A nice second day, didn't brought the whole project further as 
I targeted, but better than I thought. Now, tomorrow, hopefully
I have all the bits working to finally build some story.

Today i wrote a simple and crappy platforming engine, coded a tile map and painted a tilesheet. Also i did the movement code of all the enemies and the protagonists.

We decided to make a tower offense game.

You are an evil alien, who wants to conquer the earth by “mutating” its inhabitants into evil creatures who fight their own race.

The challenge is to mutate the right creatures at the right time, to overcome the defense lines of the earthlings. Because of the endangered human species, you have to choose your resources carefully.

Day 1 Brainstorm for possible creeps:

• Fast Food Restaurant:  Spawns “Fatsos” who roll their way through the defense lines. They are extremely durable.

• Sports Court:  Spawns “Runners” who are very fast but weak.

• Tavern: Spawns “Drunks” who attract the fire of the towers.

Screenshot day 1:

At last I have time and I am mentally capable (opposite to yesterday -- long story...) of coding. So, Idea. I will try to go 3d. It is probably going to be big failure, but who cares (I care, actually).
The idea is inspired by wonderful game Proun (http://www.proun-game.com/) + child toy named shape sorting (http://www.google.lt/search?tbm=isch&q=Shape%20Sorting) + Stupid TV show "Hole in the Wall".

So, player will drive throuth track with shape and it needs to switch shapes (mutate) to go through holes.

As I was reading other participants diaries I find out that my idea has some similarities with malcolmt entry (http://www.pyweek.org/e/Hermit-42/). I hope he do not mind it.

You can try Proun it is pay as You feel type of game. Just be careful -- do not miss pyweek because of it.

(So it's not from the game, but it's from the community, sort of...)

I now have textures, and most of the time was spent making that super cool background!

Hi all!
So, I was supposed to say something? Better late than never I guess...
Anyway, what a good news the theme anouncement was!
The theme "Mutate!" was the first to draw my intention, and it stole me a few hours of sleep, while my sick mind was trying to apply the "Mutate!" theme to a bullet hell.
So, my game was mostly ready in my head before the start of the challenge!
In addition, I was working on a bullet hell reimplementation right before the compo, so, I have everything in mind. It leads to really similar code organization (mostly the same module names, and the code is probably similar in a lot of ways, too), and things like that, but I did not look at my original code *at all* once Foreseen Yet Unpredictable Invader started.
Now, I'm going back to add a few missing gameplay features in FyUI!

I just got my first rendered level:

  I'll be adding textures next, and with my system of level saving, it should take minimal effort to code

The advantage of having your Tile class (Block in my case) subclass pygame.sprite.Sprite is that you can then use spritecollide with a sprite the size of the screen to get all of the sprites you need to render! It does raise some confusion about coordinates, but I think I can sort that out.

After thinking it over, my game will be simple but hopefully fun. Basically, you just wander around in a procedurally generated cave, trying to bring life back to it.
Everywhere you go, plants start to grow again. But, you'll have to fight off the evil creatures already living there. Your only defense is the life that you cause. The more plants in an area, the more defenses you have. If you successfully bring life back to the entire cave, you win. But, the cave itself also mutates as the game progresses, so you'll have to work fast. 

I deliberately tried to make the gameplay as simple as possible since I have a nonexistent chance of finishing already, and so I could add music!

Since I'm on the process of encoding a timelapse video of my current programming, I thought I might as well write the initial diary entry and pitch my idea.

My game is going to be inspired from the DNA duplication process. The basics is that you have 4 blocks ATCG (actually, the players have (AUCG)), which may be connected as A-T, A-U, C-G. Each player has a hand of cards and a hand of nucleobases, and they share a target sequence to be matched with those blocks.

Your goal is to get the higher score in a battle, getting +1 points for every nucleobase that you play that matches the sequence, and losing 2 points for every nucleobase that you play that doesn't. The sequence is encoded from left to right, always, and you may play from one to as many nucleobases as you hold, but they must be played in the sequence that you hold them, rearranging is not naturally possible.

In addition to the nucleobases, you have your cards, which allow you to draw more nucleobases to your hand, possibly pushing out of your hand excessive nucleobases, steal nucleobases from the opponent, switch your current nucleobases with new ones (or the opponents nucleobases), clear your whole hand (might be useful if you're trying to get them special cards), play more than one nucleobase sequence in your turn (if you want to play a large number of nucleobases, but want to skip some in the middle, that may be a good one), or play more cards. Maybe some other cards to come.

As soon as this bit is solidified and tested, I'll finally start my UI programming (I've been testing the stuff on command line so far).