The large text was a link to this blog post. How had they figured out the servlet’s address and the parameters that need to be passed in?! Then I saw this:

	<!--
	<p>Enter your high score:</p>
	<form action="/sign" method="post">
		<div>
			Name: <input type="text" name="name" />
			<br />
			Score: <input type="text" name="score" />
		</div>
		<div><input type="submit" value="Post Score" /></div>
	</form>
	-->

I had forgotten to delete some test code that I had written! The code provided an an input for name and score, and then you could submit it. I was handing people a method for inserting high scores to the board.

I deleted the large post, giving the person some credit at the bottom of the scoreboard. I’m starting to think that a lot of the other highest scores were created that way, but no one else pointed out the flaw. I’m considering a move to weekly high scores so that it’s more competitive, and I don’t feel bad about deleting the current high scores. I’m going to have a talk with Alex about it, and try to whip up some GQL to make the whole process automated.

First I created the high score app, which you can view here. Google’s tutorials for AppEngine are quite nice. They guide you through creating a guestbook that people can sign using their google account. From there I changed several things: instead of taking a guestbook post, the app accepts a score; instead of having users log in with their google account, the player can enter any name he likes (arcade style, without the three character restriction). I then changed the default page to look more scoreboard like, displaying the top 20 scores in descending order.

The most difficult part was getting the extension to integrate with the app. The first step was to add permissions to the extension:

  "permissions": [
    "http://avalanchescores.appspot.com/"
  ]

From there, interaction with the app was done through an XMLHttpRequest:

function submitScore() {
  var highScore = document.getElementById("HighScore").innerHTML;
  var name = document.getElementById("name").value;
  var url = "http://avalanchescores.appspot.com/addscore";
  var parameters = "score=" + highScore + "&name=" + name;
  var req = new XMLHttpRequest();

  req.open("POST", url, true);
  req.setRequestHeader("Content-Type", "application/x-www-form-urlencoded");
  req.send(parameters);
  window.open("http://avalanchescores.appspot.com/","High scores!","");
}

The XMLHttpRequest just sends the parameters to the servlet’s location, and the java code that handles new scores takes care of the rest. Overall I was really impressed with AppEngine. Setup was extremely easy, especially once I successfully installed the Eclipse plugin. There’s even a neat dashboard with usage statistics that make me feel like Zero Cool from Hackers.

Code for Avalanche itself is now on github, which also works like a charm.

I will be leaving for Germany on Friday, where I will be studying abroad until mid-August, so I probably won’t be updating much.

Braid

Passage

Judith

Gravity Bone

I was going to write short summaries of these games, but I think it’s really best if you go play them. The unifying characteristic of these games is that they are short (with the exception of Braid), beautiful, and convey a narrative that wouldn’t be as effectively delivered in any other medium. These games really affected how I think about video games, and what can be done with them.

I think, personally, that game overs are a thing of the past. Heavy Rain is not about failing or succeeding, Heavy Rain is about the journey… not to replay, but to continue the experience, whatever [the player's] choices… The four playable characters can be in great danger sometimes, and sometimes they can die. But when you lose a character, this is an information that the game takes on board, but the story continues… You bear the consequences of your actions. I think, personally, that game overs are a thing of the past. Heavy Rain isnot about failing or succeeding, Heavy Rain is about the journey… not to replay, but to continue the experience, whatever [the player's] choices… The four playable characters can be in great danger sometimes, and sometimes they can die. But when you lose a character, this is an information that the game takeson board, but the story continues… You bear the consequences of your actions.

In Heavy Rain, the player is in control second to second, he tells the story through his actions. All this is done in a very fluid, seamless way, with no cut scenes, no big flashing sign to make decisions, and this is what makes the game really unique… Heavy Rain is not a videogame anymore in my mind, because it breaks with most of the traditional paradigms, but it’s fully interactive.

This shit didn’t come from the marketing department, I’ll tell you that much. This guy is a fucking wreck – he’s barely making it through the day. The clutch of the creative process, as experienced by Cage, is indistinguishable from a mental illness. Heavy Rain is a symptom of his condition, and he is expelling it from his mind piece by piece in an effort to save his own life.

The game’s purpose is to be something quick that you can play while doing something else, or waiting. Of course, music playing in the background can be quite annoying if you’re trying to multitask, or are already listening to music.

Here’s what the original code looked like:

  <audio src="resources/shortgametune.mp3" loop="true" autoplay="autoplay">
  </audio>

This code just loaded the song, and played it on a loop. At first I added the built in html5 audioplayer so the user could pause/play, and adjust volume. Overall I found that too distracting, so I built a mute button. Here’s what the code looks like now:

  <script type="text/javascript">
    function mute(){
		var audio = document.getElementById('song');
		if(audio.paused)
			audio.play();
		else
			audio.pause();
    }
  </script>

  <audio id='song' loop autoplay>
    <source src="resources/shortgametune.mp3">
  </audio>

There isn’t a cornucopia information out there on the audio tag, so hopefully this might help someone if they’re looking to do something similar. I’ve also updated the source code that I posted earlier after adding the changes, and cleaning everything up a bit.

You may have noticed the slick code highlighting on the above samples. I looked around for a while trying to find a good code highlighting plugin for wordpress. The one that I am now using is SyntaxHighlighter Evolved. The best explanation of how to use it is here.

I uploaded a small chrome extension this evening. It’s a small game called Avalanche. I coded the game together with my friend, Alex Langenfeld (who is a genius). It’s based off the classic avalanche game that everyone enjoyed on their TI-82s throughout high school and middle school. As it says in the description, the game is based on code from this tutorial, and uses an art asset from this forum post.

The platformer code from the tutorial provided the basics for animation, collision, asset management, and game state management. First we boiled the game down, taking out extras we didn’t need. We then added code for the falling icicles, code to generate the icicles, score management, and a couple other things. The worst part was dealing with the pixel art and animations. Overall it was pretty simple, and fun to develop!

The best part of making a chrome extension was how simple it was. All we had to do was include a simple json file with some metadata, and chrome did the rest! The process of adding it to the chrome extensions website was also extremely simple and rewarding. Check out the source code if you’re interested!

I’ve been trying to start a small project to teach myself ruby. Unfortunately, I decided to do this right when _why disappeared from the internet. I had been planning on using Shoes, his ruby gui toolkit, but all his sites disappearing has made it a bit more difficult.

I spent a couple days wrestling with getting everything working. It retrospect it took an embarrassing amount of time once I knew what to do, but when isn’t that the case? Here’s the steps I took in case anyone else is interested. I did this on Ubuntu, so I use apt-get. For other os’s, check the why repository mirror on github at http://github.com/whymirror.

1) sudo apt-get install ruby-full :: I was missing some dependencies required for shoes when I didn’t get the full ruby install the first time I installed it. I spent a lot of time hunting around for what I was missing, but this seemed to take care of everything.

2) sudo apt-get install shoes :: I originally downloaded shoes off of the github mirror, but using apt-get makes the process much simpler.

From there I recommend reading why’s shoes tutorial/guide/novel, “nobody knows shoes”. This guide disappeared along with everything else related to why, but it has been reposted in other places. This post has download links for both “why’s poignant guide to ruby” and “nobody knows shoes.”

The most notable thing about the experience for me was the control scheme. I was reading some comments and came across one that said something along the lines of “How can you release a first person game nowadays, and not have wasd and mouselook controls?” I found one of the most compelling aspects of the game was the control scheme, which by most standards today would be considered poor. This lead me to ask myself the question, “is it okay for a game to have bad controls?”

Controls are an essential part of the gameplay experience, perhaps even the most central aspect of it. Controls are the vehicle for making the player experience something. Most games revolve around feeling good at something. Guitar Hero is enjoyable because it offers an interface of control to make the player feel like they’re a rockstar. The Splinter Cell series is good because its controls allow the player to have the experience of being a spy, sneaking into bases, assassinating people, and just being an all around badass. These controls give the player a sense of responsibility and achievement in return for their actions; the controls aim to lower the player’s abstraction from the game, and immerse them in the feeling of accomplishment.

If movies were like games, every movie would be an action film. Most games offer little emotionally besides feeling like a badass. Sure, a character with a hamfisted relationship to the player might die in a cutscene. But in the end, you still blast off heads with ease and conquer evil. Movies don’t always end happily. More importantly, film characters aren’t always in control of their destiny. No matter how hard they try to achieve something, they fail.

This is something that Judith achieved. It took control away from the player, and made them feel awkward in their world. In retrospect, it’s amazing how fearful I was every time I had to stop and slowly turn with the arrow keys to see what was in a newly opened room. I wasn’t able to do what I wanted: I couldn’t strafe to the side and take a peek into the next room, and I wouldn’t be able to quickly turn around, and run out of the room. Films always have this effect, where the viewer has no control over what happens. Games only try emulating this by creating cutscenes. Instead, they should use the strongpoint of the medium, control, to evoke more complex feelings from the player. I want to fail while I’m in control of the game, because actually failing always has a greater impact than being told that you fail.

Most importantly, this is not an excuse for “poor” control. By this I mean, controls should achieve what they set out to do. If you’re a space marine blasting away thousands of aliens and you’re supposed to feel boss, don’t have unresponsive controls with poor auto-lock. Designers should explore control. Exploring control in games is like experimenting with the cinematography in a film. Movies are not games, and innovation should not be approached in the same manner. I’m not saying art direction, camera, and story are unimportant, they’ve just already been explored more deeply than control. Controls are means to a narrative end in games, and if they don’t deliver that, then they’re not doing their job.

I’ve decided to learn to use Emacs. This came after reading one of Steve Yegge’s rants where he proclaimed “Emacs is the 100-year editor.” In addition to this he says “All of the greatest engineers in the world use Emacs. The world-changer types. Not the great gal in the cube next to you. Not Fred, the amazing guy down the hall. I’m talking about the greatest software developers of our profession, the ones who changed the face of the industry.” These look like some bold statements to me, but you can’t argue with the great software that Emacs has helped create. After reading several other blogs, and lots of comments, I seemed to gather this from the community: Emacs and Vim are old. They used to be the best way to get things done, and can still be incredibly powerful today. However, they are not essential to coding as some zealots would have you believe.

So why learn emacs? I want to see what all the rage is all about. After reading how powerful it is, the potential to never have to touch the mouse, and learn some lisp along the way to fully customize it; it just sounds like something fun to do. When I’m moving around my OS, or working on something, I love the long periods where I can alt-tab, ctrl-tab, and copy/paste my way around without having to touch the mouse. I always crave more hotkeys for doing things. From what I’ve seen, emacs has more than enough hotkeys to keep me sated.

Why not Vim? I had started trying vim about a week ago, before starting to explore emacs. It seemed neat, but the separation of normal and insert mode definitely bothered me a little bit. I didn’t like having to take my hands off the keyboard to hit escape in order to move around and actually use vim how you’re supposed to. After starting to use emacs, I immediately preferred the ctrl-x commands. At my level of understanding, the differences between the two aren’t really apparent yet either. All I’m doing right now is moving around using the keyboard, and using some basic commands (yank, kill line, jump up and down pages, etc.), so it’d be foolish for me to say that one is better than the other.

I’ll keep updating with my progress, and how I feel about emacs. Maybe I’ll switch to trying vim again, maybe I’ll abandon both. But for now, I’ve got my Emacs cheat sheet taped to the wall, and I’m excited to give it a shot.