If only generative AI never produced utter nonsense!
Personally I think this is a solid litmus test. Are you giving input, or a prompt? Do you expect consistent output (the same Japanese text should always translate to basically the same English text, the same misspelled word should always give the same suggestions to fix it), or are you expecting creativity out of the system? There’s definitely creativity involved in proper translating, but I don’t expect Google Translate to provide it.
There are still some edge cases, but I think the vast majority of them are pretty clear. And when I wanted to get into some of those edge cases last year (using a Markov chain to generate AI-sounding gibberish), I just asked the organizers and got an answer.
I am tempted by the La Petite Mort category, but is it reasonable to think that one can create a game with a parser using PunyInform in less than four hours? Any other type of game seems unrealistic in four hours, but I am slow at this kind of task.
I suppose that since the category La Petite Mort is displayed in French, everyone knows what this expression means.
I’m fast, but I’m not that fast. It’s hard to do a parser game in 4 weeks let alone 4 hours. Remember that the 4 hours doesn’t include thinking time and design time, only coding time.
Would time spent on thumbnail/cover art count towards the 4-hour time limit? Does it matter whether the cover art is shown in-game vs. just on the itch page? What about art that could be considered part of the styling (e.g. decorative borders for text boxes)?
It’s tricky, but definitely possible! The key is to do as much planning in advance as you can, and play to the library’s strengths, relying on default built-in behavior wherever possible.
I have written a parser game in Inform 7 in 4 hours for Petit Mort before–so it can be done! (I spent another hour after the comp closed on fixing things up, but the official EctoComp version was just under 4 hours.)
In addition to what @Draconis notes (“as much planning in advance”!) I would recommend focusing on a very small map with a limited number of available actions and pretty basic descriptions.
I’m planning to participate in ECTOCOMP for this first time this year. Really looking forward to it!
While we’re on the topic of advance prep work for La Petite Mort games:
Am I correct in thinking that creating a pretty detailed outline ahead of time is “fair game”? Say, for example, a list of story beats and specific actions the player may take in a given room/scene. (Stopping short of writing actual prose text or code that appears in the final game, of course.)
I have a concept for a choice-based game that I might try to write for the category, but it’s complex enough that I can’t really imagine making the four hour time limit without that sort of advance planning!
Yeah, my usual strategy is to make maps, flowcharts, puzzle diagrams, everything I possibly can before I launch into the actual code and writing. I did the same thing for IFComp this year and it seems to have worked out well!
Same here. For the game linked above, I had several pages of handwritten notes with all this sort of stuff on it. The “4 hours” measured in ECTOCOMP time was “writing, coding, testing, and fixing” time. (And as some people pointed out in the contest, I still did not do a great job of mapping. Apparently I get east and west mixed up a lot. )
Im planning on entering a Petite Morte game called The Last Gunfight of Joseph Smith. It’ll be based mostly on primary documents about Joseph Smith’s death and incorporate a mechanic allowing you to switch between viewpoints (either two columns of text or a button to flip between them, so including articles by his enemies and by his followers). It’ll be fairly grisly, as the background story includes dismemberment and child death (and ends in a assassination).
I think you almost have to have a good idea on where you want the game to go before you start a petit morte . (Well, I find I need to anyway). You’ve only got 4 hours writing/testing/editing time which isn’t much to get a functional game out there. I usually do all my brainstorming, decide on a storyline/major choices, make up an image to set the tone etc to try and get everything as organised as possible so I can just start writing/coding when I hit the timer. I also encourage noting which parts of the game might be able to be added in/dropped if this is something possible. I had one occasion where I had to drop a scene from lack of time. It’s not ideal (it did mess up my stat balances a bit) but at least I got it done within the time frame. (Ideally it’s the other way around, try not to over commit, and add more content if there is time to do so… which there often isn’t, but that’s what post comp releases are for .)
)
. (Well, I find I need to anyway). You’ve only got 4 hours writing/testing/editing time which isn’t much to get a functional game out there. I usually do all my brainstorming, decide on a storyline/major choices, make up an image to set the tone etc to try and get everything as organised as possible so I can just start writing/coding when I hit the timer. I also encourage noting which parts of the game might be able to be added in/dropped if this is something possible. I had one occasion where I had to drop a scene from lack of time. It’s not ideal (it did mess up my stat balances a bit) but at least I got it done within the time frame. (Ideally it’s the other way around, try not to over commit, and add more content if there is time to do so… which there often isn’t, but that’s what post comp releases are for 
.)
