I loved playing Mindmaze, the trivia game inside Encarta '95, and was upset that I couldn't find a similar game anywhere online. Distraught and a little nostalgic, I recreated it as WikiMaze in HTML5. I've tried to keep the game true to its classic form with a slightly updated take on the graphics and gameplay.
Can you talk more about how you create the questions?
I'm guessing the answers are always article titles, and you look for articles that have a neat leading summary which is easy to perform anaphor-substitution on. But, how do you pick the candidate wrong answers, and have you developed any other tricks for handling problematic/ambiguous summaries?
I'll be writing a series of blog posts about the architecture of the game as well as more info about the API at http://blog.wikimaze.me/
At a high level you are mostly right. The incorrect answers are generated using http://www.freebase.com/ (now a Google product)
Lots of pages have to be thrown out, but after analyzing thousands of articles it was relatively straightforward to know when you had a good page versus a bad one.
Mindmaze was a favorite of mine in my younger days; I'm glad to see it revived. This is just for my nostalgic needs, but I'd love to see the original music here :P
There is cache manifest support, so after initially loading the game it will save a bunch of questions and you can play offline. When you reconnect to the net it will sync your score and maze progress.
It saves between 10-20 questions, but it could easily go much higher. The cache.manifest says "even if I am offline, still load the page". The localStorage is a database that holds the questions.
I'll be writing a series of blog posts about the architecture of the game, as well as more info about the API
This is amazing, thanks for putting it together. Really fuels my already-intense Wikipedia addiction - I was one of those kids that read encyclopedias for fun and still get into Wikipedia reading sessions with dozens of tabs open.
If you guess incorrectly twice, it will show you the right answer. If you click a third time, it will take you to the Wikipedia article for that answer. Sounds like you got sent to the Wikipedia article but it didn't load?
Looks like it, yep. I'm having the same problem on other sites though so it might be a haywire Chrome extension or something. Still may be worth looking into (latest stable Chrome on mac osx) - good luck!
afterthought - perhaps open the wikipedia article in a new tab.
I'd love to see a more interactive, "HTML5-y" (sorry) Wikipedia, like Encarta back in the day. More prominent audio/video; interactive maps, charts, etc. Sure, we essentially have the same content on the Web now, but not tied together like it could be.
And of course, more MindMaze couldn't hurt.
EDIT: Processing[1] seems like a perfect fit for such a thing. Open, fairly easy to use, focused on interactive graphics, and compiles to JavaScript so it works with most browsers (and Java too, so can even include IE6 in the party)
Personally, I'd like wikipedia to stay the way it is, but others can build on top of wikipedia's content. They are having trouble as it is, with getting more contributors and spam.
If you want to keep your account "secure", login with Facebook, Twitter, or Google. I know that might turn some people off so I included the username login as well.
Future versions will let you set an optional password to "save" the username.
Meh. You could also tie the user to the device or browser and avoid a first time user entering the game with 2k points and the map nearly complete. Give a new user a good experience and maybe they will use your login mechanism and become a repeat user. Otherwise they might just think the game is broken and never return. I wondered why it said I had answered over 40 questions when I only answered 2. Makes for a poor user experience.
Will there be a way to play via browser that breaks it out of the phone interface? I'm having a blast, the only thing that bothers me is playing in the small area inside the iPhone.
42 comments
[ 3.7 ms ] story [ 113 ms ] threadThe questions are generated on the fly from Wikipedia. For any folks that are looking for an easy place to get trivia questions, check out the API (e.g. http://www.wikimaze.me/api/v1/get-questions/?limit=5)
Give it a spin and let me know what you think!
http://itunes.apple.com/us/app/wikimaze-free/id475157639?ls=...
<3 HTML5 apps!
I'm guessing the answers are always article titles, and you look for articles that have a neat leading summary which is easy to perform anaphor-substitution on. But, how do you pick the candidate wrong answers, and have you developed any other tricks for handling problematic/ambiguous summaries?
At a high level you are mostly right. The incorrect answers are generated using http://www.freebase.com/ (now a Google product)
Lots of pages have to be thrown out, but after analyzing thousands of articles it was relatively straightforward to know when you had a good page versus a bad one.
And the maze part is surprisingly fun given that it is basically pointless / orthogonal to the rest of the game.
And does it use cache manifest to store the questions, or localStorage?
(I've been working on both lately at work, so I'm curious how yours works!)
I'll be writing a series of blog posts about the architecture of the game, as well as more info about the API
http://blog.wikimaze.me/
afterthought - perhaps open the wikipedia article in a new tab.
And of course, more MindMaze couldn't hurt.
EDIT: Processing[1] seems like a perfect fit for such a thing. Open, fairly easy to use, focused on interactive graphics, and compiles to JavaScript so it works with most browsers (and Java too, so can even include IE6 in the party)
[1] http://www.processing.org/
There's also a site called 'Ultrastudio' that aims to supplement encyclopedia articles with illustrative applets.
[1] http://qwiki.com
[2] http://ultrastudio.org/
Half of all the proceeds go to Wikipedia for providing such great questions.
https://venmo.com/gleitz?txn=pay¬e=for+awesome+trivia...
I guess I always figured Google would give AppEngine some separate allocation/reverse DNS like EC2.
Future versions will let you set an optional password to "save" the username.
thimmy is right though gleitz, he is wayyyy better than you
Very interested to see how the API develops, would love to use it for http://quizipedia.org
EDIT: I just made the leaderboard.