Welcome to Game Club, where Joanna Price and myself exchange letters about a game we’ve jointly played. This week, we’re discussing 2064: Read Only Memories, the remastered version of Read Only Memories from developer Midboss. All the letters are collected here for your perusal, as well as some of the many screenshots I took. As usual, the letters are chock-full of spoilers.

(Also, I’m using alt text in the images).

Letter 1

Hey Joanna,

I have a lot of thoughts about this month’s game, Read Only Memories. For those who haven’t played it, Read Only Memories is a Kickstarted graphic adventure game closely modeled after Hideo Kojima’s Snatcher (specifically the 1994 Sega CD release, as opposed to the original 1988 PC-88/MSX-2 release). Which is to say: it’s a very Japanese idea of a graphic adventure, with an interface designed for a keyboard/console controller rather than a mouse, and with a disproportionate amount of time taken up by talking heads (a sort of precursor to what would eventually become the visual novel genre). The action takes place in Neo San Francisco in 2064, with every cyberpunk trope you can think of making an appearance and driving the plot (more on that later). It also shares Snatcher’s easy, logical puzzles and general avoidance of fail states.[1] The developers have gone as far as to give the game the same graphical fidelity as Snatcher (actually, a bit worse!) which is probably a budget consideration as much as an aesthetic choice. The game was recently re-released as 2064: Read Only Memories, which adds voice-acting and a new epilogue.

In theory, I’m that target audience for this game. I’ve played and enjoyed Snatcher[2], I like pixel art, I like cyberpunk, and I am particularly forgiving of small-budget indie games when the developer’s earnest enthusiasm shines through (as it very much does here). And yet while playing Read Only Memories, I had trouble getting into the immersive space, and alternated between boredom and eye-rolling.

The primary, and most obvious, issue with Read Only Memories is that it’s so indistinct. A lot of games we’ve covered for game club put a disproportionate focus on setting, plot, or character; ROM focuses on all three equally, and they’re all overly familiar. The setting is the most generic, pseudo-futuristic cyberpunk city you can imagine; the only thing that sets it apart is its intentional rooting in San Francisco (where the chief developer lives), and even that is largely window dressing and doesn’t really address any situations unique to San Fran. It occurred to me over the course of the game that somehow, every starving artist and full-time hacktivist managed to afford rent in SF, despite the fact that the little info we’re given tells us that SF has turned into a corporate state where all public services have been privatized.

The great thing is that I didn't even catch this image on purpose. FRAPs was set to auto-capture at certain intervals.

I’ll admit that this is a nitpick; the problem is that every single element of the game is nitpickable. Characters are decently written and sometimes even charming, but all suffer from being fairly obvious. The plot unravels really slowly, and reveals itself to be third-grade Neuromancer knockoff when it gets to the reveal. The big conspiracy (SPOILER WARNING) is that a distributed AI will monitor people’s use of the future-internet and change what they see based on their perceived preferences and/or what those in power want them to see. This isn’t cyberpunk; this is real-life 2017. It’s called big data, and love it or hate it it isn’t exactly a hidden conspiracy.

This situation is made even worse by over-obvious referencing. The anti-hybrid group is called Human Revolution. The update that gives all ROMs sapience is called Wintermute. There’s a guy named Leon Dekker who (UNSURPRISING SPOILER) isn’t human. It’s artless and blew whatever immersion I had every time one popped up.

On the one hand, this is a really stupid name for an activist group. On the other hand, lots of real-life activist groups have pretty stupid names.

The game does exactly one interesting thing with its world-building, and that has more to do with its crowdfunding origins. This wasn’t Kickstarted by an establishment developer, but rather a group of conference-runners, who created and hosted GaymerX, the first large-scale gay gaming convention. It was thus inevitable that it would be pitched to the GaymerX crowd as a game that would be unusually inclusive of varying gender and sexual identities. Its novel way of doing this was imagining a future in which oppression along these lines had totally disappeared.

It works well. Apart from being pleasant escapism, it can dodge heternormativity by not making a character’s gender or sexual orientation their defining status, merely a sidenote. Likewise, the characters don’t all have to be written as oppressed. There are a million ways they could have been tokenistic, or agenda-ed at the expense of fleshing out characters, and it deftly avoids all of the obvious pitfalls. It’s a pleasure…until it undermines the entire premise by focusing a lot of its runtime on the oppression of ‘hybrids,’ humans who have modified themselves with animal DNA.

Basically: it goes out of its way to use a framework other than identity politics to look at the lived existences of characters with real-life identities, than invents a fictional identity and brings the framework right back. It digs itself a deeper hole by having the anti-hybrid group called the Human Revolution, which instantly recalls Deus Ex: Human Revolution, a game that did a remarkable job of fleshing out the arguments on both sides of the debate and making them both compelling. Here, the anti-hybrid people are portrayed as a mix of malicious bigots and ignorant dolts.

I have a lot more criticisms in this vein, but they all basically boil down to the same thing: an earnest, competent execution unwilling or (more likely) unable to push beyond the obvious. There is something to be said for comfort food, and at its best this is indeed good comfort food; one of its hidden charms is writing an enormous amount of verb-noun interactions of the sort graphic adventure games had, and I enjoyed using my carton of spoiled milk on every character and piece of scenery in the entire game and getting a unique, often punny response. But then I find myself in another piece of uninteresting exposition, clicking through dialog sentence by sentence, and my warmth turns to exasperation.

That’s more than enough for the first letter. Let me know how it went for you!



There is a unique response for using spoiled milk on every hotspot in the game. There are hundreds of them. It's pretty incredible.

Letter 2

Hey Dylan,

I think I have a lot of the same complaints, but how I think of it is that it’s just too predictable. I don’t know how possible it is to write a story in which an AI does something unexpected anymore. I believe that the “every possible ending” solution is a really terrible way of approaching the “it’s all been done before” thing, like it feels like story infidelity. The game has several endings, and players who have finished one or two of the endings but don’t want to play through the rest of them, can see a table of them here.

My problem with the story is thus twofold: in the first place, as you note when you say it is “indistinct,” I found myself not super enthralled, and often able to predict the story. You call this a nitpick, but I would argue that it’s a big problem. In a narrative heavy game, not committing to an interesting narrative is a problem, and if you’re retelling an old story — which is wonderful thing to do, don’t get me wrong — it’s important to do it in a way that remains relevant and immediate to the player. Secondly, to my mind, instead of committing to the narrative (what I am calling “story fidelity”), this game simply offers all the endings. I think “an ending for everybody” is a poor way to write a story, and a bad reason for branching narrative in a game. It struck me as a lazy solution to an admittedly tough challenge — writing a great game that is so story focused. I mention the toughness to explain that I am not sitting here fuming at the writers or developers, I’m just noting that the game didn’t make the grade for me. I don’t believe that it is right to “No Man’s Sky” a developer, if I don’t like a game, that doesn’t mean that the devs are terrible or that they’re required to do anything about it.

Turing, Humanities Professor.


I’m curious why you think this way of not using identity politics is particularly interesting, given that games really only became extremely political not that long ago, as far as I can tell. The older story games don’t feature political views very heavily even when they express characters or groups with political views. But perhaps that differentiates indie, comparatively low-budget games from games with either Big Name Devs or from Big Name Companies. Perhaps one of the ways indie gaming has gotten a foothold and an audience is with political agenda. At any rate, I didn’t find the lack of identity politics all that refreshing because as you rightly pointed out, the framework is still there.

I actually¬†loved¬†the voice of the AI Turing, which made up sometimes for the bleh dialog. One thing that drives me bonkers is when the player is kept from important information because the character who gives it just won’t get around to saying it before you go through X number of dialog trees which tease it. The other highlight for me was the number of interactable objects, and the narrative that went along with them. Normally I’m not click happy, especially if I’m really into the story. But I was kind of bored with the story, so I spent some time clicking around and I enjoyed that.

I think for me, my time would have been better spent on a different game. I liked J.U.L.I.A. Among the Stars mechanically more, and it had a similar AI vs human trope and similarly developed (or undeveloped) characters. [N.B. the actual plotline of the J.U.L.I.A. is not similar, but they have questions, themes and a strong narrative focus in common] That said, this might well be a good game for people who don’t have a ton of experience in narrative gaming. Like the “Settlers of Catan” of story games. Players who have spent even a a little time with euro games move quickly past Settlers. But Settlers is an excellent introduction to euro gaming for people who have played mainly American family and party games, in that it presents many of the common mechanics about as generically as possible — when playing future games, various mechanics will be familiar. I felt this way of about Read Only Memories, I thought it presented a lot of common features of story based indie games — dialog heavy, developed plots, lots of clickable area, puzzles, etc — without doing anything particularly original or new. If it were the first of this general type of game I had played, I’d probably have had a much better experience. But for someone with that experience already, J.U.L.I.A. Among the Stars has similar narrative themes, and meets most of the same player interests and needs, while having better mechanics and more draw, in my opinion.




Letter 3

Hey Joanna,

To take things in order: while I agree with you that the game is too predictable, I actually had no problem with the multiple endings – and was only tangentially aware they existed, since I only played the game through once. Multiple endings have been around for as long as video games have, because in a game that responds to play choice, it is only natural that the conclusion of the game changes based upon those choices. I agree with you that they’re a weird fit for Read Only Memories, because by and large this is a linear game without substantive or difficult choices, but in the abstract lots of games have successfully implemented multiple endings, and I don’t think that implies a lack of commitment.[3] But of course I can only speak to the ending I got. Did you play the game multiple times? Did you ‘save and reload’ to see what would happen with different endings? What drove you to do these things, and did it negatively impact the weight of your original ending? Have you played other games that you thought handled multiple endings better (for me, Cyanides’ Game of Thrones comes to mind as a game that let me choose the perfect ending to the aspects of the story I was focusing on)?

But overall, I think we’re on the same page: the surest way to genericism is to try to be all things to all people, and Read Only Memories seems unwilling to alienate any potential audience by being in any way distinct. To answer your question about how it handles the politics, it is true that games have historically been ‘apolitical’ – but they have done so by basically only showing and focusing on certain segments of the population (white men who use violence to solve problems) instead of looking at humanity more broadly. This was not a commentary or even self-aware, just a combination of the developers unconsciously recreating their own normal, and publishers consciously ‘playing to the [existing] market.’ So it’s unusual and unexpected when a game founded on the promise that it will do the *exact opposite* attempts to use the same framework (of characterizing its world and characters as unremarkable and normal, rather than loudly declaring itself more diverse). As someone who plays a lot of indie games, I can promise you this is the not the norm. But, as I mentioned and you so eloquently characterized, it uses a different lens for the exact same framework, so it’s only interesting on the surface; and in a sense it’s even worse, because that framework is intentionally unexamined by virtue of being diegetically unremarkable. If nothing else, there’s a lesson here: applying the thoughtlessness of the majority to minorities may be subversive, but it’s still thoughtless.

I definitely came around to Turing over time, and agree that ultimately the voice made an unremarkable character memorable, and is a good example of how distinct performances can really lift a modest script. As for your concern about dialog trees – it didn’t really bother me that much because I’m the kind of person who wants to follow every dialog tangent anyway, but in general it is always annoying when any work acts as if the player is dumb, or when the player character/narrator lags behind the player/reader in what they perceive.

In the epilogue, Turing becomes addicted to Team Fotress 2.

I appreciate that you are trying to give the game an out as a ‘starter game,’ but ultimately I think that there are simply better-suited games (as you yourself note). The Settlers of Catan[4] may be incredibly vanilla, but it’s some of the world’s best vanilla, with a simple-yet-deep design refined and polished through playtesting. Read Only Memories is mediocre vanilla, and the world is so awash in good-not-great, straightforward point-and-click adventures that I’m not really sure why anyone would start with this one unless it was really important for them that the game featured a lot of queer characters.

Incidentally – if you’re looking for a game that goes above and beyond in exploring the questions of AI, I’d recommend The Talos Principle. This puzzler (penned by Tom Jubert and Jonas Kyratzes) brings tons of relevant philosophical and scientific writing to the fore and tries to go about as deep as a game can go without being completely inaccessible to a lay audience. It definitely doesn’t skate on the inherent appeal of big philosophical questions.

Finally – I know we’re putting this letter series on ice for now, and if we ever do any future games writing it will probably be in a different format. I wanted to say what a pleasure it’s been to hear your insights on these interesting games. There’s something wonderful and comforting about sharing the solitary experience of playing single-player games, and I look forward to whatever we do in the future.


I will never tire of this meme.

Letter 4

Hey Dylan,
I think you’re right that many games implement multiple endings successfully. The linearity of this plot line does not inherently mean there is a strong commitment to the story, but Read Only Memories doesn’t have a whole ton else going for it– it’s not the excellent graphics or amazing combat, the open world or the fantastic multiplayer. It’s a story driven game. From that perspective, given that the plot doesn’t branch a whole lot, it seems like the designers either had to commit to the story or make a less good game. For me, the six main endings were disappointing because they felt like a last minute decision to not commit. There are a few “secret endings” that I read about later, and those are a horse of a different color for me. The difference is that with the original six endings, it was taking away from the central thrust, but secret endings add another layer of interaction as opposed to diluting the story. In particular, due to the philosophical nature of this story, it seems to me that what I call “narrative fidelity” is even more important. The game is asking interesting questions but it feels a little bit like wandering around when it doesn’t “bring the player home,” I end up sort of left with the feeling that I never finished engaging with the story. The first version I played, Turing hates me at the end of the game, and I had a previous save that I called up and did things differently. The other endings I just read about.

It’s just a really strange story to go the generic route with. It’s not like you have a dwarf, an elf, a bard and a warrior mage trying to save some fantasy land here, I didn’t even know you could be generic about the question of sentient computing. So it’s kind of intriguing in a way, and kind of hilarious, this weird commitment to tepid storytelling. Also the use of Turing was interesting, because the very first thing the player does upon encountering Turing is distinguish the character as AI — but the Turing test is not to create an AI that is as intelligent and sentient as people, but to create one that is indistinguishable from people. This haphazard philosophy thing is so weird, I find it kind of amusing. It may be the most permissible in terms of identity exactly because it is so accommodating. It does make me wonder if they were actively trying to avoid a fight.

I own the The Talos Principle but I’ve never played it. I think J.U.L.I.A. and Read Only Memories have a similar lightness to them, they ask about the same amount of work from the player, address many of the same themes, and J.U.L.I.A. is just a better game. Which is to say: I do think they’re comparable. I donno about The Talos Principle, but I’ll take your word for it!

This completes Games Club for now, I believe. But I look forward to reading your future writing, nonetheless. Thanks for writing with me!

Randomly touching everything really is the highlight of this game.

  1. [1]This was radical in the context of American adventure games in 1994 – it was only a few years earlier that The Secret of Monkey Island had invented the death-free adventure, but even that still had plenty of difficult, adventure-game-logic puzzles. Of course, in 2017, Read Only Memories is just doing what everyone else is doing
  2. [2]Though it’s Kojima’s Japan-only follow-up, Policenauts (1994), that is near and dear to my heart.
  3. [3]One of my pet peeves in video games is a particular bad way of implementing multiple endings; making the default, achievable ending a ‘bad ending’ that is intentionally unsatisfying, and forcing the player to replay the game and jump through various hoops in order to get a positive ending or (worse) and ending that actually makes sense. For whatever reason, this has traditionally only been done with Japanese games (perhaps because this is perceived as increasing the ‘value’ of the game). But thankfully, Read Only Memories doesn’t do this.
  4. [4]Since renamed ‘Catan’ by the company who publishes it, in order to create a Brand it can slap every which way – nevertheless that it makes the game sound like it’s about geography rather than the settlers!