Finally ported my ofxWordSquares to Python, leading to an improved wordsquare generator. The progress has been fairly gratifying, especially the implementation of a Dawg (directed a cyclic graph) for prefix search.

Oddly enough (or perhaps not so oddly, knowing me), this progress, the amount of squares I can now generate, and the amount ahead, that still remains out of grasp, are more depressing than anything else. Perhaps it is a counterwave, a moment of low spirit after the fit of work and tension required to achieve that result. Still, it does feel I am distracted, focussed on technicalities rather than art, and still utterly in the dark regarding my future, etc.

The problem is this: sure, with a bit of patience and disk space, I could have a database of 4-lettered squares with a size in the hundreds of millions, but I’m not even sure I could do anything with it. My other project WordSquaresAI was already difficult to manage with a database of 200’000 squares (next steps: improve that), and I ran into difficulties when I tried to get into the artistic part of it, namely when I had to find specific squares I would want to exhibit as works of poetry.

The two problems, of quantity and of quality (finding good, beautiful squares), remain unsolved, and I don’t see how they would go away any time soon. It does feel that I ought to change my strategy entirely. Namely, even more radically than in my previous turn, when I turned away from my focus on AI to complete old projects a few days ago, to put artistic practice at the absolute core of what I am doing every day, and put everything else in the background. Start with a basic random function if I have to, but (re)introduce play and artistic creation asap, at this very moment.