Hebe

I am, as I think basically everyone in the world not on the OpenAI or NVIDIA payroll must be, heartily sick of LLMs. Substantively I think they suck when employed for anything resembling creative work; environmentally, turns out the companies have been lying and things are even worse than the already-quite-bad picture that’s been painted; economically, they represent yet another attempt to consolidate wealth and power with the owners of capital at all of our expense. But more than that, the discourse around them is wearying – having to think about this stuff and engage with those conversations even in leisure activities is pretty frustrating. So I should disclose that I went into Hebe, which bills itself as a puzzley adventure through Greek mythology, and also commendably acknowledges that ChatGPT wrote some of the text, with some trepidation, both about what I would find in the game and what I’d need to write about it.

There’s good news and bad news, I suppose. The bad news is that those Hebe-jeebies were justified (Hebe’s the Greek goddess of youth, so her name rhymes with Phoebe – geddit?). All of the game’s room descriptions are overlong and mention lots of unimplemented objects, the prose glops about, weighing everything down like oatmeal laced with lead, and by the end my eye would start twitching whenever I read the words “flickering” or “serene.”

The good news, such as it is, is that I didn’t find it too hard to ignore most of that. The game is clogged with empty locations that are just there to pad out the half-dozen places boasting self-contained puzzles, and so it’s relatively easy to just glide through them and concentrate on the interactions that seem like they were written by a human being, just as I skimmed the overlong cutscenes. Similarly, the author’s offered a helpful INVESTIGATE verb that tells you what items actually exist to be interacted with, so you don’t need to play the guess-the-hallucination game with ChatGPT if you don’t want to. And turns out that under all the AI cruft is – well, a perfectly ordinary undertested, wonkily implemented game that manages to boast a bit of charm, the kind of thing that would be a perfectly respectable starting point for a new author who didn’t yet understand the level of polish a parser game requires, but for the LLM use.

Let’s start with that bit of charm, as I’ve given Hebe a bit of a hard time so far. It’s clear that the author took the theme seriously, bringing in a lot of fringe detail from Greek mythology including but not ending with the choice of protagonist – Hebe’s both the second-most-famous cupbearer to the gods and the second-most-famous wife of Heracles, so I appreciated her underdog energy. The game proper involves visiting various sanctuaries and temples across northern Greece to find the Olympian Gods, who’ve been defeated after a surprise attack by the Titans and left depowered and chained behind various puzzley barriers – and then venturing into the underworld to find Heracles and bring the fight to the Titans. Again, there’s a fair bit of attention to detail, with numerous cities and ports implemented, a visit to the Pytheia, minor naiads given a supporting role, and plenty of obscure bits of myth getting a name-check.

The thin implementation means that the names are often all that you get, however – “you see nothing special about the Charon’s boat” is an enthusiasm-killing phrase to read. And the seams between the LLM stuff and the chattier human-written propose are sometimes comically sharp:

> x aigle
Aigle radiates with a golden glow, her hair like cascading sunlight, and her eyes shimmering like the first light of dawn. She is the embodiment of brightness and warmth, her presence illuminating the garden with a serene, golden aura.

> talk to aigle
“I’m so relieved you’re safe! Now go show them what you’re made of, Hebe! Just like old times!”

A game written entirely in the latter style might feel a little silly, but could have some zip to it; the juxtaposition with ChatGPT’s overwrought descriptions just creates bathos.

As for the substance of the adventure, the puzzles are largely old chestnuts. There are a couple of codes, a put-the-right-object-in-the-right-place one with a poem providing the hints, a guess-how-heavy-the-unmarked-weights one… None of them break new ground, but the classic are classics for a reason and they could be fun to work through. Unfortunately bugs and incomplete implementation make many of them way harder than I think they’re intended to be – the weight one stymied me for a long time due to the fact that I wasn’t clear that there were two scales, not one, and I couldn’t directly interact with the first one (“which do you mean, the scale or the small scale?”); the object-placement will softlock you unless you get it right first try, because the game incorrectly thinks one of the slots stays full even after you remove an item from it; and the endgame seems to have gotten especially little testing, as accessing it requires going through an unmarked exit (tip to other players: try IN/ENTER when you’re near the Necromanteion) and then the climactic conversation with Heracles is made awkward by a YES/NO are-you-ready-to-proceed choice that doesn’t work (to continue with the service journalism: say YES and then manually type DOWN afterwards). And there’s a lot more besides; see the transcript for the gory details.

But again, pretty much all of these issues are familiar ones – heck, I’ve committed some of these sins myself, and but for lucking into experienced testers for my first game could have wound up with similar egg on my face at my debut. And as I’ve said, I do think Hebe comes from a place of real authorial excitement, some of which occasionally comes through. So it would be easy enough to just wrap up with my typical remark in these cases, about how I hope to play a much better second game from this author. Which I do!

My weariness at LLM discourse can’t prevent me from adding, though, that I think use of ChatGPT was an especial disservice to the game. Beyond weighing down the enthusiasm that’s often one of the best elements of a debut game, the use of AI I think might have created some bad habits that contributed to the overall weak implementation. As I mentioned, the long location descriptions include a fair bit of (bland) scenery detail, and call out sounds and smells, so it’d be easy enough for an author to review what ChatGPT spit out and feel like their bases were covered – but none of the scenery is implemented, and LISTEN and SMELL return their default responses throughout. The difference between the prose styles also makes it really obvious to the player where there’s stuff they should be paying attention to and what they can safely tune out – as an author, though, I don’t think you ever want the player shutting off their brain. I don’t think I can say with a straight face that the version of Hebe that didn’t use an LLM would have been significantly higher-quality, but it would have been a clearer reflection of the author’s vision, and probably a much much better learning experience to boot.

Adapted from an IFCOMP24 Review

Why does anyone do this reviewing thing? No one reason, obviously. For me it started as a simple impulse: to try to give something back to the community in advance of asking it to consider my own efforts. It quickly got a lot more complicated. It turns out that the prospect that my words might help someone refine their art gives me hope that I have more to offer than raw snark and good intentions. Underpinning a lot of it is admiration for the medium and the artists that continue to transform it beyond anything dreamt of in the early days. There is so much negativity in the modern age, an opportunity to find things to gush about makes me just a little more resilient and centered.

So yeah, it’s all about me.

The common thread to this miasma of feelings is connecting with the work of another human, then further connecting with humans that have also explored that connection. So. What does this mean in the encroaching age of generative AI? This is a work that embraces new technologies to produce art, acknowledging its debt to automation to produce text. But most IF, especially parser IF, IS text. Where does human authorship stop, and machine authorship begin? Is there a line where machine authorship reduces the human part of the art? At what point am I inadvertently connecting with machine? And why on earth, given the things that motivate me to hammer out words for you, would I want to do that?

There is an argument of ‘so what? What does it matter if the work makes you feel something?’ Ok, fine. But what if it doesn’t? What if the words are capably rendered, the scenario clearly and adequately painted, but ultimately just flat? Then what? If authorship were unambiguously human, I would endeavor to show where and how that impression developed or missed the mark. But if it is because machine? I have NO interest in providing feedback to a machine that in the best case, has no way to digest my observations, and in the worst makes itself BETTER at a human endeavor I wish it weren’t involved in in the first place.

This is a Greek Myth IF, where as the titular protagonist you are asked to free your god brethren by solving IF puzzles. Last few years there was a spate of art that recontextualized and transformed Greek Myth in fascinating and revitalizing ways. This is not that, this is a pretty straight-ahead representation. Find some trapped gods, solve puzzles, on to the next. The gods themselves have no particular character or personality hooks, no neat twists, and rarely escape their familiar lore. If fact, if NOT for that lore their characterization would be nearly non-existent. How much of that is AI, and how much author choice? It certainly seemed to be missing a spark of some kind.

It isn’t helped that the gameplay is demanding in the least satisfying way. Early on, the difference between traditional cardinal directions, ‘go to,’ and ‘sail to’ is unclear. Its nouns are wildly uneven in their implementation - meaning most small details respond with ‘you see no.’ This trains you not to poke too deep. Until some puzzles REQUIRE deep dive into nouns no more or less prominent than their neighbors. NPCs, arguably the MOST human-adjacent aspects of IF, are similarly completely shallow (dare I say, robotic?). They have information to impart, but with almost no character voice of their own. Interactions outside that functional purpose generate a ‘you get no response’ Even when asking about, say, a trapped spouse they have just asked you to find!

The effect of all this is to highlight the mechanical moving parts at the expense of idiosyncrasy and unique human voice. Then to try to hide those parts behind capable text that more obfuscates than enthralls. The combination of all that is that puzzles are much harder than they should be - depending on if you poked at the right noun or not. It was pretty clear what needed doing in most cases, but the mechanics of finding missing pieces to do them were obtuse. In one case I literally turned rings to a near-random combination and it worked. In another I waited until the solution presented itself, just waited. The combination of obtuse yet also anti-climatic was off putting for me.

It also hit what seems a pretty big bug. Per the text in one location, both the Agora of Thebes and Mount Olympus were N. Going N though took you to an empty location. I think this made the game unwinnable (intrusive if not unplayable, per my rubric), as a pair of gods needed to complete your rescue were clued as being there. I spammed some commands just to see if I could power past to no avail.

I’m not thrilled that my first review of COMP24 comes across so negative. There is every possibility that being told AI was involved colored my response, I leave that to the reader to decide. There is every possibility that the work’s shortcomings have nothing to do with AI at all, and just needed more refinement. Between the flatness of the scenario and characters, and uneven puzzle implementation I guess I would RATHER attribute these things to AI. For sure, I want more humanity in my art!

Jeez, first game of Comp, and I am spiraling into existential angst and techno-paranoia. Buckle up folks, I’m turning into a curmudgeon before your eyes!

Played: 9/1/24
Playtime: 2hrs, score 30/maybe 90? (4 gods rescued)
Artistic/Technical Ratings: Mechanical/Intrusive Implementation gaps
Would Play Again?: No, engage IF for different thrills

Artistic scale: Bouncy, Mechanical, Sparks of Joy, Engaging, Transcendent
Technical scale: Unplayable, Intrusive, Notable (Bugginess), Mostly Seamless, Seamless

In this game, Kronos has imprisoned the Gods, and you must rescue them two by two.

This game acknowledges it was written with ChatGPT, so I won't spend much time (if any) discussing that, as it seems the purpose was to make sure descriptions were descriptive, and they generally were. I knew that it would hallucinate, so I ignored most of what the text said except for objects that were easily interactable with (and a command INVESTIGATE let me know what those were, most of the time).

The map is several different cities, each of which can be moved between fairly easily. Sometimes the exits list were incorrect (like S vs SW) and one, the Necromanteion, isn't listed (you have to ENTER when you're north of the city that is near it).

Puzzles are generally complex code-type puzzles. I used decompiling to figure out rules for some of them. Players will need to know they can ROTATE something COUNTERCLOCKWISE or the game is impossible to beat. Player's should a know that you might be able to put things on a statue's head when you can't put things on the statue itself or the shelf that is on the head. A certain YES/NO question glitches if you type YES, but you can just type YES then NO and it treats it like a YES.

The HELP command here is useful, because if players don't know they can WAIT 11 HOURS, they might have to type Z dozens of times.

Overall, some of the puzzles were fun. I liked the one with flowers. What this game needs, in my opinion, is more careful puzzle testing and more bug-fixing in general. Having one dedicated gametester or several less dedicated testers who report bugs and an author who has time to fix those bugs could make this game a lot of fun. The puzzles are the main draw here, and the overall story idea, with everything else as set dressing, so I'd love to see them shine even more.

(I do have a better impression of this game because the author was open about using ChatGPT rather than hiding it. It is often clear that an author used ChatGPT, and if they do that and don't disclose it, it gives me a much more negative view of the game).