I was apparently one of the few people that genuinely realised, from the item’s description,
that I had to fold the paper into something. Mentioning the crease seemed too intentional to me for it to be otherwise.
Not that it did any good, because the implementation of that action failed spectacularly, and after trying umpteen different ways of wording it I gave up on it, thinking I’d had the puzzle figured out wrong.
Turns out I had pretty much the same experience with the game itself. It’s implementation fails spectuacularly, and after trying umpteen different ways to progress I gave up on it, thinking I’d had the game figured out wrong. After reading the walkthrough, I was proved right. I’d originally thought the game was fair and fun.
Do I sound bitter? That’s how I came away from the game, so I see no reason to hold back.
I also assumed there was something going on a bit like that, but I would never, ever have guessed the actual phrasing, or what you would want to use such a thing for.
I am notoriously slow at these things. But while, based on the description, I might (but probably wouldn’t!) have tried to use it to carry powder or something like that, the actual task and phrasing I would never in a million years have got.
[spoiler]But I thought it was pretty unfair from the outset. I can just about see how I might be able to kick the stone even though I can’t touch the stone. I imagine chains that hold my hands up high. But how is it that having been unable to get my hands down on the ground to touch the stone, I can get them down to pick up the wand?
In the airplane case, I don’t think there’s anything at all to tell us that Gwydion – who is happy for us to wander around through a secret passage we have previously been unaware of, but not willing to let us in his bedroom though he will have us in his dungeon – is going to be so distracted by a paper dart.[/spoiler]
[spoiler]I wasn’t sure what to make of this, mostly because I wasn’t sure what it was trying to do.
It’s an old-fashioned paper-based CYOA work, of the sort that gives you explicit choices of the “To greet Harold, turn to 34, to avoid him, turn to 87” type. The gimmick is that as the game progresses you get given secret codes – keys that you can add to the explicit choice, in order to open up an alternative route. So, to extend the example given above, you might be told that if you want to hit someone instead of talking to them, you add 10 to the choice for conversing, in which case you would be able to turn to 44 to hit Harold. TIT uses this gimmick in the context of a malfunctioning time machine. Gradually you acquire more and more codes, and in the end you may actually be told just to roam through the book finding small hidden adventures within. Thus the determinate structure of a CYOA is broken down.
I don’t know enough about CYOA to know if this is a new idea, though I can’t imagine it is. Anyway, TIT deploys it in considerable depth, and it seems quite clever. It must have taken careful checking to make sure all the numbers ended up right. The result is a reasonably interesting story, told in plain but agreeable prose. It’s not got much depth of character, but it’s clearly not trying to have.
But I don’t see the point of the gimmick. Effectively, it turns the player into his/her own virtual machine. You have to keep track of all sorts of potentially-addable numbers, and trail around through the PDF (easier, perhaps, if you print it out, which I didn’t). Perhaps if one were nostalgic for CYOA books this would be a pleasure. But I’m not, and I found it rather a chore. I’d far rather let a computer do this sort of dirty work for me.
Except, of course, a computer wouldn’t produce quite the same effect. The key effect here is that the “codes” give you access to hidden choices, which sit alongside the explicit choices, whereas hypertext games (the closest electronic equivalent of CYOA) can only deal in explicit choices, and parsers generally make all choices implicit, so the precise form of interactivity offered here is unique.
Perhaps this is the point the author wanted to make. If so, I cannot feel it was worth making. However interesting in theory, the actual experience is more about scribbled notes and quick mental arithmetic than about new and interesting forms of interaction.
I came away with the impression that this was trying to be clever. Entertaining too, I’m sure (it’s not a troll piece, and it’s not just showing off) – but first and foremost clever. I found it reasonable clever, and reasonably entertaining, but in the end not sufficiently either to make it a real favourite.[/spoiler]
[spoiler]Time-travel and CYOA have been close relatives from the beginning. The very first gamebook by Choose Your Own Adventure founder Edward Packard, Sugarcane Island (from whence Twine’s Sugarcane format), features a small but explicit time-loop; time-travel was perhaps the single biggest theme of the CYOA series. Puzzles centrally about the mechanics of time-travel have definitely been done, with the most pure, refined examples being done by Jason Shiga (Meanwhile is the major one). And the work sometimes credited with having the idea of CYOA, Borges’ The Garden of Forking Paths, is centrally interested in how the structure messes with the sequence of time and causality.
The mechanic ‘perform a simple mathematical exercise to discover which page to turn to, thus preventing you from cheating unless you have the required knowledge’ was used in a number of books - I don’t know the first, but I think several Fighting Fantasy books used it. And the trick where certain nodes can only be reached by ‘cheating’, and you’re meant to do this, is a major feature of Inside UFO 54-40.
So, Trapped in Time is very much a homage. I don’t know of any particular game that uses its specific structural mechanic - heavy use of mathematical operations to elaborate on a loop - but conceptually it’s a small embellishment on very well-established forms. So I think your appraisal of it being clever is about right: it’s a bit like a functional toy car made out of twisted paper-clips. Nobody’s going to be very surprised by the idea, and it might not, in the end, be the best toy car in the world: but it’d take considerable craft and care to get it right, and you can see how it’d be a fun project.[/spoiler]
Part of my childhood. [emote]:)[/emote] And since I read TRANSLATED versions, and in some books you had to add up the number of the letters in the word that comprised the item you were trying to use, or some such… well, those translators had to work extra hard to ensure the math worked.
Those were the days. Translations were actually good back then. Now it’s all crap, everyone thinks they know English because it’s so widespread.
[spoiler]The “machine of death” (apparently an internet celebrity, which has somehow passed me by) is a sort of oracle, which cryptically predicts one’s fate. The conceit of this game is that one receives one’s fate, and then has the opportunity to explore a scenario in which (perhaps) that fate comes to pass. As it happens, I managed to dodge the bullet on each of the two occasions I played, even with some replaying, though I suppose I should be glad.
I quite enjoyed the two mini-stories I played through. The writing is competent, and I thought the game did a good job of playing on the various expectations created by the allotted fate. But I didn’t feel it was completely comfortable in its skin. Most notably, there seemed to be some scenes where it felt that this was trying to recreate something of the experience of the parser, but in hypertext. So a page will sometimes consist of a basic description of a room, with a large set of links to things that one might examine or exits one might take. These sections didn’t seem to me to play to the real strengths of the hypertext interface.
But that’s really a quibble. This isn’t a piece that’s trying to do anything cutting edge, technically or artistically. But that doesn’t prevent it from doing what it does in an entertaining way.[/spoiler]
[spoiler]Two stoned people go to a grocery store. You get to choose a drink, some chips, some cookies, and some ice cream for them. You get to hear their comments and sometimes their reminiscences about what you have chosen. Then, as a check out person, you get to listen to their comments as you ring the merchandise up.
There’s not much of a story here, not even a backstory. And for the few minutes I spent with it, I found this neither unpleasant nor thrilling. It’s competently put together. It’s not terribly written. But I didn’t get any sense that it wanted to do anything much, and I slightly resented it for its lack of ambition.[/spoiler]
I’m done. There are a few games I haven’t reviewed. I don’t plan to, because I don’t think I have anything worth saying about them.
I don’t have any earth-shattering final thoughts either. I had a good time with a lot of the games. I thought three stood out: Coloratura, Solarium, and Their angelical understanding, because these seemed to me to manage to combine technical facility with artistic ambition, so that they promised a lot and fully achieved what they promised. But there were lots of other games with many things to admire in them, and I’d like to thank the authors for the time and effort they take to make things for us to play.
The Comp seemed notable for the number of games which sought to explore or discuss very serious topics, many with considerable success.
I’m aware that the entries this year provoked quite a lot of discussion about the possible decline of the parser and/or the rise of hypertext, especially Twine. I thought there were strong (and weak) entries from both “schools”. I didn’t feel any difficulty judging them against each other (or, at least, no more difficulty than I have judging Tex Bonaventura against, say, Ollie Ollie Oxen Free). There are things the parser can do easily that are difficult in hypertext, and vice versa; and there’s a marked difference in both between the most accomplished authors and others.
I have a slight sense that Twine users seem to have more energy, more urge or licence to experiment, and perhaps less “baggage” than those working in, say, Inform have. The parser-based games seemed, as a group “safer” than the hypertext ones (artistically, not technically – technically I thought the parser games this year were particularly accomplished). Perhaps this is my imagination, or perhaps it’s a function of the relative maturity of parser-based IF and the fairly extensive set of expectations that has been built up.
I hope that the criticisms I have made in these reviews from time to time have not been too bruising: I truly appreciate the work that goes in to these games, and I know how easy it is to snipe from the sidelines. All of the works I have reviewed are labours of love. It shows. Thank you all.
As I did last year, since I don’t have a blog, I’m going to post short-ish reviews here. I’ll probably spoiler tag them more or less in their entirety. I will largely play the games in a random order.
A note to authors: (1) I’m sorry if I am critical, knowing how much effort you put into your work, but a review is not worth much if it’s not honest. Criticism of the work is not intended to be criticism of you as a person. (2) I don’t think it’s possible to be entirely objective in reviews – in a competition as broad as this it’s inevitable that some things will simply not be a particular reviewer’s cup of tea. (3) Sometimes, in the pressure to play so much, I manage to miss or misunderstand something badly. If that happens, I’m sorry. (4) Seriously, whatever I say or think about your game, thank you for making them.
Summary: hugely ambitious, but fundamentally flawed
[spoiler]This appears to be a huge game. I can’t imagine that anyone could complete it in the time allowed: even following the walkthrough and skim reading I doubt it would be possible.
I’m afraid, however, that I didn’t finish; indeed, in my own play, I hardly even started.
The immediate reason for my loss of patience was as follows. I was asleep in the trench, when I was woken by rats. Fair enough: trench rats were a real problem. But these were killer rats. Not fair enough: rats ate corpses, yes; but they were not killers. Anyway, I am told I must fight these beasts. I decide to shoot them. Certainly against regulations: shooting rats was regarded as a waste of ammunition. But, instead of an admonishment I got … game over. Apparently, instead of shooting the rats, a “shotgun pellet” had killed a colleague. But what was I supposed to be doing with a shotgun? I could have tolerated even this; but I was then told I could not undo.
If a game implements sudden and irrational endings without warning, and then (selectively – i.e. deliberately for this very purpose) disables undo, I know it is a game that really doesn’t care whether I am having a good time. So I quit.
I then went back to the walkthrough, to see if I could at least understand the experience the author wanted me to have. I hardly scratched the surface. The full walkthrough features around 700 commands. There is no way that anyone is going to get, spontaneously through that in 2 hours, or approaching it. Even simply to type and read the text would not be possible. I spent about an hour or so working through a chunk of it.
There are, to be frank, huge problems here. There are problems with the writing, for which the motto might be “tell don’t show”. There are problems with the mechanics, with non-standard verbs, non-standard use of standard verbs (“drop pants” is required; “remove pants” fails). There are finicky sequences in the pursuit of mechanical realism (opening things, closing things, picking tools up), and then entirely unwarranted departures from them. So, for instance, when you want to eat some beef, the game insists you tell it to pick up a knife and fork. But when you are assailed by gas in no-man’s-land, it just puts your gasmask on and off for you. There is inconsistent use of “talk” and “ask/tell”. There are problems with random death (from which, in fairness, you can recover: but until you realise they are just random, you are liable to assume you did something wrong, when you may have made exactly the right move, and just been randomly killed).
These are basic craft things. The fundamental problem here seems to me to be that the author has in mind a story he wanted to tell, but assumes the reader will follow seamlessly along. Given the ambitious depth and breadth of the game, it’s not surprising that there should be problems; but they are stark and frequent. Much, much, more testing was needed.
And, I’m afraid, I’m not really sure about the story. I’m not against historical fiction, and although a pretty over-travelled road, WW1 is not without interest. But do we need to be required to drop our pants and shit in order to persuade us that latrines were unpleasant? And, apropos that sequence, what are we to make of this:
Well, actually, if it’s a choice between what are here coyly called “my physical requirements” and my life, I would choose my life. As between embarrassing defecation and death, give me embarrassing defecation every time. But the game, having presented a false choice, proceeds to insist that we make it absurdly.
And for all this attempt at realism, the historical accuracy is only skin deep. Did killer rats claim many victims? Nope. Did a platoon have three privates and a corporal? Nope: the smallest unit which would have been under the control of an NCO was a section, of 12 men. Would the British army have allowed a US citizen to join up and serve in US uniform? I doubt it (nor, I imagine, would the US army have been over the moon about it). Would a “posh” person with a good education destined for the diplomatic corps end up as a private? Very unlikely, and would have stood out. Does a corporal salute a sergeant major (with the “usual salute” no less)? Nope. Would a person have used the word “Robot” in 1916? Nope: its first recorded use is 1922.
Now you may say “These things don’t matter.” But, if you are aiming to write historical fiction with, it has to be said, the distinct air of the pedagogue, these things do matter; and they can all, quite easily, be checked. This is not presented as a game which simply happens to have a WW1 background: it keeps emphasising and explaining it. In this respect, the author’s approach seems to be exactly the reverse of what the best historical fiction does: instead of a deep and very accurate understanding of the period worn lightly, it shows a shallow and inaccurate one, worn on its sleeve.
I wish I liked this more. A vast amount of work must have gone into it: there is masses of text, and extensive plot (one can guess some of it if one reads the walkthrough: but 700 commands to get to the end — that’s a monster). The trouble is, it was not the right work. To write a decent story whose “ideal” walkthrough extends over so many commands would be the work of months, I should imagine; testing would be nightmarish. Even more so when there are numerous NPCs, and where the action is intended to move sequentially and fast. Frankly, I think the game has bitten off more than it could chew, and the work that needed doing — research, editing, re-writing, writing all that text that hardly anyone ever sees to respond to plausible but incorrect commands, and testing, testing, testing — that work, unfortunately did not get done.[/spoiler]
[spoiler]Is there any cluing about how to get into the office? I found this puzzle (and others like it, that require you to REVIEW random photographs without motivation) frustrating.
The other thing I found a bit odd was the narrator’s nonchalance at all of the supernatural things happening around him. The first time you see something creepy in a camera flash, I was surprised and a bit spooked; but then it kept happening over and over again and I found it merely annoying. Those snippets would have been more effective if they only happened, say, 10% as often. Also, at one point the narrator gets ambushed by a box full of disembodied hands, and he doesn’t even flinch; even ‘x’-ing the remaining hand afterwards, you get a description suggesting an animated, disembodied hand is the most natural thing in the world to find inside an abandoned mansion. Huh?
Finally, the game desperately needs a binder of some kind that automatically collates the various notes you pick up.[/spoiler]
[spoiler]I don’t think it’s specifically (or adequately) clued, and by the time I got there I had looked at the walkthrough – but I had already picked up the idea of taking loads of pictures and waiting for the “tingling” to tell me which to review; but I agree the cluing could be stronger (as I think I pointed out). Indeed, if one even had a clear indication that it was to the study that one should be especially attending, one would at least know where to focus.
The nonchalance I think is justifiable. Consider the PC. If the PC is the sort of person who takes the haunting to heart, then they are going to leave pretty quickly – so to me it makes sense, if you are seeing lots of these things, that you are pretty phlegmatic; this is not inconsistent with finding some aspects (e.g. the nursery) more disturbing, though in general my own experience was not of “high tension” but of an engaging puzzle environment.[/spoiler]
Summary: A carefully constructed Twine-based puzzle: click here to exit
[spoiler]The contortionist is a puzzle-based escape game, made in Twine, but with a very open interface in which you can select from a range of actions to perform (such as moving, taking things, wearing things, opening things and so forth) and in which you have and acquire an inventory of objects to solve puzzles. I played it to (several) deaths, and finally, with the help of the walkthrough, to a satisfactory ending.
It’s quite nice. For the most part the text was well-written and error free, the puzzles were reasonably coherent and (my resort to the walkthrough notwithstanding) fairly hinted, and the whole thing is set in a dystopian future which, while it’s not going to win any prizes for originality or perceptiveness, kept the game together and gave it enough of a narrative edge to make it satisfying for me. I was especially impressed that the two central puzzles were not simply of the “find and object, use it” sort (though they appropriately involved that), but needed more thinking.
And yet I was quite disengaged when I played it. And I think the reason is the interface.
This sort of thing is, I think, a natural for the parser: it’s the kind of thing the parser does well. A link-clicking interface is, for me, much less satisfactory. The thing about the parser is that, once you have an idea of what it will let you do, and provided the game is well programmed, it keeps out of your hair. This virtue is also its vice, because until you know what it will (usually) let you do, a parser can be sphinx-like, so for a novice the availability of links, which function like a sort of on-screen menu of possible actions, seems a boon. And even for an experienced player, the moment when you cannot quite manage to communicate, when you have to guess a verb or a construction, re-ignites that novice frustration.
Links avoid this, but at the price of insisting that the “mechanics” of the game are always on show. And instead of a natural flow of text and command, you have screens which are constantly clearing and refreshing, and I lost the sense of flow. Moreover, if one actually looks at the content of each screen, a good part of it consists of things which are relevant to the interface (links to click), not to the story. It becomes a rather laborious process, especially frustrating during the frequent replays that are needed to make progress. (This is very much a game where you learn from your mistakes.) In a parser-based game, to move I am likely to type a single letter; here I must click once to say I want to move, then click again to decide where I must move, and it begins to get in my way, to get in the way of what I think they call “immersion”.
I don’t find this at all with a more conventional Twine set-up, because there the links are, as it were, integral to the text: in considering my “next move” I am considering the text itself. It is the separation of text and “interface” that is presented here — my attention effectively shifting from text to input interface in a way that I found distancing.
And of course the other thing this inevitably produces is a risk of going to mow. I thought the author largely avoided that, particularly by arranging things so that often a particular sequence (talking to someone, or examining something) was necessary in order to open up the next stage. Still, with all the options on display at every stage, the sense of exploration is bound to be reduced.
I wouldn’t want it to be thought that I didn’t enjoy this, because I did. I liked the setting and the puzzles, and I was happy that it was generally a game that seemed happy for me to be basically a nice person (not a saint, but a person who cared appropriately for others and treated them decently). The only element of the story I was unhappy about was the ending, which I thought was too much of a text dump, and somewhat saccharine. It’s a technically proficient demonstration of how Twine can be used to produce a puzzle-based work; but in the end, for reasons that are perhaps purely personal, I found myself wishing that the interface would get out of the way more.[/spoiler]
Summary: A programming exercise which does not really hold together as a complete game
[spoiler]This is a rather depressing thing to read in ABOUT text:
Explaining “cliche-ridden tropes” does not make them OK. If the collection of puzzles really had been “held together” by a story, things might have been better; but the puzzles are only held together by a story in the sense that the random assembly of elements at a potluck supper are “held together” by a tablecloth. There’s no really coherent theme, and the story (which is pretty sparse, and itself a loose amalgam of generic fantasy elements) does not truly unite the puzzles, it simply happens as they are solved.
Nor can The Tower stand on the strength of the puzzles alone: there are a few things to find by searching, a few things to find out by reading, some simple object manipulation. The playing experience is not very smooth. To take one obvious example, if you are going to require me to arrange paintings on hooks, it would seem desirable that if I try to hang painting A on a hook already occupied by painting B, the game should silently remove painting B first, rather than forcing me to do the whole thing manually. So even as programming exercises these are not really fully polished. (The hint system, on the other hand, seemed quite solid and helpful.)
The writing does not save the piece either. It’s sometimes awkward (I don’t think English is the author’s first language) and never especially vivid or engaging. The Tower is indeed not much more than a container for a collection of cliches.
At the end of the piece, in what is presumably a reflection on the work itself, the PC finds himself drawn to a painting that is, in terms of subject-matter and execution, merely mediocre. It’s a strange combination of realism about the work’s quality, and optimism about the effect on its audience. But for me, I fear, it was only half right. I’m glad the author wrote some practice pieces. I hope he’s learned useful things while doing it, and there are certainly some nice touches here (like the hint system). But overall this cannot be regarded as successful.[/spoiler]
Summary: A mature and subtle exploration of mental illness
[spoiler]If you haven’t read any reviews, your first experience of this will lead to a surprise: what presents itself as two language versions of the same story (in Scots and English) is in fact two different, but loosely corresponding stories: a fantasy question (in English) and the story of a person suffering from mental illness (in Scots).
What does this give you? A structure such as this is almost set up to invite you to “compare and contrast”: the English text is smooth, easy to follow, heroic (but, when all is said and done, unrealistic, vacuously heroic, faintly ridiculous: the emptiness is deliberate); the Scots text is rough, difficult to follow, mundane. But, at least for me (as an English person), the Scots text is also, simply by virtue of its dialect, exotic. One story represents escapism; the other a form of imprisonment. And yet, at the same time, both depict different modes of retreat from reality, and different forms of loss of rootedness in making that retreat. On another layer, one text can stand in a metaphorical relationship to another: a maze as a metaphor for a panic attack (a clever and successful technique, I thought); a question for some power-conferring object as metaphor for the search for a way to get through the day intact.
All of this is true, and interesting. But in the end I wasn’t completely won over, mostly (I think) because I didn’t find either of the texts sufficiently interesting in itself, and although the interplay between them added a good deal, it didn’t amount to enough to hold my gaze. This is not so much a failure of design, as a feature of it, for at least as I read the work one of its points was that mental illness is experienced as a grinding distraction, a dead weight. But it meant that while I was thoroughly impressed by the intricate way this had been put together, but I didn’t really find myself caught up in either story. So the experience for me was too much intellectual — too much a matter of admiring the way in which the two stories had been connected and woven together — and insufficiently visceral.
The walkthrough and author’s notes here are very informative. The latter quote a post (a “bangin post critiquin the systematisation o mental health in gemms like Depression Quest”) which asks “What would it look like to create a game about experiences of madness that does not use medicalized discourse?” As Sam Kabo Ashwell notes in his brilliant review, this is a loaded question: “most Twine games that deal with mental illness or suffering take a strongly interior, experiential approach”. If anything, as it seems to me, the target here is not just the medicalisation of mental illness in games, but its romanticisation. Raik provides a dramatisation that neither glamourizes nor prescribes. The experience it depicts offers no epiphanies, no prophetic insight, but a demoralising struggle with the mundane. It offer no window onto any sort of medical world; but nor is its madness freighted with the voice of the oracle. The result is balanced, sensible. If that assessment incorporates a hint of the negative (sensible as in dull), it is also intended to be strongly positive: this is a thoroughly intelligent piece, acutely observed.[/spoiler]
Summary: a parser-based story which is promising but rough around the edges
[spoiler]Jesse Stavro’s Doorway is a parser-based story. I played it to an ending, though I made fairly heavy use of the walkthrough in the endgame (and I doubt that I would have completed in time without it).
Although there are puzzles, they are not really the focus of this game; none of them requires much thought or experiment, and they largely serve as pacing devices. They are generally well-integrated into the story elements of the game (i.e. they feel, within reason “natural”, at least so long as you don’t think too hard about whether a pit-bull could be permanently subdued by a cardboard box, for instance). They did, however, require a certain level of mind-reading on some occasions. There were also too many areas which resembled, if they were not actually, mazes for my taste.
Still, the puzzles are not really the point here. The point is the story, and the key question is how the game delivers that. I thought the structure was good: the device of travelling in time (on this occasion through “portals”) is not exactly cutting edge, but it’s effective. The idea that you are searching for a particular person (Jesse Stavro) provides the occasion for this travelling, though I was not quite clear why I was so keen to find Jesse. Less successful, I thought, was the backstory, which I found slightly confused, and which was somewhat forced on one through an insistence on reading notebooks; but it didn’t really get in the way.
Another structural element that I thought worked effectively was the use of “chapters”, where solving a puzzle would open up not a new area of the map, but essentially a completely new location to which one moved. One problem that parser IF can have is creating a sense of dynamic changes of space and time, and the chapter format achieved that.
I was impressed by the attempt to include a large number of NPCs. There was almost no point where one was left wandering alone: the player constantly encountered new people, who moved around and to some extent at least participated in the action.
Finally, the setting seemed reasonably fresh. A trip through California in 1977 to a Grateful Dead concert is, as far as I am aware, a novel premise, and that is welcome. Some effort had been made to capture the sense of time and place.
However, for all those positives, I don’t ultimately think it’s a successful game, for three reasons.
First, although the premise of the story is sound enough, the story itself isn’t really very good. Put it this way: if one took the result of a playthrough and turned it into a short story, it would be dull. The PC and NPCs are too colourless; the need to find Jesse seems insufficiently pressing; the villains insufficiently villainous. It lacks what I think is called in some circles a “story engine”:
There’s plenty happening, but there’s no overarching reason for it, and the backstory, which might provide that reason, really doesn’t. And some of it’s really rather unthought-through: a secret door that appears whenever anyone dries their hands is not going to stay very secret for very long, one might think.
Second, there are implementation problems. It’s not so much that the writing is bad (it’s not: it’s flat, but it’s fine), but the implementation is just too shallow and sometimes buggy. There are many too many empty locations, with no purpose except the faithful reproduction of geography, for instance. It feels like these were a chore to write: “Damn! A house must have a bathroom. And a bathroom must have a toilet, and a mirror, and a sink. And now I’ve got to describe them, so I’ll just put down something that is vaguely atmospheric but won’t encourage further exploration.” And that, presumably, is how one ends up being told that the toilet “reminds you of a glazed donut” (sadly, EAT TOILET produces the library-standard “That’s plainly inedible.”).
An author who insists on including unnecessary locations is making a rod for his own back, and the effort that goes into producing lacklustre descriptions of irrelevant places would be better spent polishing the locations which really matter. Instead, they end up under-implemented.
The same is true of NPCs. A good NPC is hard to find, harder to make. For instance, to provide any sort of conversational depth is notoriously time-consuming. The NPCs here are pretty scrappy. They have few things they will talk about, and many apparently plausible topics yield nothing better than “There is no reply.” They are largely inert; they rarely take any sort of conversational (or other) initiative; and they will mostly simply stand around letting you do things if you choose to do so. I can see that it would have been a huge task to implement seven or eight convincing and active NPCs, but the game needed more than this. As with the locations, we seem to be being offered quantity over quality.
Finally, there are simply too many rough edges. For instance, you spend a lot of time with a man called Riley. But even after you know his name, his every movement is announced by saying “tripping guy comes in from the east” — without even proper capitalisation. Conversely, on other occasions, a character’s name is announced before you should know it. Sometimes the screen is almost full of empty default announcements about what various characters are up to. On another occasion, having been told that you can see “a woman packing a large trunk” EXAMINE WOMAN produces the odd response “You’d have to go into the northeast room to do that,” and EXAMINE TRUNK produces “You can’t see any such thing.” The woman in question is then, when you enter the room, identified as Lydia (before you have spoken to her) but described in conversation as “a strange woman”, so that — weirdly — you are counselled to “ask a strange woman about a strange woman” (that should be “ask Lydia about herself”).
So, in the end, I don’t think this works. It needs to be more tightly plotted, more tightly written, it needs to have more of a sense of obsessive perfectionism to it. It needs both to be cut down (fewer places, fewer NPCs) and worked up (more detail, more conversation topics, more attention to the precise way the text is produced). As it is, it feels like an early draft rather than a fully finished product, albeit a promising draft.[/spoiler]