Reviews by Sam Kabo Ashwell

Set in a mildly steampunky high-fantasy world, First Snows establishes a strong opening tone: a grim, dark winter, a town whose inhabitants are curt or suspicious, a sinister mystery, a certain emotional numbness.

As so often, the promise of the initial mood is the best part; the emotional numbness and suspicious inhabitants turn out to be the product of scanty characterisation, and as the mystery unfurls, the prose proves inadequate to render shock and horror. The story does nothing to build on your initial motivation, and the worldbuilding doesn't become any deeper after the early game, so by the endgame - it's about medium comp-sized - you're still wondering why you should care.

There's a shade of overwriting early on - The clipped, abbreviated sentences were his normal manner of talking, and you responded in kind with a brief “No” - which suggests a more character-oriented plot. This fades rather quickly as you get into the game proper. That's fine; if dialogue's not your strength, avoiding it is a legit approach.

Mechanically, it's pretty old-school: you will get stuck quite early on unless you vigorously search for crucial items, NPCs are stand-offish, there is timed death. There's a nod to the Enchanter scroll-based magic system, but the game isn't long enough for this to become a regular feature, or for it to really qualify as the Thorny Old-School Puzzler that it's drawing on. But it's still difficult enough that you can expect to be stuck quite a lot, and for me it didn't generate anywhere near enough motivation to stay with it long when the going got tough. (I finished with a walkthrough -- hat-tip David Welbourn.)

So to me this felt as though it had at least one foot in the era when people would and did persevere with games even if they were sort of dull, just because they were games. Perhaps that's overly harsh; in the mid-90s, say, this might have earned some quiet praise and a place somewhere near Wearing the Claw. And I'm a long way from being the ideal audience for this. But still, it feels like a piece with the soul missing, lacking an essential Cool Thing to make you stick with it.

Eric Eve's games tend to be solidly built and to follow well-established, orthodox design patterns; The Elysium Enigma, a sci-fi mystery, is no exception.

Enigma is structured around simple, traditional puzzles in a fairly non-linear arrangement. There are three major NPCs, all responsive conversationalists. In terms of design, tech and careful diligence, it's an impressive piece of work: its threads never seem to trip over one another, the plot inobtrusively avoids ballooning without making play feel confined, and conversation updates smoothly with knowledge and plot advancement. Interaction is very much of the traditional variety: discover hidden items, find keys and passwords, fiddle with electronic devices, find a vehicle to overcome a barrier, and so forth. (While there are more involved puzzles for a higher score, getting a winning ending is quite easy). Although you're exploring a village and environs, it's a largely deserted environment, major characters aside, and the player's adventurer-style trespass, vandalism and theft goes largely unremarked.

The weak points of The Elysium Enigma lie in the writing. I don't want to paint too strong a picture here, since for the most part the prose does the job that's required of it. The standard IF fare - descriptions of rooms, objects and actions that straightforwardly negotiate these - is mostly good solid work. (There are occasional quirks of overspecification where more natural speech would have worked better, but nothing egregious.) Where it falls short is character writing.

Characters are used effectively to deliver key information and direct action, but this often comes across as highly artificial. (This is partly because of the brave decision to keep the three central NPCs on-stage and highly responsive for the entire course of the game.) Even allowing for this difficulty, they're all rather two-dimensional.

Take the protagonist, Andrew: tall, athletic and good-looking, a bit contemptuous of Elysium's locals. His reactions to events generally go undescribed, which is a standard approach: show what the player character sees, imply their inner state, avoid directly reporting it. But when we do get hints of the protagonist's state of mind - his final words at the game's conclusion, for instance - they can seem massively off-key.

And this is a problem, because the NPCs aren't incidental to the story. Indeed, it deals with perhaps the most difficult of NPC dynamics, romance. The game's problems in this department are, I think, perfectly summed up in a single moment. It takes a little buildup and is spoilery as hell, though, so bear with me.

(Spoiler - click to show)Early in the game you encounter Leela - young, attractive, apparently an outcast from the village. She's a wide-eyed, curious ingenue; she asks you for food and clothes, bathes naked and then continues to follow you around in that state. Once you've provided her with food and clothing, she expresses romantic interest in you. If you ask about a relationship, she strongly suggests that she'll have sex with you a little later. A little later, you're exploring an underground bunker; she throws herself down on a mattress. You're discussing the implications of the exciting mystery you've uncovered. At this point, she's dressed in a sheet held together with a couple of safety-pins. And...

...and nothing. Not so much as a fade to black. There's no way to initiate sex directly, and more circumspect methods (more kissing, lying on the mattress) seem, in context, awkwardly chaste.

I'm definitely not arguing that there has to be a sex scene here - you could fade to black, have her reject the PC, articulate some motivations for the PC to keep it in his pants, or rewrite the scene so that it didn't lean so obviously in that direction. But as it stands, this makes the whole scene seem like an awkward lapse in characterisation.

More generally: Leela's character is a recognisable Type from SF of a certain awful era: a wide-eyed ingenue, in need of rescue, childish, curious, sexually liberated yet virginal, spirited yet biddable, given to following the hero around. Now, arguably the story's point is that this character is a fiction; perhaps the intention was that Leela's character was meant to look like an implausible male fantasy. But this is rather undermined because Leela's true identity, Anita, is the other side of the same coin, the Cold, Calculating Bitch. Now, this is a boring stereotype, to be sure, but it also kind of torpedoes the emotional impact of the game's final Big Choice. Andrew has to choose whether to bring Anita in or let her go; for this choice to have weight, it requires the player to be invested in Leela/Anita. The problem is that Anita, the real one, is less complicated than Leela: all we see of her is the heartless schemer of the official report, and the spitting ball of hate when she's captured. For me, the choice of whether to let her go read less like a moment of anguished indecision and more like relishing a moment of power over a bad woman.

The game initially suggests that it's going to be social science fiction. You have to deal with a strongly technophobic culture, and the implication is that you're balancing the need to respect that culture with other concerns: the need to protect individual rights, political and military objectives, and so on. In the event, though, this isn't explored as much as the initial setup might suggest; you only really deal with one Elysian, there's no story incentive not to violate their cultural norms, and the real plot is about espionage. The Elysian culture is rendered at a Star Trek planet-of-the-week level of detail: one big cultural hook drives all the conflicts, and everything else is a bit generic. (Of course, a detailed Le Guin-ish culture-building piece really isn't what the author is interested in, even if I'd like him to be: the real focus is geopolitics and espionage. (Spoiler - click to show)But Leela's deception relies on you misreading the genre, thinking that you're in a Culture vs. Individual story rather than a Great Game one; so perhaps the mystery could have been preserved for longer by continuing to develop a culture-oriented plot.)

Finally, the central premise of the plot feels a bit off.(Spoiler - click to show) We're meant to understand that Anita plans to seduce Andrew in order to extract tactical information from him. Her mission is covert observation; the risks of exposing herself are very high, so the information should be a) very valuable, and b) unobtainable by safer means. In the event, she doesn't get all that much information from Andrew, and most of it seems like things that could have been learned covertly (if perhaps not so quickly). So the upshot is that Anita seems like an incompetent spy.

So while I could respect Enigma as a piece of design and implementation, I found it very hard to enjoy as a story.

This is a game that is centrally about trolley-cases: ethical scenarios with binary options, designed to get at the heart of a problem of ethics. Rather than circling around a single issue, it aims to give a survey of an entire landscape; as philosophy IF goes, this is an approach closer to The Chinese Room than De Baron. The difference is that where The Chinese Room presents its material as a goofy wonderland, The Test makes some effort to present each scenario with a degree of serious realism. I've written elsewhere about Test's shortcomings as a philosophical piece, so I won't harp any further on that subject.

As a work of participatory narrative, however, it's also a bit disappointing. The premise makes it very clear that every scene will boil down to a single binary choice; as in CYOA, this can easily lead to disengagement from the rest of the material, particularly when the material is intellectually or emotionally challenging (which trolley-cases damn well should be). The fact that the scenes are obviously unrelated one-off scenarios also makes engagement more difficult: it's easier to take a decision seriously when you expect to have to live with the consequences for a while. The game aims to create realist, flesh-and-blood characters, thus lending more weight to its scenarios; but it's hard to develop a sense of attachment to a character when you're aware that they exist for a single purpose and will be discarded once that purpose is complete.

Finally, the ending, in which it transpires that (Spoiler - click to show)the protagonist is an artificial intelligence being evaluated for ethical attitudes, is too brief to be satisfying, and feels like a cheap narrative justification for a hodge-podge structure. To be really interesting, it'd have to explore things considerably further - what sort of world has a need for robots with a varied range of ethics? what sorts of things would result from following the ethical compass you've defined? how do these values conflict or cohere?

Given its design premises, Test is competently executed; but those premises make it prohibitively difficult to accomplish its goals.

It's not hard to discern the game that this wants to be: a setting-focused, slow-recovery-from-amnesia, evil-science-secrets-uncovered, atmospheric-horror piece akin to Babel, drawing on a potent setting: the ruins of the Chernobyl Exclusion Zone. S.T.A.L.K.E.R.: Shadow of Chernobyl and its sequels have explored very similar territory, albeit with more guns.

Pripyat has some bells and whistles - music, a scattering of graphics - but the game itself reads like the rough outline of a parser IF game, rather than a complete implementation of a choice-based one. The protagonist's actions are all adventure-game standards: examining scenery, reading diary pages, collecting inventory items and using them on things.

It's very short: every possible action is a key plot-coupon action, except for one that instantly kills you. Horror, and especially this kind of setting-reliant piecing-together-the-fragments horror, needs a slow build to be effective. The impatient pace means that setting and atmospherics -- which would seem like strengths of the premise -- get neglected. And the speed of plot delivery means that a perfectly serviceable plot ends up feeling facile.

So there are some sound instincts here, but a great deal more work needs to be done.

Written for the Ludum Dare game jam, Evolution: A Parallel Narrative is a game about the shaping of a god, most akin to Ex Nihilo. It is not really about evolution as a biological process, or even about analagous selection-and-mutation processes; but rather the more general word, applicable to Pokemon and presidents, that means 'development' or 'change'.

As might be expected of such a rapidly-created work, there are lots of flaws: the prose is awkward and full of errors, suggesting very quick writing and possibly a second-language author.

Initially a normal person, you discover a Rift Engine which sends you through a series of disconnected, hastily-described science-fiction sequences, each resolving in some binary choice; these, in turn, grants you some kind of massively powerful ability. By the end you have become a godlike being.

Some of these choices are conceptually clear - philosophers vs. warriors, for instance - but the choices between different scenes are hugely opaque, like the one between String Theory or Mars. Most of the choices involve choosing which side of a conflict to support, suggesting that you're becoming the patron of causes or peoples.

These abilities don't interact, or become relevant at all within the story; rather, the end just assumes that you have become intensely powerful, and determines the result from your Morality score. This is a shame, because the powers you get, and how they might function together, are the most intriguing thing about Evolution. (But of course it would have been a vastly more challenging effort.)

Ultimately, this boils down to that rather boring variety of story - the game with Good and Evil choices, at the end of which you are either irredeemably selfish and destructive, or a pure and righteous hero. This is a model that's both boring and arbitrary; and it doesn't always fit well with the actual choices you're given. (Choosing peaceful herbivores over violent carnivores is presented as moral: but this feels like a category error of sorts, a case of anthropic thinking.) Often you get far too little information to make choices that seem as though they'll have massive impact. The impression you come away with is that gods are monstrous, charging headlong through the universe and screwing around with it on whims, without ever really understanding any of it.

Finally, the effort expended on the early game - discovering the Rift Engine, and a trad-IF-like sequence about repairing it - seems wasted, ill-suited to a CYOA format and largely irrelevant to the real focus of the game. Of course, these failings - both of surface polish and core design - are hardly blameworthy in a game jam piece.

Figaro is a very small example game, introducing a single point of theory and not really aspiring to any larger artistic goal. It lightly depicts an imagined scene from The Marriage of Figaro, taking an approach very similar to the examples of the built-in I7 manual.

The standard role granted to IF player is to identify with and serve the interests of the protagonist. You might have some influence over what those interests are, but the purpose is much the same.

A different approach, far less-explored, is that of drama manager: the goal there is not to reflect the agency of the protagonist, but to make decisions about the story, some of them extending well beyond the agency of any character. This is territory well-explored in RPGs, where improvisation is much easier; in computer games, the examples I'm aware of are all, like this, very brief.

Figaro presents three choices of three different kinds. One is a flashback choice that has major implications for the protagonist's character, rather like the character-creation choices in certain CRPGs or many ChoiceScript games; such choices often ignore strict agency (such as choosing your gender), and may even imply some changes to the world, but their proper locus is still the character. Another is a traditional agency-of-protagonist choice. And in between there's a choice that bears no relation to protagonist agency at all - which character is your wife carrying on with? Figaro demonstrates, albeit minimally, that all three kinds of choice can co-exist in a narrative game, and that having several kinds of choice can be more interesting than being restricted to one.

Nonetheless, my main reaction to the piece was that there are vast numbers of narrative computer games determined by direct protagonist agency, and a decent number with a strong element of retrospective or character-creation agency; but there just aren't very many dramatic-agency games of any significant size. Within its limited range, Figaro is well-elaborated, allowing for a very broad range of outcomes - but it's far broader than would be practical in a larger work, and doesn't really address the problem of how to design drama-manager choices in a longer piece.

Like many concept games, I came away with a feeling of mild dissatisfaction, because explaining the concept is the easy part. This game is roughly the same thing as a conversation about theory in the pub; it introduces an idea, but doesn't grapple with the (much larger) problems of design and implementation. Which is fine, as far as it goes; but it makes you want a whole lot more.

Retro-Nemesis is a game that was originally written as a hidden feature on the Get Lamp documentary. It didn't work out that way, so Robb released the game in a more conventional manner.

So this is a story about Robb's IF friends, and as such it's very much an in-community kind of story; you play Robb, who is lured by the diabolical Jason Scott with promises of more screen-time in Get Lamp into a road-trip to set fire to Adam Thornton's second house in Canada. Madcap road-trip hijinks ensue in rapid succession and everything goes to hell, with the barest of nods to interactivity.

If you're a fan of Robb's writing in general, this is an entertaining few minutes: it's all over-the-top Achewood-esque capers, viewed through a thick haze of sex, booze, retrogaming and sketchy Americana. If that's not your thing, you're probably going to feel a bit like a Mormon at a baccanal, sitting quietly in the corner and waiting for all the stuff you don't grok to be done with.

Filbert and the Broccoli Escape is an illustrated children's story adapted to a game; the lazy protagonist uses magic to try and get out of eating his vegetables, and finds himself on a (brief) miniature-kingdom adventure.

I dislike it as a game for straightforward structural reasons, and as a piece of kid-lit for more aesthetic, personal reasons. Let's start with the former.

Broccoli Escape uses Quest's hybrid parser/menus/hyperlink interface. I'm sceptical about this interface in general; but Broccoli Escape doesn't make confident use of it. Early on it flirts with a Quest-ish approach with option menus based on objects, but it quickly switches to something much more like a straight-up choice-based system. It doesn't use either smoothly; where it wants to offer straight-up CYOA choices, it awkwardly forces them through verb-noun commands that make no sense and confuse the transcript. Nouns and choices are capitalised in an ungrammatical manner.

Even imagining the work as a vanilla CYOA, it's pretty clear that Broccoli Escape was made by a static-fiction author with little or no game-writing experience. This is a pretty common species in CYOA generally, and the game falls into a familiar pattern: a single linear story wherein all the apparent choices are blocked off, except the one that leads to the One True Path. Worse, a lot of these blocked responses fall into that bad old IF pattern: offer an interesting option, then deny it as stupid or obviously unfeasible. (This specific thing is less justifiable in CYOA than it is in parser IF; it's ruder to refuse an explicitly offered option than an inferred one.)

It bears repeating: there is no point in adapting a work to a new medium unless the work grows in the process. Perhaps the idea was that in an online, gamified format it might reach more people; but gamification for its own sake is worse than useless, and an ebook might have been a more suitable (not to mention more widely-used) format.

Now, on to considering it as a book. (Important caveat: I'm speaking here as an adult who enjoys well-crafted children's books, not as a child or a parent.)

I have a number of nitpicky annoyances. I'm not a fan of the art style: it's all scratchy shading and blobby newspaper-funnies eyes, without the overflowing exuberance and fun-to-explore detail that I like best in children's illustration. And I always rather liked broccoli as a child, and deplore the unsubstantiated libel of its good name. (Aubergine is another matter.)

But running through these complaints, there's a general feeling of blandness. Broccoli Escape wants to be quirky and imaginative, I think, but more than that it wants to be safe. Filbert is a white kid from a middle-class two-parent family; the offending meal is gravy, mashed potatoes and America's Supreme Court-approved Designated Vegetable Which Is Unpleasant Yet Healthy. Filbert's problems are minor, his conflicts mild by the standards of kid-lit: he dislikes doing chores that every child dislikes, but never seriously clashes with his parents over this. There are touches of the comfortably-old-fashioned. (The author's parents probably read a printed newspaper. Parents of Filbert's generation overwhelmingly don't.) The central fantasy, of becoming very small, is a very standard one and isn't elaborated in any unexpected directions; and the whole fantasy plot is just one chase scene. Magic works (or doesn't work) as rhyming couplets. There's no problem with any of these as individual elements: together, they add up to something rather dull.

A missing-person case draws a detective into a cultic mystery. The genre is Chaosium's familiar twinning of 1920s noir-flavoured detectives and the Cthulhu mythos; indeed, the author acknowledges that the game is based on a particular Chaosium scenario, The Secret of Castronegro. Rather than the standard Lovecraft New England, the action here takes place in small-town New Mexico.

Small-town New Mexico in the 1920s seems like a strikingly non-standard setting with a lot of potential, but Castronegro Blues is not very focused on descriptive writing, or character writing, or on prose in general. Some people have complained about the profanity in this series; while I generally don't give a fuck about that kind of thing, I kind of sympathise. It's not that there's a lot of swearing, really; it's that the rest of the writing is so utilitarian, so placeholderish, that the crassness really sticks out. Similarly, the protagonist is kind of an asshole: he's uninterested in and contemptuous of his surroundings in general and the people in particular (there are lots of NPCs, almost all of whom are one-line stereotypes). Now, asshole protagonists can absolutely work, if they're interestingly complex or have redeeming qualities or are just entertaining; but the nameless detective here is a blank, except for when he occasionally, suddenly comes up with something curt and mean-spirited - the entire description of a desk sergeant is "He looks like an idiot." So while I'm generally a big fan of asshole PCs, it was uncomfortable to spend time around this protagonist, and I didn't feel as if this discomfort was in service of anything.

(It's possible that he wasn't even intended as an asshole. Maybe he was intended as a gruff, laconic Marlowe type, and the author didn't manage to capture all the elements thereof. But what's left is pretty much just asshole.)

Player interaction is not a focus, either. The process of detective work is conspicuously just a framing device, and this is true of the game's approach to interaction generally. For much of the time, solutions spring into your hands in a great hurry; on the other hand, UNDO is forbidden, the game puts you in situations where death is hard to avoid, and there are no warnings about keeping saves.

What this game is focused on is story, the unfolding of a Lovecraftian horror mystery. But in its enthusiasm, it hurries: police and witnesses pour detailed accounts on you, helpful library books spring into your hand. Both mystery and horror are genres that are reliant on pacing, on the slow build, on taking care with the delivery of crucial information to the player, on the power of the unknown. It's more difficult to pull off when, as Castronegro does, it includes a lot of stock subgenre elements (creepy Indian tribes, old local families with human-breeding plans, secret cults); the result is that this game tips its hand way too early, before there's time to develop a sense of threat.

There's enthusiasm here, but that enthusiasm has yet to get translated into a labour of love. This section, from early on, tells you pretty much what you need to know:

There is a knock at the door but the knocker doesn't wait for an answer. The door opens and a beautiful woman walks in. "Thank you for seeing me, detective." You invite her to sit and she does.

That's skipping over a lot of annoying little steps - opening the door, greeting the woman, offering her a chair - that the author thinks are boring. And a lot of them are boring! But the baby gets thrown out with the bathwater: the character herself motivates the protagonist and thus the whole story, so she very much needs to not be boring. What we get really amounts to just a statement of her narrative role - 'you know how this kind of story goes, fill in the details yourself.' That's really boring.

(On the plus side, the titles are getting a lot better. The Surprising Case of Brian Timmons is pretty feeble as titles go. Castronegro Blues could be a Tom Waits song. Ill Wind is just great.)

Witch's Girl a browser CYOA framed as a children's story, and combines gently snarky humour, simple but charming illustration and a turn-to-page-N conceit to suggest a picture-book. Two small girls, Oblivia and her best friend Esme, encounter a witch who directs them on a time-travel adventure to save their (slightly hazily-defined) fantasy world.

The emphasis here is rather more on the adventure part than the saving-the-world part; in a few important respects the story resembles a game of make-believe more than a work of authored fiction. The long-arc plot is conspicuously an excuse to run around and have adventures, and is often temporarily forgotten in favour of more immediate distractions. There's a pleasantly childlike focus on the cool stuff that's happening now; but there's also a certain snarkiness about the whole enterprise that gives it a more disaffected grown-up feel at times.

Structurally, a degree of puzzle difficulty is added by a time-travel mechanic; in order to find items to solve gateway puzzles and advance to the next stage of the central plot, you need to use the witch's cauldron to travel back to earlier points of the plot and choose alternate paths. This means that you have to consume most of the game's content in order to win. Towards the end this devolves into lawnmowering through all the options, but it's less annoying than it could be; finding the last few items took about the right amount of time, as far as I was concerned.

My feeling was that it could have benefited from a little more use of state; it's a game about collecting items that doesn't have an inventory, which can make it a little hard to keep track of things. And it'd have been nice if there had been slightly more illustrations, more evenly distributed through the text. But overall, a pleasant experience.