Aisle

Aisle is one that I finally broke down and played a few years ago. It did not impress me, and I didn't find the central innovation (i.e. one-move game) to be very interesting as a mechanic. A recent re-examination of the work and its reviews in greater detail changed my opinion of it -- for the worse.

The game's tagline calls itself: "an instant in the life of a man" [emphasis mine]. Likewise, many reviews mention "the story" [again, emphasis mine]. But there is no single story. This is made clear by the "frontispiece" displayed upon opening Aisle, which at its end states: "Be warned; there are many stories and not all of the stories are about the same man."

As many reviewers note, extended interaction bears this out. It's not possible for the PC to be just one man due to conflicting information presented in various "endings." (I put scare quotes around the term because it seems clear that no one outcome is intended to represent a complete experience, thus the interactor is almost certain to "replay" many times before deciding to stop.)

Obviously, on a purely structural basis, the interactor of this work is expected to try at least dozens of different moves in order to observe the various bits of past and/or future that each new "ending" produces. Theoretically, that structure does offer an interesting possibility, in that the motivated interactor can try to suss out particular details and build up a picture of the whole possibility space of the protagonist's future as well as the complex of factors in his past that shape his present. To my mind, the kind of meaningful question asked and answered by a work of this structure should be: How can one action now change the future of this person, and what are the limits of that change? (As Duncan Stevens puts it, the point would be "to explore the central character and take a look at the various possibilities available to him from one point in time.")

Unfortunately, since this work is not consistent in the details that a diligent player uncovers, any meaningful exploration of those questions is abandoned, and the whole concept reduces to an exercise in style. Sean Barrett's I'll does a good job of making the point in an exaggerated manner via rhetoric-in-action.

The central premise of the scenario is weak and seems particularly ill-suited to a weighty treatment of the possibilities opened up by the novel mechanic. (As the review by Ben at Trotting Krips notes: "It is difficult to turn your life around in the middle of a Winn-Dixie.") The inciting incident is that the protagonist has a moment over the fresh gnocchi at the end of an unpleasant day. Prior to this, he seems to be just have been going aisle by aisle through this, his regular grocery store, in a run-of-the-mill bit of shopping. Is this the first time the store has stocked fresh gnocchi? Has he never noticed this item before? Has he forgotten that it was there? Why does *this* time seem so significant? The work offers no answers -- does not seem to have anticipated the question.

... Or maybe it has. Perhaps the added significant context is the proximity of the brunette, whose appearance reminds the PC of a woman, Clare, who features so prominently in many responses.

The work is noted for its 100+ unique "endings," but the range of supported actions is not well-correlated to the situation, making it hard to predict the effect of many commands. For example, if >SLEEP is tried by the player, it is because he or she is in "lawnmowering" mode and that is a default verb, not because it makes any sense for the PC to actually go to sleep. At least some responses use unique verbs that are unlikely to be discovered without a walkthrough or decompilation. It's hard to know what to make of responses to commands such as >GIBBER or >DROOL, which are highly unlikely to be tried without external prompting.

Interactions with the woman tend to send the "story" on an unexpectedly dark trajectory. Stevens notes that the PC "often treats apparently normal conversational gambits as an excuse to act psychotic." While I immediately agreed with this sentiment, that description is, on balance, overstated. The author usually chooses an unpleasantly surprising spin on the interactor's choice of actions involving her, even those that seem innocuous, in a manner that strikes me as a deliberate exploitation of the coarse-grained interaction afforded by the ASK/TELL model, which collapses many potential variations of topics into one keyword or keyphrase response. Since the woman is by far the most interesting part of the environment the player is certain to stumble over one of these warning signs soon enough, perhaps in response to >ASK WOMAN ABOUT PASTA or >ASK HER ABOUT HER CART. Sure, asking someone about the contents of their cart at the grocery store is an unconventional conversation starter, but does Barlow think that the PC's subsequent actions truly represent the player's intent? Compare those two responses to >ASK HER ABOUT HERSELF or >ASK HER ABOUT GNOCCHI for similarly unpredictable results in what is presumably intended to be a positive direction.

Inevitably, the interactor reaches a point in exploration where, to quote Stevens again, "options for civilized behavior run out." If one stops before that point (and has by happenstance missed the yellow flags presented when interacting with the woman), then it is indeed possible to come away with a sense that this is a single story dominated by themes of "romantic isolation" and "melancholic longing" as labeled by Jimmy Maher's review, or even one that is "beautiful and tragic" as described by manonamora's. However, the work itself does not stop there, and a completionist will soon discover huge swathes of additional "story" that are very dark indeed, with possibilities ranging as far as assaulting the woman in the store or the revelation that the PC is a murderer. When one bears these darker outcomes in mind, it becomes more difficult to interpret the "positive" outcomes as good things, after all. (In fact, on closer inspection, even some of the presumably-positive outcomes involving the woman seem tinged with a sense of predatory manipulation.) As one progresses through less and less likely significant responses, the PC becomes steadily more repugnant.

Between the single-response constraint and the author's capricious imputation of the intent behind any given command, I'm hard pressed to say what it is the player is supposed to be actually doing after loading up Aisle. When interacting with this work the interactor is not actually exploring a character or a story -- between the limited and/or contradictory information about the context of the single choice point for the protagonist and the extremely low ability to predict the impact of that choice on the situation, this work is missing the heart of what I consider interactive fiction to be. There is nothing solid to discover through persistence, and although any one response often resonates to a greater or lesser degree with others, the player is hard-pressed to even draw firm boundaries between the various major permutations of the "true" situation, which seem to bleed into each other in places. In the end, we are only exploring the author's momentary whimsies until patience and/or interest runs out.

As noted in the review of Mike Root, there is an alternative lens for evaluation of Aisle, which is that the work's purpose is to "play the player." The best evidence for this interpretation is the way that certain clusters of responses do paint a semi-consistent portrait of a certain type of person with a certain type of past. However, these responses do not seem consistent enough to look like deliberate planning on the author's part. For example, if the player focuses on mundane goals, the commands >EXAMINE GNOCCHI, >EXAMINE PASTA, >TAKE PASTA, >TAKE GNOCCHI, >TAKE SAUCE, >ASK WOMAN FOR SAUCE, and >PUSH CART are all straightforward options, but they each define significant aspects of the PC and/or his backstory in a manner that the player can't possibly predict. For example, >PUSH CART imputes an active anger to the PC over being confronted with the suppressed memory of Rome and edges slightly malevolent toward the brunette, while >TAKE GNOCCHI suggests long-settled resignation about the outcome of the events of Rome and ignores the fellow shopper. >TAKE SAUCE yields a Clare-less response implying that Rome wasn't that significant, after all, and >ASK WOMAN FOR SAUCE indirectly suggests a painful episode from which the protagonist might be starting to recover (or perhaps, taking some responses into account, something less pleasant). Although the outcomes differ significantly, I don't see any consistent logic mapping those significant differences in outcome to significant differences in the player's choice.

In summary I would say that Aisle does best when engaged with on a very shallow basis -- one can argue whether or not that necessarily means that the work itself is shallow. I think I would have liked it better with fewer implemented responses, and especially if Barlow had chosen to trim out or simply not respond to various default verbs related to violence, as it does for some of the stock actions in Inform 6's Standard Library. In that form it might live up to the praise in Jimmy Maher's review, even as it was reduced in scale to a bite-sized 15-minute novelty. I would definitely have liked it better if all endings were consistent with one another such that discerning the protagonist's total past becomes a genuine exercise in discovery, which would let the diligent player reach the point of making an informed and more importantly intentionally significant choice of the single allowed action.