IFDB page: Internal Documents
Final placement: 19th place (of 30) in the 2003 Interactive Fiction Competition
After Curse Of Manorland, it was a relief to start into a game that seemed at least to be composed of coherent sentences, although the incredibly long, clause-upon-clause opener was a bit of a red flag. Still, the premise of the PC as civil servant investigating electoral fraud in a small town seemed like it had potential, and I began the game excited and interested. Sadly, Internal Documents fails on a number of levels, and by the end of the game I was just annoyed and disgusted, typing straight from the walkthrough. Some of its problems are pretty standard, such as the writing issues — typos and grammatical errors are common, and a noticeable number of sentences just fail to make any sense.
The coding is similarly problematic. At one point, I tried to read a document that (according to the hints, anyway) contains important clues. Here’s what happened:
>read manual Chapter 1 reads: Xy Yi Z Zm x b k a c k a d k a e k a f k a g Y k a h ( k a i ê k a j y k a k 0y k a l 1u k a m 2o k a 3c 3u
I think this freakiness happens because of a particular quirk in Inform, because I vaguely remember using it intentionally in LASH to demonstrate illiteracy. However, in this game it obviously wasn’t intentional, and even the most basic testing should have revealed that it wasn’t working properly.
Another one of Internal Documents‘ big problems is one that I’d like to discuss in more depth, because I think it’s a signpost for anybody who wants to design good IF. Basically, the mantra is this: THINK LIKE A PLAYER. This game doesn’t, and falls down badly as a result. Here’s an example: early on in the game, I walk into a bar, and after the room description and a bit of incidental business, I get this:
The bartender cuts in and asks you, “So what brings you around here?”
Remember now, I’m a player. I’m trying to be immersed in the game’s fictional world, to do what the PC would do. So I type TELL THE BARTENDER ABOUT DUE DILIGENCE. In the game’s words, “This provokes no response.” So then I try to TELL THE BARTENDER ABOUT several other things, with no results. I proceed to strike out with TALK TO BARTENDER, ANSWER BARTENDER and ASK BARTENDER ABOUT <a variety of topics.> With sadness, I realize that the game has asked me a question but has no intention of providing me with a way of answering. So much for immersion.
See, as a designer, I might know that the bartender isn’t useful, and therefore give him only the bare minimum implementation. Maybe I want him to seem a little bit alive, so I give him a line of dialogue when he’s first sighted, and maybe I don’t so much care just what that line is. As a player, though, I have no idea whether the bartender is useful, so if he has some dialogue that encourages further interaction with him, I’ll take it as a cue that indeed he is supposed to be interesting. Then, when I encounter his minimal implementation, I’m annoyed and disappointed. The solution? Make the bartender surly and uncommunicative, so that his thin implementation seems like a natural aspect of his character. Your game is going to shape your player’s expectations — there’s no way around that, so it should shape them to its advantage rather than to its detriment.
On another “think like a player” point, remember that what’s fun for you isn’t necessarily fun for your player. You may really dig writing up three hundred slightly varying descriptions of the same sort of rather ordinary room, but by all that is holy, it is not fun to walk through them. Don’t set up situations where the player ends up wandering huge swaths of useless, pointless territory, or has to sit around or walk around aimlessly for dozens of moves waiting for a random event to trigger. That’s not fun, and it produces no particular emotional effect except for irritation. That’s not what players want from their IF.
Another aspect of thinking like a player is recognizing that if the game prevents a particular action repeatedly by use of unlikely coincidence, the player is going to believe that this action is forbidden and unimportant. If you think that the player will later be moved to attempt to do that forbidden thing as a way of solving a major puzzle, you’re probably wrong, especially if much more sensible solutions are disallowed. If I the player end up looking at the walkthrough, you want me to think “Oh, of course! Why didn’t I think of that?” Instead, I end up thinking, “First of all, I thought the game didn’t allow that action, and second of all, how does that make sense as a solution to that puzzle anyway?”
There’s a larger point here, which is that implementation generates expectations. If 99 out of every 100 nouns in your game are unimplemented, as is indeed the case with Internal Documents, I will come to expect a very bare-bones experience. I will not type out a long and unusual command construction, because what possible reason would I have to believe that the game would understand it? If you do expect me to do things like that, the game becomes either an intolerable exercise in attempted authorial telepathy, or else it ends up having to dole out sledgehammer-like hints (such as 1-2-3‘s notorious “Don’t you want to ask me about her breasts?”)
Before you even begin coding, think about what it will be like for players to experience your game, and if the answer is “boring”, or “irritating”, or “confusing”, stop those problems before they start. Otherwise, you end up with something that’s great fun for you, but not for anybody else… something like Internal Documents.