So i failed pretty much at this times pyweek. My first mistake was to stay awake to see the theme and then decide things in sleep deprivation. I decided to make a virus simulator where the player takes the role of a mad scientist who tries to design the most deadly virus. The simulation itself should have been a very complex cellular automata, which takes place at a map of the world (i spent some time to find the optimal map projection ;). But i don't went down this path as i knew that i couldn't had finished anything in time. So i did't make the same mistake as this guy How to fail at pyweek. So i had a new idea: i wanted to make a puzzle game based on the simulation of the virus rna in my previous idea. All went well to this point but then i lost focus. i did other things and finally i had only a few hours to code left. So i did something really simple based on the idea of Mutually Assured Destruction. The whole game consist only of pressing three buttons. If you don't want to loose you have to keep them pressed for eternity.

Pro and Cons in the feedback:
Some people didn't understand the idea.
There is a pyglet-windows bug in context of setting icons.
I should mention which version of pyglet and python i use.
The logo got someone to really want to play it.
The self drawn pixel font and the graphics are "hot".
The game is bizarre and somewhat amusing.
It cannot be called a game.
I confused some people.
It is more art than a game.

So in a summary: There are only pros! ;)
Apart from that i have to apologize to everyone who expected fun.

I found that python isn't my favorite programming language anymore :( In contrast to haskell it isn't as elegant and beautiful as before. Although it's still one of my favourites.

Hopefully i will finish a game next time!

So long,
Ernie.

Thank you all for the comments. 
Hope next time will have more time to give you a real game.
Consider this as a proof of concept.

Video of multiplayer mode in action.
Video of me playing survival mode for 4 minutes.

Things that went well

• Squish and I worked well together. I handled all the programming and game mechanics and she provided art and animations in exactly the format I requested it in. Very little time was lost in overhead costs.

Things that went poorly

• Had lots of trouble getting the game to work on OS X.
• People can't be bothered to find a buddy to play with - if you want a top score in PyWeek, don't bother with multiplayer. Check out the video linked above to see what multiplayer mode looks like.
• The gameplay is rather complex, and the game does not help the player learn to play gradually, but instead, drops the player into a difficult arena, totally unprepared for what is happening.
• Squish put incorrect instructions in the "How to Play" screen, and I didn't catch it until way too late. However, as BlueDragon pointed out, even the real instructions are confusing.
• Atoms fall down too quickly - the speed at which they fall should start out very slow and then slowly increase.
• The crazy physics glitch bugs that remained are bad enough and occur frequently enough to almost render the game unplayable.
• No score in survival mode.
• Bombs are actually helpful, not hurtful.

Next steps

• I'd like to port the game to html5 canvas + javascript. I'd like to see how lowering the boundaries to getting the game up and running affects the game's popularity.
• I'm considering creating a new game just to explore the grapple gun concept. I think that there is some complex and ridiculously fun gameplay to be had there.

This morning i was surprising for the bad rating and one comment.

I don't know why we recived so low rating, compared to “pluto the ninth planet” and “camouflage” the game is more complete, but clearly we made some mistakes.

The comment is:

“... I really object to the sexual objectification of the female
lead, starting with the picture of her bum in the intro sequence. It felt gratuitous and
exploitative, and definitely not cool.”

The opening comic is a comic and in the comics there are often sexual references. I created Pheega with the help of Milo Manara comics, these comics are full of sexual picture but I don't belive he depicts the women like objects, the female character of Milo Manara are always strong and indipendent, like Pheega. Besides I like the female body and I enjoyed drawing this scene. If you consider this picture exploitative you must consider the cute animals, cultural references and more exploitative too: they are all potential way of win the others over. In any case we do the games because we like doing them and because the peole appreciate them. If someone appreciates the butt of pheega my purpose is reached.

If I like drawing in this way I will not censor myself for possible misunderstanding due to wrong point of view.

“Only the self-censorship is worse then the censorship” Andrea Pazienza (1956-1988).

Sorry for my english.

If you went over to PyPI or Bitbucket right now, you could download a new release of Doctor Korovic's Flying Atomic Squid with a bunch of new components, including lighter-than-air flight. Just sayin'.

Thanks to everyone who played and rated my game.

Earlier this day I was walking down the room, flexing my biceps and half-screaming: "I am god of python!"... Because for a while I was co-winner of this pyweek... Thanks for misunderstanding in calculations...

Realization that I am at third place was best ego reducing therapy in last year, I think...

And third place is wonderful too. I remember playing "Mechanism" or "The FInal Huzzah" and thinking: "damn -- my game sucks...".


Things I have learned:

- I not am a god of python...
- Art is important
- The only things I can paint are simple geometric figures
- Music is important
- Do not fear to erase all progress and start over (Dinosaurs and Santa Clauses...)
- It is sad but it is safer (less DNW) to make game with pygame than with pyglet
- pyglet is still awesome
- I have trick you (players) with best scores to think my game is better than it is...


Future:

I need to go mobile for sure. Controls are perfect for touch screen. I need to port this game to android and maybe HTML5. I have no mac or idevice -- so no iOS port.

Quite happy with this one. Overall, I didn't do so well as last time (3.03 vs 3.42), but I spent a lot less time on it.

Good:
The fun score (3.6) was my best so far I think.

One of the criticisms I got was
"Needed less punishing death and more rewards for the player in progressing upward."
which is fair. I thought about having checkpoints but thought that it wouldn't work for such a simple game. If I were to expand it I think I would need to add additional gameplay elements (e.g. enemies) so it doesn't get too repetitive.

After a disastrous start, I managed my time pretty well and came up with a plan that was actually achievable. I had plenty of time left for testing, packaging etc. this time round.

I used sfxr for the first time for sound effects, which is awesome http://drpetter.se/project_sfxr.html

My hacky method of generating the map actually worked! Making it actually possible to win is tricky :/

Bad:
It wasn't particularly innovative. I need to come up with some more interesting ideas next time.

I think a slower pace and more mazelike levels would have been better, but I just didn't have the time.

I didn't use the full 7 days, so the game is very simple and short. I think next time I might set aside more time and maybe do a team entry?

Favourite entries:
Cloud Shepard: http://www.pyweek.org/e/cloud-shepherd/
Multiverse Factory 14: http://www.pyweek.org/e/multifac14/
Brainwashed: http://www.pyweek.org/e/apadentro/

See you next pyweek :)

Pyweek #3 for me done!

I am so happy with how my entry did.  Much better than my previous entries!

Things done right:
    Set goals on each day
    Use revision control (github)
    Develop artwork (tileset)
    Use existing resources well (hero sprite, music)
    Mix genres
    Good performance on many machines

Things not so right:
    Not fix annoying bugs (no time!)
    Not being clear about the requirements (python 2.7)

=============================================================================
I'm going to address a comment posted about my game because I felt it was too critical of the entry and justify myself here.  The comment, verbatim:

Dying is fun — unless it's the lousy platforming implementation that kills you. For example
getting stuck in a door and getting shot or running into an infinite loop in movePosition().
It's hard to score the art. It shows that it is lifted from various places because it does not
have a uniform style and is not put into the game neatly (try jumping while running for
example).


Platforming: yes, there are bugs.  on python 3.2 especially, the timing is totally off (not my fault) and makes the bugs worse.  for one person making a platforming game in less than one week is no easy task.  also, getting shot is part of the game, not a bug.

art: are you kidding me?  i created the tiles from another tileset (had to redraw them all completely), made the robot myself and had to customize the animations for the hero.  the art totally has a style, i'm sorry if you don't see it.  the platforming sprite was not from me, but was very well animated and took a lot of work to get the details right.  there was no jumping animation with the sprite, why be so critical if the running animation plays while running?

the innovation score of one is totally uncalled for.  i should have gotten much higher than that.  the game combined a platforming game, a puzzle game, and elements of rougelikes with the text, collecting keys, and unforgiving death.  get a clue, seriously.
=============================================================================

seems that the packaging script ignored my extension-less level files (only a problem for version 0.0.2); I should have made sure the package was complete before I uploaded :P

if you want to try them you can get them from github: 
https://github.com/bjornicus/pyweek14

or grab the whole tarball:
https://github.com/bjornicus/pyweek14/tarball/master