. Sped it up to the point where a moderately sized world (96x96) takes about .6s to compute an AStar path. That's still way too ...

Reading through the original outline for my thesis (that is, the one in 1997) and realizing there's lots of stuff in there that just isn't getting covered in the rewrite. So, today I worked on the "Patterns of Creation" section, this section just discusses the nature of generative and organizing patterns and their role in ...