Nick's BloggityThing

What I’ve been working on

I always have a million projects on my mind, but here are the top ones of late:

Angular + Organizer and Elemental Chess

To learn Angular.js I started porting my old organizer application to it. The new source code is on github now. The old app uses Google Gears for the client-side sqlite database with full-text search features, but unfortunately Gears is no longer supported by any browser and WebSQL is not going anywhere. Fortunately, Angular includes a handy full-text object search that works well enough for my needs, and porting was surprisingly simple. Unfortunately, I had to use a lot of ng:eval

I also hacked in a very simple method of synchronizing application state so multiple users can collaborate. All you have to do is wait for some things to change, serialize all the stateful information to JSON, run diff-match-patch to compare it to its previous state, send those patches over the wire via Socket.IO, and patch the state on the other end in the same way. I _.debounce’d this so it doesn’t happen too often. No idea how well this scales, but it’s ghetto fabulous so I like it. However, a much better solution would be to hook into angular’s scope.$eval for some hints on what actually changed. If anyone knows how to do this, please answer my StackOverflow question!

I also attempted to port my Elemental Chess game to Angular. I was unfortunately rather disappointed with the result. I used a lot of ng:eval and ng:class, and Angular ended up screwing these up and using old values far too often, so you’d move a game piece and it’d leave its colored background behind. I guess it’s not really meant for games. Either way, this did result in me porting the game engine to a DOM-free implementation, which made it easier to build a minimax AI player to play against. I’ll post this version when I finish tweaking the AI. It’s currently pretty slow, because I couldn’t find a better way to clone the game state for each iteration of minimax other than to selectively JSON serialize the whole thing and then re-build the board each time. Perhaps if I made a game event log and added a feature to undo your moves I could use that in my minimax algorithm instead of recreating the game state. I’m not sure how I want to do the graphics, but I’m leaning toward doing them more like the old way again only with support for dragging with touch screens so you can play on your phone.

Oh yeah, gotta love the name they put on this version of Angular: “Canine Psychokinesis”. It sure beats “Jiggling Armfat”.

Factor Blaster and Beat Ship

Two games I started on PixieEngine.com. Beat Ship was lost but I intend to resurrect it. For Factor Blaster I want to remove spacial navigation and make it more like Missile Command. Check my previous post for more details.

TopTools

Like I said before, using webfinger was probably a mistake, and I should really make this a general purpose tool instead instead of forcing a use case on it. I’d like to make this super easy to use without even creating accounts. Perhaps it would work if all votes were public and an election’s creator and moderators were allowed to police them for sock puppets. Also it needs to have a hot-seat mode where all the users can vote from one computer, so you can easily settle disputes like which board game to play next in real time with minimal hassle. And it needs to work on a phone browser!

Art and Animation!

Cartoon animation is a passion of mine. I practiced it in my spare time in high school and college. I do a lot of drawings for friends, and I even sell stuff at conventions. However, I haven’t really put together much of a portfolio online to show off my work. I need to do this.

Playing Game Dev Story has convinced me that career changes are good for developing your all-around skills and making more valuable people.

I’ve also been looking into traditional pencil and paper animation. I got myself a peg board, light board, and ordered some paper for it. We’ll see what I come up with =]

HTML5 Dev Conf

I signed up for this on a whim, and I’m glad I did. I learned about a lot of new fun things out there.

The biggest booth at the show was for SpacePort, which can extract vector art from flash files and render it in SVG and CSS or OpenGL ES. They’ve also ported the Flash AS3 API to JavaScript. This seems aimed at helping Flash developers port their apps to native mobile apps and flashless web browsers.

I attended a panel about Three.js which I’m pretty tempted to use now, despite having already used the lower level features of OpenGL and GLSL before.

I was looking forward to Max Goodman’s panel on procedural audio in JavaScript, but he called in sick. I was still able to get some advice on syncing the audio in Beat Ship, the browser-based rhythm game I started on at last year’s TigJAM. Unfortunately the source code for this was lost when PixieEngine.com was re-architected, but it shouldn’t be too hard to rewrite. This will be my next project.

Speaking of Pixie Engine, I ran into the creators and they insisted I make my experimental game Factor Blaster more like Missile Command. We both agreed that spacial navigation just distracts the player from the real fun of factoring. I think I’ll place the player at the bottom of the screen and have the numbers rain down from above. I’d love to add some more variation in the weapons too. I’d like 2 to be a machine gun, 3 to be a sweep laser, and 5 and 7 could be chain reaction bombs like in Missile Command.

I also ran into Ryan Jarvinen from Eventbrite and it seems we’re both big advocates of the alternative vote. Some time ago, I made a nice web interface for running Condorcet elections with OpenSTV called TopTools. Perhaps I could have it import EventBrite attendee information in case you wanted to run some elections at an event.

I realize that this would serve better as a general purpose election tool than it would for the niche I’d selected for it. Also it needs to be able to display ties, especially when there aren’t enough votes for a definitive winner to be selected. Also I’m a little worried about my choice of doing logins with WebFinger. It seemed like an awesomely simple idea at the time, but I’m sure it’s just gonna scare people away since there are so many issues with everyone’s implementations of it — even google.

4Fourths

What a beautiful game! I got to play with this at Gamma 4 as part of their One Button showcase. It’s a co-operative game where four players control two ships using one button each: players 1 and 3 can charge and shoot their lasers, while players 2 and 4 control the upward thrust of each ship.

4Fourths

Unlike a lot of co-operative games, this one does nothing to stop players from killing each other. In fact, it’s pretty fun to spar between challenges, and most players who pick the game up got right into shooting and dodging each other. Before long, the first enemy is lowered between the players like a giant wall, forcing them to break it up and learn how to work together without a word of dialog. But as the game goes on, bosses place less walls between the players until they must be very careful not to blow each other up.

Despite the fun of shooting each other, there are no competitive scoring elements. Also, it’s rather difficult to defeat a boss without both players’ combined firepower, and failing to do so in time means everyone loses. That makes this one of the most purely co-operative games I’ve played.

Also, I love the bright colors and boxy graphics. It’s so Nu-Lo-Fi. And it’s made in unity.

Building a Good New User Experience for Elemental Chess

People don’t like to log in to things until they’re convinced it’s worth it.  That’s a given.  My first thought was to just make a sandbox mode for my game, where you can play it purely on the client side, with a friend or against the computer.  If you wanted to play against other players online, you’d need to log in with an openid provider.

I showed off my new sandbox mode at Google IO.  It didn’t have AI or checkmate detection yet, since I couldn’t decide whether to implement that on the client side or the server side.  Google Web Toolkit seems like a nice solution for client/server code sharing that I will definitely pursue.  But then I thought, I can solve this code sharing problem and provide a better user experience, if I get a little more ambitious with my user account handling.  I’ll let users do anything they want except hang onto their history if they’re not logged in.

When a user hits the site without a user in their session cookie, I’ll create a brand new user for them with a randomized name (and plenty of invitation to change it).  This is a full-featured user, with one caveat: it can expire, at about the time their cookie-based session should.  To keep it from expiring, all the user has to do is associate it with an OpenID.  This is exactly the same thing as logging in.

Okay, but what if a lazy user who already has a saved account comes along and plays a few games without logging in?  Easy.  When they eventually do log in, we can delete their temporary account and replace every instance of it with the real one!  Talk about awesome.

Since I’m totally overloading my Login button to function as “Login”, “Register”, “Save”, and “Merge Accounts”, I think it’s appropriate to call it something else.  “Recognize Me” sounds pretty good!

Ruby Inner Methods Are Not Really Inner Methods

class T
  def foo
    def bar
      puts "oh snap"
    end
    bar
  end
end

t = T.new
t.foo
t.bar