« Elevator Park | Main | Wii will develop surgical skills »
Jul
19
Designing my own ARG (post-mortem)
So this summer I am teaching at the Academic Talent Development Program (ATDP), hosted by UC Berkeley, as usual. And as usual, the class meets MWF, with one Friday falling on July 4th. The program's policy is to cancel the July 4th session and make it up at a later date. I was struggling to think of something fun for our class to do at this make-up session, since the computer lab we are usually assigned would be unavailable.
Originally I wanted to take them on a field trip, but the only viable candidate (the San Jose Tech Museum) was prohibitively far away. So instead, I came up with the seemed-sane-at-the-time idea of designing an alternate reality game for the students to play, revolving around the UC Berkeley campus. The class would be split up into groups of 4-6, each led by a TA, and given a laptop from ATDP's mobile computer lab for accessing the online portion.
It took me about two to three weeks to come up with the entire game, from the puzzles to storyline to actually planting online hooks and goals. I looked at some existing ARGs for inspiration, but for the most part I think the game is fairly original (at least I think so -- I admit I did not do very much indepth research).
The day before yesterday (07/18) was the make-up session where the game actually took place and I was relatively pleased with the results. In general, I think the puzzles might have been a little on the tougher side, but that did not stop a couple groups from steamrolling through it anyway.
Some notes on the process of designing TAJA (TIC And Java Adventure -- the instructor of the Java class wanted his class to do it too, since he was afflicted with the same make-up day woes as us)...
- I am SO GLAD that I did iterations! I had two meetings with my TAs, where I had them playtest the game in its current state and each time, I had more stuff for them to work on. Their feedback on which puzzles were too difficult and why allowed me to throw out / improve bad puzzles and replace them with better ones.
- Along with the topic of playtesting... especially with an ARG, it's hard to get an accurate impression of how difficult/easy a puzzle is unless you have a tester who has never encountered any of the previous iterations. So, with my limited pool of testers, I basically got only one test of each puzzle, which was lame but better than nothing.
- I designed nearly all the puzzles from behind my computer screen: i.e. I never went on location to any of the places I ended up making the students go. This ended up being really bad, because for a couple of the puzzles, the answers were actually possible to find entirely online. I should've been less lazy in this respect and took more time looking for less networthy locales/sites on campus for the students to visit.
- My design documentation started out somewhat disorganized at first, but increased as I had more things to keep track of. Finally I ended up writing a formal script that kept track of all the elements, which sites were located, etc.
- The story of the ARG is relatively simple and -- yes -- fairly campy (just because the nature of the community prefers humor to drama and really good drama takes much longer to write). At the root of it was an online cult that had a secret Master Plan for world domination. It turns out that the Java instructor and I had foiled the cult's plans by stealing their only copy of the master plan, splitting it up into six pieces and scattering them all over the internet. One of the cultists, in the guise of an ally, recruits our classes' students' help in recovering the master plan pieces, finally turning on them when the whole plan has been reconstructed. The students get back by ... well, stealing the plan again and getting the Judas-cultist kicked out of his exclusive little club. Yes, it's very, very campy, but it worked out. ::lol::
- Over the entire adventure, students visit/use: e-mail, Blogspot, YouTube, Photoshop, TinyURL, any audio player, PHP forums, SendUIt, any program capable of opening password-locked zip files, and they are made to step through an extremely simple Java for loop. Answers to various puzzles are found in four different on-campus locations. I made six different pages and/or sites (including fake e-mail clients and a fake image hosting site >_< ) for students to "hack" into and registered four different (free!) subdomains in order to make these all look like they were owned by different people.
- Some parts of the game were nonlinear! I specifically structured it into three basic parts: first three branches, second three branches, and then the finale. Basically students arrive at a node which contains leads to the first three pieces of the master plan image, but they can do any of these three clues in any order they like. However, they need to have all three pieces before they can continue onto the next node. The students needed all six images in order to proceed to the finale.
- Almost everything was automated, but there were a couple parts of the game that had to be done manually... specifically, the e-mails that the cultist character sends out. The very first one is automated, but the second and third ones had to be sent out manually by myself during the game itself, as groups reached the checkpoints. I probably could've automated all of it, but was running up against the deadline and didn't want to spend time figuring out how to parse through and respond to e-mails.
- Because of the above, I was a little frightened at the idea of being completely blind to where players were in the game... how would I be able to determine when someone "won" the game and merited a victory e-mail? This was solved last minute by implementing an extremely primative, but functional, tracking system. Basically I stored a cookie containing the player's e-mail address at the very beginning of the game and then access it on my various custom-built pages throughout the game. This allowed me to see which player was at what page, which I would send in a quick e-mail to myself. It actually worked pretty well, plus I learned way more about browser cookies than I ever thought possible. :P
Thus far, the responses have ranged all over the spectrum... some students really loved it and had a lot of fun, while others have admitted to completely hating the experience. In general the reactions have been mostly positive, with a couple of pockets of frustration at the more difficult puzzles. Perhaps not so surprisingly, my students did not enjoy the one puzzle that was about Java because they couldn't solve it themselves and Java students seemed to struggle with it. (I did try to offset this by having one puzzle that I thought was TIC-centric, but since I'm not getting direct feedback from Java students, I don't know if they were equivalently irritated by that one puzzle.) Many stated that they would play the game, or one like it, if given the chance. Some said they would prefer a regular scavenger hunt.
Reactions to the story specifically were polarized. Either the players absolutely loved it and thought it enhanced the entire experience by a lot (campy as it was), and the remainder thought it was completely pointless.
Admittedly it's been a teensy bit painful to see the really negative reactions (selling your soul to make something big does that), but I guess it's good that I get used to seeing such things, since it's not going to be possible to please everyone and negative criticism is just as important as positive for improving one's work.
Overall, this has been an extremely positive design experience for me, in coming up with good puzzles, learning how to get useful feedback from testers, planning and documenting, and actual development and execution.
Formal thanks to my TAs who playtested for me and put up with my nagging questions about "how they felt when...". Also to my co-instructor, who trusted that I knew what I was doing even when he had no idea what I was proposing at the very beginning. XD And finally to Alex, who followed some of the players around with a mini-DV to get video documentation of the experience.
