The Midnight Room (Part 1) by Early Day (aka Pray for Plague) was released about a week ago. This post was initially about my failure to get past the first chapter, but since then I've managed to complete the game so I’m converting it into a mini-review.
I should preface my comments with the caveat that I haven’t played Next Door Girl: April, which I understand covers how April and Donald's relationship began. If I had, I might have different opinions about their respective characters (particularly Donald, who I took a dislike to in this game). However, my feeling is that if the information in NDGA was vital to understanding Midnight Room, it should have been included in this game as well. Not doing so needlessly makes the game less accessible to anyone who didn't play the prequel.
So, what is Midnight Room about? Only the first part of the game has been released so far, and the various plot threads have barely started to coalesce, but I would describe it as a mystery cum thriller. The mystery in question surrounds the shooting of Donald Hardeman by a masked intruder. The protagonists are April Cameron, Donald's much younger lover, and Raymond (aka Razor), Donald's estranged son. I'm somewhat reminded of Dangerous Assets, partly because of the dual protagonists and similarities in plot, and partly because both games are reluctant to let the player discover any plot-relevant information. The major difference is that so far Midnight Room's plot feels a lot more realistic (although the references to BunkMobile and Bunkmail strain that a little) and its efforts to block the player aren't nearly so blatant. However, in terms of its ideas and concept, I think that Midnight Room is one of the most interesting and original games to come along for quite a while. Unfortunately, those ideas are somewhat let down by poor execution.
The most obvious problem is the writing. It’s not bad by any means, but its good points are frequently obscured by the large number of errors. These include spelling mistakes, run-on sentences, sudden changes in tense, misused words, and so on. Once I started noticing these issues I couldn’t stop noticing them, and it really detracted from my enjoyment of the game. I know that I spend a lot of time banging on about the importance of proof-reading, but it really is the easiest way to improve the visible quality of your game. To be fair, it does appear as though the text was spellchecked, but there are an awful lot of problems (eg. homophones, random capitalisation, etc.) that a spellcheck won't catch.
Once you remove the errors from the equation, the writing is actually pretty good. Certainly better than that of Lamont Sanford or Chaotic, and vastly better than Vachon (the ne plus ultra of bad writing in AIF). I'm generally a bit sceptical of the use of multiple protagonists in IF, GoP3 being a good example of why. However, the fact that Midnight Room is written in a more personal style highlights the differences between April and Razor, and makes playing each of them a more distinct experience. If I have a criticism of the writing in Midnight Room it's that there doesn't seem to have been enough consideration of the differing requirements of interactive fiction as compared to static fiction. As someone who hadn’t played NDGA, the one that caused me the most problems was the slow reveal of the first two chapters. In static fiction the idea is that you don’t tell the reader everything at once. Instead you drip feed them the most interesting information in order to give them a reason to turn to the next page. However, unlike a reader, a player must make decisions on behalf of the protagonist. If they don't have the necessary information, they're reduced to guessing, which is not going to make them feel very involved in the game. That's slightly less of a problem in parser-based IF, because the player can at least explore the PC's environment to find the information they need. However, in CYOA you're expected to make plot-advancing decisions straight away. By the time that Midnight Room got around to telling me who April and Donald actually were, I’d already made several decisions on April’s behalf based primarily on guesswork. Evidently they were the wrong decisions, since they led to her getting killed. That's not the kind of thing that motivates you to keep playing. It may be inelegant, but in interactive fiction there’s a lot to be said for telling the player things up front.
In a post on his blog, the author mentions that he expects Midnight Room to be criticised for its lack of player agency. In itself, I didn’t find that to be much of a problem once I knew who the characters were. The fact that Midnight Room is a Twine game meant that I went into it with lowered expectations of how much agency the CYOA format would allow me. In practice, the stat-tracking system actually does a good job of providing the illusion of agency and disguising how linear the game really is. However, that system is not without its quirks. It's unclear exactly what each stat is supposed to represent, and which actions make them go up and down sometimes seems very arbitrary. For example, hitting someone in the groin reduces April's 'sexuality' for some reason. In the same vein, lying to the intruder increases April's 'honesty' (that might be a bug, but it's hard to tell). More egregiously, if April says that she doesn’t love Donald, the game takes it upon itself to decide that she’s lying, which is the one instance where I would complain about the lack of player agency. In a longer game there is the risk that this system would start to feel restrictive by preventing the player from acting 'out of character'. Within the context of this game however, I think the bigger problem is how little effect the PC's stats actually have. The differences between different stat levels are almost exclusively cosmetic, and as a result I stopped paying close attention to them. However, the fact that it is an aspect of the game that the player can visibly affect with their decisions makes it a useful source of player agency.
The problem I actually had was with the amount of *character* agency that April is given. Like most female protagonists in AIF, she’s almost entirely subservient (and deferential) to the male characters. This begins with Donald who, in keeping with the fact that he's old enough to be her father, treats her like a child. In the first chapter he ignores the fact that she has responsibilities of her own in favour of getting her to have sex with him. In the third chapter it's revealed that he’s made April the executor of his will without bothering to consult her about it, which is kind of a dick move given the potential legal problems it exposes her to. Not to mention that although Donald has bragged to his friends about April, he hasn't bothered introducing her to any of them, which makes me wonder if their 'relationship' extends beyond the bedroom. Consequently, his declaration of love seemed to me at best unconvincing, and at worst an attempt to emotionally manipulate a naïve and inexperienced girl. April does get to be a little more proactive in dealing with the gun-wielding intruder. It’s ultimately ineffective of course, but at least she’s not completely passive. Unfortunately, she then spends the third chapter waiting in the hospital while the other characters (who are all older and more knowledgeable than she is) are actually doing things. This only emphasises that April is the least mature character in the game by a considerable margin, and led me to dismiss her feelings for Donald as being little more than the childish infatuation of someone who has a thing for older men (cf. her reaction to Detective Keenan). It’s not that April isn’t an interesting character, and if this were static fiction she would make a good viewpoint character (with the expectation that she would get some character development as she 'grows up'). But for a game she’s far too passive for me to consider her an interesting protagonist at the moment.
Raymond/Razor, the second protagonist, is much more proactive and makes April look even more passive than she is. While April is moping in the hospital, Razor is breaking into Donald’s house and looking for clues. If anything, he’s almost too competent. Unlike my first experience with April, where I got her killed multiple times, Razor effortlessly deals with a knife-wielding prostitute. Likewise, he doesn’t encounter any difficulties when he’s breaking into his father’s house, although to be fair a lot of that is down to the fact that Donald apparently didn’t consider home security to be of any importance. Even an 'encrypted' file is no obstacle as Razor magically has the right 'tool' to deal with it. The problem is that while it might be nice to be so uber-competent in real life, it’s much less interesting in a game because it means that the player never has to struggle to achieve anything. Razor is also a victim of the slow reveal. When he's introduced it's not immediately obvious who he is or how he's connected to the story, which was one of the reasons that I initially didn't get past the first chapter. We do eventually discover that he's Donald's son, but that's about the only solid fact that the player does learn. His business is only described in vague terms, we never learn why he’s estranged from his father, and so on. These are things that Razor knows but which can't be explained to the player for some reason. NGDA might hold the answers, but that's not a good reason for denying them to anyone who hasn't played that game. The result was that I never became that interested in Razor because I was never allowed to learn anything interesting about him. The limited information that is provided just makes him seem like a generic badass, right down to his silly macho nickname.
In technical terms, Midnight Room is free of obvious bugs, although there are a couple of unobvious ones when you look more closely. There are also a number of continuity errors, sometimes within the space of the same paragraph. The third and fourth chapters try to mimic regular IF style gameplay, by allowing the PC to move back and forth between rooms. Frankly, this doesn't seem to be something that Twine is particularly good at. It's often not clear which part of the text is the room description and which part is something new happening which makes it all a bit confusing.
AIF has a tendency to just rehash the same tropes over and over, so it's refreshing when someone appears with new ideas and a fresh approach. The May to December relationship is an excellent example of this, since it's not something that has ever been dealt with in AIF to my knowledge (although I was disappointed by how superficially it was ultimately handled). However, so far Midnight Room is only *potentially* interesting because so much depends on how part 2 resolves the story. As I’ve mentioned, there’s very little definitely established about the plot. The most we learn is that Donald might have been involved in something shady, and he might have been cheating on April. I understand the desire to prolong the mystery for as long as possible, but it’s a risky strategy in my opinion. The danger is that the longer the solution is delayed the more likely it is that it will be anticlimactic when it is finally revealed. However, even if the mystery does have a satisfying solution, if April continues to be a virtual bystander while Razor barely needs the player's help at all, it still won't be a satisfying *game*. The level of ambition and originality that the author has brought to the first part of Midnight Room does give me some hope that he can pull off the second, but the proof will be in the pudding.