Project ideas?

I really need to take on some type of techy project soykaf I've been rotting in for the past few weeks and I can't deal with it anymore, I really want to take some BIG soykaf on but I'm just a beginner so I don't know like anything. any ideas(USER WAS WARNED FOR THIS POST, You should always elaborate on your opinions rather than just spilling the soykaf.)

The best thing about tech is how it interfaces with your favorite virtual experiences! just look for something you enjoy, (movies, games, music) and find ways to interface with it. Eg, you like movies? set up a jellyfin media server with radar/sonarr to manage torrents downloaded with qbittorrent. You like minecraft? learn how to code in it. You like music? Try writing a program to rip your favorite songs from youtube, using yt-dlp or something. There's a lot of things you can do so don't just say you dont know anything and ask people on the internet for help. just go and do it

>>76513 first, consider just going on and learning the language/languages via books and doing exercises availible in them for better understanding of how things should be done. if you just need a project practice your skills, do some casual useless stuff. you didn't mention what language or even what field of software development you are studying, so choose something for yourself depending on your interests, just like the guy in the post above said.

>>76516 thank you so much honestly :3

>>76513 learn standard ML ������������������������������������������������������������������������������������������

>>76516 >put your limited time into consooming more entertainment slop I've been thinking about but couldn't figure out a decent system to take apart arguments into their components, including their premises and assumptions they're based on in something like a handy, easy to traverse circular graph (similar to a flow chart) where also evidence supporting each claim could be attached or a similar argument with different points could be painlessly added and duplicate "argument nodes" could be collapsed into one. This would be a great project since it would teach you more about putting something complex since you have to come up with an architecture, a skill that pays off no matter what programming language you will go with and even if it doesn't go anywhere, you will make faster progress than putting some python scripts together for some home media "server". But on the off chance you succeed, you'd have a solution to something which is plaguing online discourse and could filter effectively most if not all sophistry and bots and reduce flooding to a catalog of unique views while helping people who want to promote truth but are to much of a sperg to bibliographically retain and recall facts, events and evidence or drop their spaghetti during debates. Plus points if that system can be synced with other's on a per "curator" basis. This way different ideological circles could map out their own arguments and counter arguments and more sane in the middle people then could merge and diff the middle grounds between extremes. If succeeding, this could potentially untangle a lot of the noise which is paralyzing productive discourse today.

>>76567 I'm intrigued! You said you've been thinking about such a system, but couldn't get the framework right. From which approach / what direction or with what principles did you start thinking about it, if you dont mind me asking?

>>76568 My brain just couldn't get warm on this one for the time being. But the concept is fundamentally based on the idea that arguments are a chain of explicitly stated premises and a magnitude of unspoked yet still relevant further presmises. Many of which are assumed to be known by the other person argued with but often they don't. This leads to both talking past each other despite speaking the same language or allows for exploits by ideologues and evangelists. Where sophists, just like a sales person have a path or strategy to sell you something. In this case an idea or view. You can imagine this like a chess game where each move leads to a different set of possible counter moves by the opponent. Just that it's even more complex, where sales tactics like "make them say yes" can work well or a dishonest oponent stuffs their response with a wall of premises to cause an analysis paralysis. The way chess masters beat a machine, at least pre neural networks was by getting a check mate before a certain amount of chess pieces was reached, since at that point the computer would be able to switch from "computational intuition" to using a precomputed graph of all the possible moves and outcomes and follow the available step to the scenario where it wins. A tool or system like I've described could provide a web of distilled arguments and the relation between each of those, effectively putting someone in the position of that chess computer running on precomputed moves which would make these sales pitch strategies for debates much less viable and easy to update whenever a new one is concocted. Also could this break propaganda chat bots since they, too merely run on a script. While also saving time since I see so much effort put into having the same arguments over and over again. But while honest participants think they've accomplished anything. Often this gets undone the next day by the same malicious actor or entity by simply flooding with a repetition of what has been refuted just the day prior as if that argument has never taken place. So having these arguments mapped out would save everyone a lot of time. But that's just the idea. Technically this whole thing would be relatively trivial. It's mostly just displaying some connected nodes and allowing to traverse them and adding new ones that are saved in some local db or file system. With free choice of libraries, platform and framework. But the design is the tricky part and cuts into a domain I'm frankly not good at, hence I'm interested in what others can come up with. Now I've seen some attempts at implementing just this with similar motives but I haven't been to convinced of them nor have I been able to find those projects again. But I remember one that was in development considering to use an up/downvote system and it being centralized, even if it was just a single dev server. That much I know that this is a bad idea and instead it should be offline capable with the option to synchronize someone else's argument db into your local system, with all the premise and counter premise entries, evidence data, further elaboration on the possible interpretations, variants of statements and languages and so on. Whether any point is good or not is then decided on whether someone considers it as worth keeping. And if we trust in the objectivity of reality, even if the software is obscure, the entries would be either timeless or of historical nature. Meaning usage can be revived or info databases updated by anyone at any time. Even if the initial site goes down. But I struggle to figure out how such a flow chart type of system could be implmented, or rather designed best. LLMs would likely help matching arguments in the db for either search or grouping on similarity but say one makes a statement like "there is no ethical consumption under capitalism". Now is this true? I'm not sure, nor do I want to discuss it here. I picked this example because I found the "capitalism vs socialism" debate to be one of the more agregious instances of having the same debate over and over again to the point it's just a game rather than to aggregate knowledge on what works and doesn't in a civil manner. But I do know that there's a long chain of presumtions under that one sentence. Like what can we interpret to be meant by "capitalism"? Thus this would spawn a bunch of related nodes for all the relevant definitions and then in turn another set of nodes for each on whether that definiton can be supported or not. Resulting, if filled out dillegently in a node or graph database where one can effectively deep dive into any given topic due to articles, books, stats, arguments, pictures, historical timelines and other points of evidence linked either in support or refutation of each other. Or skim the surface of multiple different topics. It's a tool which would cut through the information overload we suffer today and endless repetition of falsehoods and sea partial truths. But the exact mechanics of it are currently beyond me. I hope I don't sound too much like a schizo but I think it's very much worth a shot to figure this one out somehow.

>>76567 Standard ML (>>76566) would be great for this as it requires artificial intelligence capabilities and that is just what this language was made for! Learn standard ML. That will be enough to keep you busy for a month and since ML is so AI it will probably make you round out and realize new ideas for your project in the process as well.

>>76570 Have you heard of argdown, the markup language that does exactly what you want? I'm using it to clarify complex arguments I read in my research. Also take a look at the website kialo

Solving a problem that actually exists in your life. Any other idea is useless and not interesting by default. If you have no problem you will have no reason to continue working on it after you figured out how you can solve it, before you finished writing the code. Most of my projects are just python/bash scripts for automation, because most problems that require more were already solved.