What is a Seven Day Roguelike?
A Seven Day Roguelike is a roguelike created in seven days. This means the author stopped writing code one hundred and sixty eight hours after they started writing code.
Seven Day Roguelikes are, for roguelike authors, what 24 hour comics would be to comic authors. (cf. 24hourcomics.com)
They are also for roguelike authors what the National Novel Writing Month is for novel writers. (cf. nanowrimo.org)
Is this really the best way to make a great roguelike?
Probably not. However, the short time frame forces the developer to actually plan on finishing. Normal roguelike design is open ended. One will still be working on the same roguelike years after starting. This can become dispiriting, as one may have new ideas that cannot fit inside one's current game. A Seven Day Roguelike is a way to experiment with the genre without fearing creating another life-work. After seven days, one can wash one's hands of the roguelike.
Why Seven Days? Why not 24 Hours?
Comic authors are clearly harder core than us wimpy roguelike developers :>
Programming isn't an activity that I'd recommend be done in 24 hour marathons. Sure, it can be fun. But you don't learn much out of it, except maybe that programming with little sleep makes unreadable code.
Why Seven Days? Why not one month?
Novel writers clearly have longer attention spans than us easily distracted roguelike developers :>
How do I do a Seven Day Roguelike?
Choose a week to work on the roguelike. Post to rec.games.roguelike.development that you have started. After seven days passes, post to rec.games.roguelike.announce your successful creation. Or, you can beg for more time in rec.games.roguelike.development :> (You don't have to announce starting, of course.)
Note that while a 7DRL could be written at any time, the denziens of rec.games.roguelike.development may on occasion organize a specific week for people to accept the challenge in. This allows one to have the shared misery of knowing you are not the only one tracking down a bad pointer at the 167th hour. As a suggested schema for proposing the challenge the 7DRL Voting Specification has been created.
The first such contest was held March 5th to March 13th 2005. The second contest was held February 25th to March 5th 2006.
My roguelike took 10 days, but is really playable! Does it count?
It counts as a Ten Day Roguelike. :>
Can I use external libraries? Graphics files? Design Documents? Code I wrote in the past? Existing roguelikes?
This is entirely up to the developer.
It is recommended one has some design idea going into the project.
You should say what pre-existing code you used. The goal isn't to see who can retype existing algorithms the fastest. The goal is for people to write playable and complete roguelikes.
Remember: if you spend seven days patching NetHack, you likely will end up with something that looks a lot like NetHack, so it would thus not be considered very impressive. However, if you spend seven days patching NetHack and create an amazing new roguelike, you will be suitably honoured.
How do we judge the Winners?
The primary criterion is completeness. The resulting game sould be complete and playable. The author is encouraged to not release another version.
That being said, the only true judge of your "Winnerness" is yourself.
My friend and I want to work on a Seven Day Roguelike together...
Sure! This is definitely a Seven Day Roguelike. Keep in mind that your seven days occur in parallel. Also remmeber that adding more manpower to a late software project only makes it later...
What about licensing?
Clearly, the compiled roguelike itself must actually be released. No good claiming you have it, but won't release it :>
Source code does not have to be released, but it is strongly encouraged that you release it. Heck, release it public domain! It was only seven days work, after all. (Of course, if you patched existing code, follow its licensing agreement...)
So, I've done a 7DRL, but now I want to make it better!
You'll note under the Winners section that the author is encouraged not to release another version. This isn't because we don't want to see bug fixes, or don't want to see the framework written for the 7DRL extended into an even better game. It is to try and push the participants to have something done and polished on day seven rather than "that will be added later..." It is quite sensible and good if people want to take their 7DRL (or, license depending, someone else's!) and extend/modify it into a full fledged roguelike. For example, Martin Read's MPR7DRL formed the basis for Martin's Dungeon Bash and Slash's CastlevaniaRL:Prelude evolved into CastlevaniaRL.
How many Seven Day Roguelikes have been done?
Quite a few! See the 7DRL category or the following detailed lists: