This morning I posted a bugfix update to Fanorona. When an Android app crashes, the user has the option of reporting the crash to the developer; it turned out Fanorona would sometimes crash if you touched the screen outside the board.

This morning I published an ad-supported version of Fanorona. I don't know yet how many people are downloading it (the Market's stats update only once per day), but so far AdMob says they've gotten 462 requests for ads. (Only 82 of them have been filled, though, which means I've earned only a penny. But hey.)

Last night I published my second Android app, called Buzzphrase. It's a random mission statement generator-trying to be a high-quality mission statement generator, with a wider variety of output than most you see. I built it around a grammar engine; that is, it knows a bunch of grammatical structures and how to glue them together and plug in ( Read more... )

I finally released Adder, the language I've been working on off and on for the past two years.

Adder is Python with a Lisp; Adder : Python :: Clojure : Java. It integrates smoothly with Python-Adder code can call Python code, and import Python modules, and vice versa.

I wanted to try out Scala (mostly to have a real programming language on Android), but I've run into a snag: the compiler is much too slow. I was OK with Scala 2.7.x on my desktop at work (4-core 3GHz Woodcrest-a Xeon version of Core 2), but 2.8.x on my laptop (2-core 1.6GHz Core Duo) was insanely slow--105 seconds to compile a 6-line program. ( ( Read more... )

Today I published my first Android app to the Android Market. It's Fanorona, a traditional game, probably SCA-period, from Madagascar, played on a 9x5 board:


( Read more... )

Someone at work just posted that Talk Like a Pirate Day is coming up on Sunday. Most years I forget it, so I went to add it to my Google Calendar. I used the "Quick add" feature, and typed in "Talk Like a Pirate Day 9/19 every year", then went to see if it had understood "every year". Yes, it did ( Read more... )

About the article I linked to a couple of days ago, claiming that quantum computers do not help with NP-complete problems: someone who seems to know more than the Ars Technica reporter has pointed out some problems with the paper. Plausible, anyway.

Just seen at Ars Technica: it's now been proven that quantum computers are no better than ordinary computers at NP-complete problems ( Read more... )

Yesterday it occurred to me that, if the people at Fusion Garage had been Hitchhiker's fans, they wouldn't have named their poorly reviewed tablet the JooJoo.
So, for instance, when in a recent national speech the Financial Minister of the Royal World Estate of Quarlvista actually dared to say that due to one thing and another and the fact that ( Read more... )