After some brainstorming, I have a game idea that sort of works with the theme, and some idea of the minimal set of features needed to get a playable game, which is always a good target to have.

For reasons I may regret later, I decided to see if I can bend kivy to my will. I still haven't wrapped my head around the exactly how everything works, and that'll hurt when I try and hook up actual gameplay, but I did manage to get it to the point that I can display a maze.

Hopefully I'll have some time to trying to get this to run on android, but that's a TODO item for later in the week, and thus may be dropped.

Not much time to work on the game today, but the game has gained a protagnist who can run around the maze and advance onto the second maze. Of course, this isn't a game yet, but it's still progress in the right direction.

I should have more time available tomorrow, so hopefully I can add an antognist and start adding actualy gameplay mechanics.

I had intended to add actual gameplay mechanics today, but I got distracted.

Instead, I spent some time tweaking the graphics, although that needs further work, and also poking at getting the game running on my phone. After some running around in circles, I managed to get it running. It's not ideal - the interface feels clumsy, and I'm not sure how to fix that, but it at least works.

Now I really do need to add more gameplay, and also finish the level editor.

I didn't get as much as I'd hoped today. I added some very simplistic movement for the antagonist for testing purposes, but it's hardly a challenge. I also started turning the level editor into something useful, although there's still a lot of work to be done there. I also hooked up won / lost and a basic intro screens. 

Hopefully tomorrow will be more productive, and I can get most of the gameplay into place.

So I managed to finish the level editor today, added a couple of extra features to the maze and also tweaked the nemesis AI to be approximately what I wanted - directed enough to be something a threat, but still really stupid. All the components for a game are now there, it just needs more content.

I'm not going to have much time to work on this on Saturday, so I'll need to get it into submittable shape tomorrow, but with the level editor working, hopefully I can get it to a reasonable number of levels quickly.

I should try add some sound to the game. Kivy's documentation suggests it will be fairly straightforward, so hopefully that won't take much time.

I added several levels, fixed a few bugs and gnerally got closer to something submittable.

I added sound, although it sometimes doesn't play the desired effect. I'm not quite sure why, and I probably won't have time to figure that out before the deadline. I also added some configuration options, so you can enable and disable sound, and choose the starting level (which is mainly a debugging aid, but it's also the sort of feature that may be useful suring judging).

At some point in the last couple of days, I broke the game on android, so I fixed that. It's rather slow on my phone - I think I know what I've done wrong, although I may not have time to correct that before the deadline. It's also a bit painful to play using a touch interface, but I think I can improve that without too much effort.

I found time to do a few little things, although nothing excessive. I fixed a couple of bugs, and reworked some of the code to something a bit more suited to kivy's requirements. I also finally got around to hooking up kivy's caching infrastructure to cache images and sounds, which should also help performance a bit.

I also spent a fair bit of time checking that I could get the game running on windows. This took rather longer than expected, as I had to tweak my virtualbox setup to support OpenGL to get kivy to run there. I tried, but ultimately abandoned, the recommended approach to building a windows installer, as that broke in strange ways and debugging it looks like being a major pain. The solution I came up with isn't very elegant, but at least works in my test cases.

As far as the final result goes, I'm a bit ambivalent. I'm reasonably happy with what I accomplished during the week, and, as an exercise in learning kivy, this has definately been useful. I still haven't quite wrapped my head around the correct mental model for working with kivy, but I'm a lot closer than I was.  I'm not that happy with the actual gameplay, though - I never found time to work on the extra puzzle elements I wanted, so the game eded up too simplistic and lacking in whimsy, although it has some promising ideas.