This is a list of several projects that i've worked on that have ended up not working for one reason or another. This happens sometimes—it's an inevitability of experimentation that results will not always be what is wanted. Nevertheless, such results can still be useful—if nothing else, they can serve as a lesson on how to not do things. For that reason, i feel like it's worthwhile for me to publish the results of these experiments of creation anyway. It can't hurt.
(Note: Not only will i certainly fail at more projects in the future, i also genuinely cannot remember all the projects that i've failed in the past. For that reason, this document is inevitably going to end up getting updated over time, both as more projects fail and as i remember projects i had previously forgotten. However, it will not be updated with any sort of consistency or regularity—this includes the possibility that it will never be updated again. For this reason, i would recommend against intentionally seeking this document out again. This also precludes the existence of an RSS feed or whatever for updates to this document.)
I've written and put online some short stories in the past. However, in that time, i've also come up with some ideas for short stories that didn't end up working. In some cases, they died in the idea phase, and in other cases, i tried to write them and only then figured out that they didn't work.
One of these ideas was a short story that would consist of a lone adventurer going off to take down what they had been taught by their community was a great evil, failing and getting captured, learning that the group that they had thought was evil was instead one that was sympathetic to the people that the community which the adventurer had come from was oppressing, and defecting to their side. This is something that i think would work a lot better as a novella instead of a short story, which is part of why i think i couldn't get it to work. However, i also found myself unable to get it to work because i couldn't determine how to pull off the learning step of the plot subtly enough that the story wouldn't come off as being a thin covering over a personal rant.
Another one of these ideas was a short story set long after some theoretical apocalypse in a small, peaceful community that had sprung up in the wake of the apocalypse that persisted off of the remains of the apocalypse by moving around their settlement to new places with frequency in which, shortly after one of these moves, the initial expedition team stumbles upon a single lone well-armored skeleton in a small well-fortified base and brings it and some of the materials found at the base back to the current location of the settlement, whereupon such materials are investigated and reveal that the corpse was that of a doomsday prepper. I couldn't figure out how to pull off this story in such a way that the suffering that the person had gone through that had lead to their death would be portrayed in a sympathetic light instead of a blameful one.
The International Obfuscated C Code Contest is an international contest of obfuscated code written in the C programming language. I sought to write some entries for it that i would have submitted the next time the contest opens (it's been a while, but it certainly doesn't seem to be dead).
Unfortunately, i couldn't figure out how to do any of them. I don't have much to say on any one in particular, so here's just a list of them:
SIGUSR1 and SIGUSR2 signals. This wouldn't actually work because of how signal transmission works, but could be done with realtime signals instead, but i wouldn't be surprised if someone has already done that.SIGFPE and then dividing by boolean conditions. This is not very creative.DYNLOAD.c, RISCVASM.h, SELFPROCESS.c, and UNICRAP.c, and i don't remember what any of these ones were supposed to be.This project demands a more thorough explanation.
Warning: This program used the Simple DirectMedia Layer, SDL.
New Ithkuil is a constructed language engineered for precision and expressiveness in communication while still being terse to the point of being usable. (It is at this point that i should probably point out that it has been a long time since i've interacted with it.) It was created to improve on its predecessor, Ithkuil, by being more learnable and regular, while also being expanded so as to fix some other flaws that had been discovered with time.
Nevertheless, there were certain facets of the language that unavoidably needed to be designed to require a speaker to just memorize stuff. I decided to write a computer program with the intent to assist in this. The premise was that it would ask the user questions about the various facets of the language (which facet to work on memorizing would be selected by the user) that were procedurally generated based upon JSON files describing each element of that facet of the language and how to construct questions from them.
The project didn't get very far at all—all that i managed to get done was the aspects and the cases of the language. This was in part because i was much less experienced with programming, but also in part because i was trying to write the program during a time in my life where i had much less time and energy to do anything.
I don't intend to ever come back to this project. I am much less enamored with New Ithkuil than i used to be. But if i did come back to it, i would do a lot of things differently—not only have my standards for what constitutes good code changed, but i also wouldn't make this a GUI program (in particular, SDL had no business being used here), i wouldn't use JSON for the assets (though this is in part because there is no good JSON library for C), and perhaps the biggest change…
…i wouldn't make the program at all. You may have noticed a certain problem with the program: it is entirely pointless. There isn't any good reason for it to exist—not only could you probably just use flash cards if you really wanted to, but it's also much more effective practice to actually, y'know. use the language. You're going to end up memorizing stuff by doing that anyway, and it also turns out that using the language is excellent practice for using the language.
The source for the program is available under my old GitHub account that i don't use anymore. Be warned that GitHub is owned by Microsoft.
Warning: Always use a privacy protector such as Tor Browser when using any Google services.
In the Google search engine's recent pivot away from giving results and towards giving “answers”, there have been some search terms for which Google will provide its own implementation before the search results. The classic game of snake is one of these terms. But this is no mere implementation: it has a number of extensions to the core gameplay available for futzing around with.
Having previously seen a description of the process of implementing a mathematically optimal algorithm for the standard game of snake, i found myself curious about what differences the extensions that Google's implementations of snake would make to such an algorithm, and so decided to set about constructing an environment in which i could experiment with such modifications to the algorithm.
This project suffered from similar problems to the previous one in this list, those being that i was less experienced with programming and had little time or energy to work on this. But in this program's case, the single biggest problem with it was that i decided that the best way to implement the ability to experiment with the different extensions to snake provided by the implementation was through making available an API for other programs to use. This was not a good idea—nowadays, if i wanted to create a game that was both human-playable and machine-playable, i would probably just have the program source be as fit for modification for machine-play experimentation as possible.
I also later found out that people had already implemented mathematically perfect algorithms for the extensions to snake provided by Google's implementation anyway, so this program would be wholly redundant.
The source for the program is available under my old GitHub account that i don't use anymore. Be warned that this program uses SDL.
Y'know how when you start up Minecraft, there's a little yellow bit of text pulsating tilted underneath the logo? That's called splash text, and for some reason, i decided to create a library for that.
Who was the target audience of this? Splash text is a thing that is done by just one game. Even weirder is that i way overcomplicated this by including a bunch of different parameters to configure for selection from files in an overcomplicated format, when the one game that does this just picks a random line in a plain text file.
Not only is the only sensible way to do this dead simple for anyone to reimplement on their own in a way that will better fit their need, but even if you do want to use an existing overcomplicated thing, there already exists one: it's the program called ‘fortune’.
Here's the program's source if you want it, i guess. I don't really know what anyone would do with it.