Toby’s Nose (Chandler Groover) is a main-festival entry in Spring Thing. You play Toby, the dog of Sherlock Holmes, who helps solve crimes through his powerful sense of smell. (I’ve thought for years that a scent-detective/forensic-investigation IF, influenced by Patrick Süskind’s Perfume, would be a good idea; Porpentine’s Nostrils of Flesh and Clay begins in that direction but quickly becomes something quite different.) The game opens on the final scene of the mystery, as the suspects gather in the drawing-room – thus offering Toby an opportunity to give them all a good sniff.
The game has an acknowledged influence from Lime Ergot, a surreal, telescoping game by Caleb Wilson. Toby’s sense of smell doesn’t just give him detailed information – which is an obvious approach – but seems to transport him into the places whence the scents originated. Picking out individual elements within such scenes can zoom in on them, or even make the mind’s nose travel from place to place. This sneaks around the usual scope mechanics of parser IF, since Toby can smell everything at once; and this gives the setting a lurid and dreamlike quality, a sense of drifting through ephemeral space rather than inhabiting a solid world.
Sometimes this means that the game is unrealistically specific about some details – how would you know that that chalk odour was used to draw too-perfect pentagrams? – while being overly coy, for the sake of the mystery, about certain information that a dog would immediately pick up on (mostly to do with who exactly performed an act inferred from smell.)
The sense of smell has strong associations of disgust, guilt, sickness and contamination, social inclusion and exclusion, class. (“Must be a king.” “Why?” “He hasn’t got shit all over him.”) Perfume starts low and climbs upwards: the monstrous Grenouille is born into squalor, filth and desperate poverty, and uses his preternatural talents to climb both to higher social status and to ever more rarefied aromas, and ultimately to a very Dionysian kind of apotheosis. Toby’s Nose goes in the other direction, beginning in the upper-class security (and comfortable genre expectations) of a great house’s drawing-room; and proceeding from there to Gothic churchyards, filth-swamped slums, opium-dens. Toby is obviously a Tobermory, an animal uncoverer of hidden sins and hypocrisy, an uncoverer of socially inconvenient truths.
One of the big impressions that I got from this game is of teeming clutter; it is common to open up a detail and discover a half-dozen or so more details crowded into it, often in lists. The Victorian UK was a world unprecedentedly glutted with consumer goods; one of the big ways that the UK feels different from the US is how bloody much of the stuff is still in circulation. When I first learned about the Victorians at school, the big difference about how that was taught (off-putting to me at the time) was its heavier focus on social history, generally through the lens of everyday artefacts. (Connie Willis gives a similar impression of interior decor in To Say Nothing of the Dog, of homes crowded to the brim with bric-a-brac and ornate furniture).
And this means that your challenge is a panopticon problem: it’s not that there are too few clues, it’s that there are too many. Everybody is up to something, for sure, but which of their secret crimes were related to the murder? The big problems of mystery IF are all to do with how the player figures things out: do they have enough information? can they pull it all together, and how do they indicate to the game what they know? can the plot advance without them knowing what happened (bad) or does it get stuck indefinitely if they miss a crucial connection (worse)?
In my playthrough, I was able to gather up a great deal of suggestive detail, and to make some very strong guesses, but not quite draw it all together into a conclusion. Probably I had missed some important detail – but if so, I didn’t know what to do except keep combing back over existing content. Or else I hadn’t mentally connected some of the details that I already had. The only way to show what you know is by barking at a suspect, but on a meta-game level, I wasn’t quite sure about the assumptions at work here. If I misidentify a subject, does that mean that I got it wrong (since Holmes is infallible) or that I just don’t have enough evidence yet (in which case, how does Holmes know that I do?) I felt that, perhaps, it would have been useful to produce an occasional summary of how bits of information connected to one another – “these two letters used the same ink, which might suggest the same author,” that kind of thing – if only to suggest how strongly we should read certain linkages.
Regardless, this is an appealing, evocative game and I enjoyed it considerably, for all that I couldn’t solve the damn thing.
One small disambiguation issue, in the singular/plural category:
>smell prostitute
Which do you mean, the prostitute or the whores?
These Heterogenous Tasks 
Barking at the wrong suspect causes you to lose the game. It’s the “Holmes is infallible” route. The solution is rather obvious once you’ve collected the relevant information. But, I ended up basically brute forcing it because I wasn’t sure whether I’d solved it or not; it would have been helpful to know in advance that I’d know when I had the solution. (Though the scenes that play out when you bark at the wrong suspect are illuminating and/or amusing too.) I thought it might have been useful to ‘gate’ the solution — with the parser actively blocking your bark unless you’d uncovered the evidence. But I gather that the sheer number of relevant details and the time and technical constraints on the story’s construction precluded this.
Many of the not-related-to-the-murder clues are in service of establishing alibis for non-murderers, which is a little vague since Toby’s time sense is weak and it’s difficult to trust it to the precision necessary for alibis. It works ok, though, because the ultimate solution to the murder is not through a process of elimination.
I can’t believe that error with the prostitute is still there! That means I sent Aaron the wrong file, because it doesn’t happen with the Play Online version I’m hosting myself. Hopefully I can get the update in place fast enough so that other people download the right one!
Regarding the ink, it’s stored as a clue in your inventory to suggest that it’s important. Only two written documents appear in the game, and both use the same ink. Adding more text to confirm that they have the same author would, I think, be too obvious. Toby remembers nothing inconsequential.
If you misidentify a subject, that means you’ve got it wrong. After all, you’re not presenting evidence for Holmes to draw his conclusion from. This is why you could theoretically bark without gathering evidence at all. Holmes has already put the puzzle together himself. Your bark is just the final test, as he explains. And taking that leap by barking at who you believe to be guilty is really essential. Telling a player that they’ve got it right in advance, before they’ve barked, would deflate the game.
However, if anyone can’t solve the mystery and does want to know how the clues fit together, feel free to contact me and I’ll give you the answer! I’m CMG on intfiction.org.
Thank you for the review!
Regarding the ink, it’s stored as a clue in your inventory to suggest that it’s important. Only two written documents appear in the game, and both use the same ink. Adding more text to confirm that they have the same author would, I think, be too obvious.
To me it felt pretty circumstantial, because two people could easily have used the same kind of ink (even, if they lived in the same house, the same inkwell!) And maybe I’m wrong about this, but my vague impression was that india-ink was at least somewhat commonplace in the period; if it had been a rare or specialist ink, I might have felt more confident. (And this is just an easy example; there were other inferences that I was a good deal less confident about.)
What I mean is, there are two ways of thinking here – the puzzle-y mystery kind of thinking, where a highlighted clue necessarily leads to important information, or the real-world-logic mode, where it doesn’t. And it’s within the rules of mysteries for the former kind of assumption to be undercut – ‘what if the murderer and the person who concealed the murder weapon were different people?’ So while I knew that it was important, I didn’t know how much confidence to place in it as a puzzle-piece. And the trick is that it’s difficult to test that kind of assumption in a game – Holmes can set up a trap to confirm that the boot-black is only pretending to be left-handed, but that kind of creative experiment is really tough to pull off in IF.
This is why you could theoretically bark without gathering evidence at all. Holmes has already put the puzzle together himself. Your bark is just the final test, as he explains.
Ah. So the way my meta-gaming screwed me was the assumption that it shouldn’t be possible to brute-force the puzzle without gathering enough information first.
And taking that leap by barking at who you believe to be guilty is really essential. Telling a player that they’ve got it right in advance, before they’ve barked, would deflate the game.
I agree, but I think it would have been helpful to confirm some of the mid-level connections. Towards the end I felt I had so many connections at the decent-guess level that things felt a little like a tower of cards, and I wasn’t confident about building bigger conclusions on top of that.