I have a three year old daughter. In her three years of life, I have recorded hours and hours of high quality audio of her. I've also encouraged her to bang on keyboards and drum pads and guitars, and recorded all of that, as well. I then took that material and chopped it, mutilated it, sequenced and mixed it, and released two albums of what I'm calling "classic industrial music", all under a pretend band name, with the credited musicians being my daughter, and our teenaged dog, who is also featured on many of the recordings.
Naturally, I then released these two albums to the world, where they have sold roughly 0 copies. This doesn't bother me, though; that was never the point. It is, in essence, a long-term elaborate "performance art" piece, that is intended for my own amusement, and possibly my daughter's amusement, once she gets old enough to "get" the joke. In the immediate time, I get value out of seeing her enjoy playing instruments and making sounds. And I'm giving her the experience of making music, in whatever way she sees fit, before she ever gets taught a bunch of nonsense about what ways are OK and not OK to make music through more formal training.
Any project that naturally holds your interest, is probably worth pursuing, regardless of the overall value to humanity. At least for a little bit. Because you're generally going to walk away with knowledge you didn't have prior to starting it, and benefits you wouldn't have predicted had you not gone down the path at all.
Oh yeah, also; don't take advice from me. I make horrible choices, then I commit to them.
You have vastly overestimated how much money that one sale will net me, but I will be donating twenty dollars of other origin to Planned Parenthood, as agreed upon.
I collect bookmarks of things I want to read, but inevitably will never have time to read. Sometimes I go back and look and the webpage domain is expired, or the page 404s. But I still meticulously folder and tune my bookmarks
I’ve never admitted this to anyone, but it feels good to get that off my chest
I think that needs to be coupled to something implementing Douglas Adams idea of an Electric Monk:
"The Electric Monk was a labour-saving device, like a dishwasher or a video recorder. Dishwashers washed tedious dishes for you, thus saving you the bother of washing them yourself, video recorders watched tedious television for you, thus saving you the bother of looking at it yourself; Electric Monks believed things for you, thus saving you what was becoming an increasingly onerous task, that of believing all the things the world expected you to believe."
I did the same for a while, but it was a mess (700+ unsorted bookmarks on my main computer, 100s more on others).
I tried shaarli, but soon after I tried to build something myself, and I created share-links.
It's an open-source Django app that you can self-host, and that lets you store (and share!) links, titles, descriptions, and tags. Then it display them in a nice way (for me : not much css, a simple page with no js).
It took some dozen hours to get to the point where it's really usable, and it still have problems now (comments are not moderated, I just realized that you can't add a description in links or tags, but I will fix this soonTM).
One cool feature is to set your browser homepage to the url that loads a random page : each day I get a cool article to read/concept to discover! (here's my own instance random link url: https://links.l3m.in/en/random/)
That's my useless side project (because shaarli already exist and it's way more mature).
The almost is the interesting part here, since it's not "never" anymore :P
Another goal is to be able to retrieve links with ease. I haven't added all the tags on my links yet since I lack time (1800+ right now and counting!), but I should be able to retrieve this cool article about Python decorators, or the article from the cheapskatesguide.org that inform us that bluedwarf.top is launched :P
I really hope that one day there will be multiple share-links instances that will allow people to find interesting links, and then when they thought that they have found all the cool content they discover that there's more websites containing hundreds of fresh or old links leading to even more interesting content!
If websites were downloadable like a (versioned) document you could create your own offline bookmark web (archive), which would be neat. I guess scraping could help with getting the contents to disk, but it would probably be kinda broken.
If you can then direct your search engine to discover content from this archive I would perhaps more often (re)stumble upon this neat discontinued blog that's stuck 5 levels deep in my folder structure.
I feel seen. I have bookmarks folders named “To Read $NUMBER” from 10 years ago. Why do they end by a number? Because I create a new folder every time the previous one feels overwhelming.
From time to time (~6 months) I review all of them. Quickly delete most, make effort to process the rest and save the 'few' that I will not take an immediate action on a topical markdown note.
e.g. Let's say that I find a book interesting. I go to '~/fun/read/read.md' paste the URL, delete the bookmark and forget about it.
Oh boy... I'm the king of useless projects. From time to time I get an idea, and develop this whole idea to an mvp or fully complete it and then I'm to dumb to sell. I don't know why I always have to develop my shitty ideas.
There's a typo on one-folder.com/en/home/: "With one folder are your worries about your documents gone!" -> "With one folder aLL your worries about your documents are gone!"
I create monthly mixtapes of leftfield house, techno and disco, and spend an insane amount of time digging through recent digital releases and then mixing/uploading to YouTube (now working on adding visuals with Disco Diffusion). What blows my mind about this whole process is there is so much new music published daily…
In an upcoming mixtape the intro track has been on YouTube for 3 weeks and has received only 3 plays. I want this project to unearth such classic electronic tracks and get them out to a wider audience. I can’t see myself ever stopping this project or ever regretting the time spent.
You just got yourself a new subscriber, cause I’m trying to get back into music curation and DJing like I used to when I was in college, and this is right up my alley as far as genre and finding fresh tracks!
Most of my work is stuff that I feel very few people find interesting.
Yet, I treat each project as if it is a Fortune 50 corporate initiative; with ultra-high-quality code, reams of documentation, full test suites and/or harnesses, etc.
I do it, because the occasional project that I do, that matters, benefits from the habits I establish. I've been working on one of those, for the last couple of years (closed source, so I can't link). It's coming along great.
Oddly enough my least valuable project was my most successful (so far, I hope!). It was a tinker project while I was learning Laravel, no target market at all.
It was very specific - a database/API of YouTube videos and metadata for a YouTube group, the Yogscast.
It ended up being used as the backend data for a viewer-controlled cinema that let users vote on and pick videos to watch during livestream downtime to keep the audience engaged during a charity drive.
Can't take much credit, of course the actual video playing code was the meat and potatoes of the operation during downtime and the actual stream content during the daytime, but it was vaguely involved in raising millions for charity haha
Regrets none, only reason it's not still about is that it turns out YouTube don't like you making a database of YT videos that you allow others to access, they killed my API key eventually
I would imagine they call the YouTube API directly now. The reason it didn't in the first place is it went from random idea to implemented in about 12 hours originally (h/t Akhawais!)
I have a couple apps that only I use or are used by a handful of friends. For hobby projects, I never regret building things that I think are useful but I can't be assed to try to market them to others.
1. Mototripper - live streams my location when I'm on a long distance adventure motorcycle trip so my kid knows I'm still alive and moving.
(e.g. <https://www.mototripper.app/track/~knobbies> - a trip through Finland, Sweden, and Norway I took a couple weeks ago.) Built with Sveltekit.
2. Vatinator - an app to apply OCR to Estonian receipts to claim VAT reimbursement. Built with NextJS.
The UX is pretty terrible, honestly. But I thought I could get away with it since I didn't expect anyone to sign up.
I also built https://sudokurace.io based on an idea my wife and I shared while talking. We talked about it for years and I finally decided to build it, only for us to play it for a day or two and now we never really touch it.
This one is one of mine from way back in the day, a 6502-based ternary computer emulator with an assembler, a simple operating system, and a compiler for a dialect of C adapted to base 3: http://tunguska.sourceforge.net/
It had a user base of like 3-4 russian ternary computing enthusiasts, dunno if it still even compiles. Fun project though.
As with all of my projects (even those that bring value to others), I do them for myself so as long as I find value in them I'll keep putting time on them. No regrets.
I kinda find my writing funny, very mathematically rigorous and ambitious. http://tunguska.sourceforge.net/docs.html
Not really sure who I thought the audience would be.
You can make it into a nice product. Just make the state save-able (including comments) and share-able by a link. Would be a good tool for designers & their customers to quickly leave some comments.
Cool site, nice one. There's potential to save a lot of time when going back and forth between cropping tools. One of my usual workflows is reading big PDFs and I want to pick out the parts that are relevant to me. I could use a highlighter, but that wouldn't filter out the rest of the document, so I typically use a screenshot tool to grab specific areas and save them as pictures. But it can only make one screenshot at a time, which is a lot of overhead.
It would be cool to enhance this site so that it takes a PDF with multiple pages and allow you to do the same for that. But that's probably a lot of work compared to uploading a single picture file!
I am still working on my never to be finished at this rate dream 4x-god-sim. The mindset to have with projects like that is entertainment. I enjoy the process. I could be reading sci fi or watching TV or tinkering in the shed or playing a sport or playing with a dog or raising children or whatever instead; those are also useless passtimes that are done for their own sake. I choose this one (among others), why not?
I also learned a lot of new tech from such useless projects in the past, but that is just a nice little extra, just like with the other ones
My girlfriend sells self-made clothes online. In 2015 I made her a a small Tapestry 5 webapp to manage her orders and, this was the crucial feature, to determine payment status from her copy'n'paste online banking's transactions list.
Today, it has integrations for Paypal, Etsy, Shopify, DHL and HBCI (no more copy and paste, finally) and basically covers the whole order cycle with minimal fuss and gives her a quick way to publish new products which she does a lot as fabrics come in / get sold out.
I guess that's not that useless but of course it simply can not be worth the time I spent on it.
Yet, I don't regret doing it in the least because for me it literally is a labor of love. Plus, I learned a lot doing it and I can go down crazy paths like this one time I made an android app to cut out the background of her product photos. Didn't turn out to be useful in the end but eventually I did use OpenCV via JavaCV for something else in the software.
Currently I am in the process of rewriting it in Kotlin / Multiplatform (JVM+JS) and again, I'm learning a lot and making the software a lot sleeker and more functional. That's not going to be worth it either, I am sure.
Anyway, I wish I had day job were I was in constant contact with users to built them something of tangible utility and value.
My timeline thing. It gathers all my crap and puts in onto a timeline. It's a more fine-grained version of scrolling to a specific date on my photo stream.
I had no expectation that anyone else would ever use it, but it suited me just fine, and I used it every day for years, both at work and on my other personal projects. It's still one of the first things I install on a new machine.
I have been reverse engineering a 25-year-old proprietary (PlayStation 1 based) system used in a number of well-known arcade games for almost a year, digging into every single detail and uncovering what even MAME developers ignored.
The end purpose was initially to enable homebrew game development but, with the prices these boxes are fetching on the used market, I know nobody is ever going to bother. Moreover, as far as reverse engineering existing games goes, the most interesting parts (i.e. the copy protection) have already been figured out two decades ago and most of the games are perfectly playable through emulation. Still, reverse engineering an unknown platform that isn't too serious about security is fun and rewarding.
Ah, the "hall of shame" HN thead I have been waiting for! Too many to recount here, but the one I probably sunk in most time was fauxjsp [1], a dev-friendly reimplementation of java server pages. I don't regret spending time on the project because it was fun, it was time I couldn't/wouldn't have spent on anyting better and I learned a bit about software architecture.
My local newspaper's online version for years was free to use. Of course, they then added a paywall which kicks in after only two "free" articles per month.
Given that I refuse to pay the $3 per month for access due to strong disagreements that I have, politically, with ownership of the newspaper, I decided to build a "paywall bypass" solution as a side project. The solution involved building a small server app that acts as a proxy between my browser and their site. It parses out the HTML representation of the articles, which their paywall script at runtime first obfuscates, then removes from the DOM. This solution, naturally, also avoids my browser loading and requesting a whole collection of third-party trackers and ads - since I'm not "rendering" their entire site, simply scraping relevant portions of the HTML.
The kicker: I pay $5 a month to host this on a DigitalOcean droplet.
It was shockingly easy and straightforward. And in three years of running it, I have yet to need to update it in reaction to an implementation or page structure change on their end.
I built Pegao[1] as an open source side project, which is a web aggregator or bookmark to group links into lists trying to solve my own problem: I had many tabs open on my phone. It didn't work, but I actually use it to remember some links or tools that I know I'll use for upcoming projects.
I have an off-and-on side project for fun. I have invested a ridiculous amount of time and some considerable money in my yard and my house just because those things make it much more pleasant to be here. I fiddle with cooking and we all eat the results but clearly I don't have marketable skills. I buy and play friggin' games . . . what a waste!
Why would I regret the things I do for enjoyment or better quality of life? For myself and others?
I actually haven't started this one yet, but I am planning to write a video game for the IBM 1401 computer, released in 1959. What complicates this project is the fact that the computer uses punch cards for input and a printer for output, making the job of designing a video game for those inputs/outputs a bit more difficult. I will probably start with implementing a blackjack game and then work my way from there to a bit more complex games.
Not exactly useless but I haven't updated it for a long time. The project is called react-qml[0]. Basically it allows you to develop Desktop application using modern JavaScript/TypeScript and/or Qt/QML.
This project helped one of my consulting client 5x their development velocity.
I had been working at GitLab, pre-pandemic, for several years and I saw how writing things down was almost like a super power to enable async work. If you start with time zone distributed teams, writing things down in issues/docs just becomes the natural way of working. I also saw that lots of other companies didn't really get it - and there was a leap of faith required to try it, because it didn't logically follow that if you write things down more you can have less meetings.
My idea was to build something that provided a really natural place to write things down, and I built and tried to sell AsyncGo as a place for making decisions in a written way. You'd set a topic, a context, and a due date, and then the magic would happen. In theory. The problem I had was that I couldn't find anyone to take the leap and try it. Companies who were interested in async already had some similar process, and the ones who really needed help writing things down didn't get it and I never found a way to communicate it to them clearly.
In the end I shut down the hosted version and put an MIT license on it. I don't regret it exactly, I learned a lot making it, but I wish it had helped more people. There's other stuff out there now that's sort of similar, and it seems they are struggling a bit as well, so I don't think the market was really there (yet). Now I'm working on a collaboration app for regular people/teams who want to get better at making videos/streams (https://www.synura.com) and I hope to apply a lot of the lessons there.
I have plenty of ultra niche projects and I never regretted working on any of them. On the contrary - i use them daily and it brings me much joy. Here are three:
https://github.com/bjesus/muxile lets me continue my tmux session on the phone, bridging the two over WebSockets. How many people use tmux extensively AND want to continue on the phone? Not much i guess...
https://github.com/bjesus/callibella is my way to sync my personal calendar to my work calendar without revealing my personal entries. It's very useful for me but less needed if your personal calendar is Google because i heard they have their own integration.
https://github.com/bjesus/air is my AwesomeWM based Interface to my PostmarketOS Kobo e-reader. Linux on your e-reader isn't a huge market share to begin with...
I spent a year and a half writing an economic history book that I knew was unlikely to be profitable. Nevertheless, it was a very enjoyable experience and I learned quite a bit.
Generating e-books from public sources[0]. I wanted to read the “Stormlight Archive reread” from Tor.com on my Kindle so I generated EPUBs. It kept going forward from there to almost a dozen books now. The most popular is the Google SRE book, but I also wrote code for JK Rowling’s Ickabog, which was published online (and taken down since - I switched to Internet Archive).
Needless to say, I’ve barely read most of the books that I’ve compiled, so it sometimes feels like wasted effort.
However, I’ve since put a lot of the learnings from these projects to use, such as while building pystitcher[1].
Not quite useless but more like a one shot use:
I spent some days building a flight/train route optimizer for traveling to N cities, staying at least M days while minimizing travel costs for a single journey in 2018. Was fun to build and we payed like 200 € for one month of traveling in Europe :)
I made a webapp to learn music sight reading [0]. Initially the goal was to help my kids pass conservatoirs' exams, which can be quite demanding, and are usually studied on books. That first goal was reached, my kids passed (this week)!
I thought this would interest music teachers, and I implemented specific deatures for them. -- Typically, music students like/love learning to play their instrument (or else they quit), but hate sight reading.
I talked to a couple of teachers: they are polite, don't express objections, but don't engage, don't call back and don't use it.
I did a ShowHN that gathered 11 upvotes. From that came a couple of users, including one who seems extremely motivated.
So all in all this has one serious user. Better than nothing! I certainly don't regret it.
I deployed a Personal Handyphone Systenm (PHS)[1] network in my home complete with vintage Japanese mobile handsets.
I have a few PHS data cards I can use with vintage computers, and a few PHS telephones that are hooked up to a VoIP line so I can use them as a cordless phone.
My favorite device I own is the DoCoMo Eggy, which I helped showcase in a recent video from Cathode Ray Dude[2].
Allows anyone to create geocaching-like adventures controlled by server-side Lua scripting. Neat features are multiplayer support, some included widgets to quickly build a UI (see[1]) and a time travel debugger to fix errors. I knew there was probably no real way to make money from that, but it was fun to build. Unfortunately reception by the big geocaching portals was mostly to no allow linking to it IIRC, so there was also no way to gain any traction at all. ¯\_(ツ)_/¯
That's... that's all I do it seems. Just like everyone else here, I create MVPs that don't go anywhere. My fallback was for my useless side projects to become portfolio pieces, but employers don't look at developer portfolios. After 10+ interviews this year not a single potential employer has visited any of my sites or apps like https://pollhub.xyz, a fun site, or https://didoe.org, a way to track who you boycott.
The hardware is a very fancy synth with a bunch of oscillators that are all tuned relative & symmetrically to each other and you can control the “spread” of the tunings at the same time you adjust the core pitch. Think of the THX sound, or the entire The Social Network soundtrack.
Actually haven’t built it in a while, should redo it in the newest swift(ui) as an exercise.
Edit: no regrets. Nobody cares but me & that’s fine.
I've spent way too much time implementing a transpiler and bash like shell that makes programming my ships in Kerbal Space Program easier. The self-hosted version is almost done. I'd probably do it again though, way too much fun to see your spaceships autonomously do things.
Please share the bot address! It's helpful to extract the videos.
Why not have the bot to parse links automatically from a Twitter account (or TikTik) to auto post in a channel?
A literal computational environment & DSL for interaction with DOM elements.
I don't regret it, because I explored a lot in the process: DOM, Canvas & WebExtensions APIs, parser generators, some JS tooling, gained general code organization skills.
I got fed up with ansible and pyinfra and created a new tool [1] to fix some of the issues I've had. I spent weeks on documentation, yet it is kind of useless. There simply are industry proven tools for the same thing already in existence - so I'm the only one using it.
A Twitter bot that posts images of macromolecular structures (i.e. proteins and nucleic acids) every thirty minutes. I spent an embarrasing amount of time tweaking the coloring and orientiation of the renderings, but I like the results: https://twitter.com/molecularblobs
Wordle helper that I threw together over a few weekends. Stick in the day’s solution and your guesses and it’ll tell you how many valid words remain after each guess.
My first stab at anything webdev after years of embedded development. Definitely worth it, LiveView is neat
I will soon find out if my inbox.for.your.domain-as-a-service has any takers once I seriously start marketing it. If it finds a market, it’s not useless but if it doesn’t, that’s a couple of years of working off and on on the idea down the tube.
A service like google maps that shows all farm ground (fields, meadows) as colored hexagons representing the crop health (NDWI), updated daily/weekly. Resolution is roughly 50sqm per hex, test case was all Germany.
Was ultra hard getting it into farmers hands but ultra fun to code. Based on satellite data
I recently rewrote it to learn svelte. No joking, I prob refresh it 10+ times/day since when I built it back in 2016. And I'm pretty confident I'm the only user.
A desktop application to learn singing. It inputs your voice from mic and shows spectrogram realtime and shows note you are singing on the graph. Also provides reference note sound on headphone so that you can sing the note accurately. Also provides random notes variations to practice.
Its currently linux only, are you a linux person? I have started porting to windows recently. Once thats done, I wanted to sell it, not sure how many would be interested.
Not much context here, but built https://daborsneeze.com/ after a random interaction gave me the idea at a conference. Adding images slowly over time but there's really no ultimate goal there.
I built a little system that measures many of my systems and alerts me of anomalies or systems approaching limits via telegram. Kinda dumb in all, so very customized but saves me time and effort.
I regret it, I feel like I should have developed something more profitable.
Mine would be a cron for "buying" free Kindle books for specific languages that I can read. Now I have more than 100K useless books in my Amazon account that I'm never going to read.
I’m spending way too much money and time making light up skis for next winter. Bluetooth control from my phone, gyroscope to make them react to my carving, etc.
A couple of years ago I started architecting transcoding/streaming video pipelines on AWS for my work. It sparked a general interest/curiosity into the internal workings of video compression algorithms, which then allowed me to revisit a subject I've always wanted to try out, datamoshing. Something about purposefully breaking codecs to generate artifacts that lets you peak at their inner workings really nerd sniped me.
The most advanced video I was able to create is this datamoshed version of Carpenter Brut - Turbo Killer clip : https://streamable.com/7d9h4b. I'm particularly proud/pleased with the transition at the 2:00 minute mark.
It still a long way from what I would want to be able to accomplish (as the moshing is still pretty random and looks good only by "chance"), but I'm pretty happy with my progress overall.
My long term goal for it would be able to create moshing of music video clips, and sync the glitching to the beat/structure of the music (using Spotify Audio Analysis feature [0][1]).
As a sidenote, a tool I discovered and has been a godsend into tinkering with video codecs is FFglitch[2]. It's a fork of FFmpeg with the ability to directly access and modify internal MPEG2/4 values (such as the Motion Vectors, macroblocks, DCT coefficient, etc). Compared to other datamoshing techniques (such as corrupting AVI files with random data, manual key frames deletion, etc) it's order of magnitudes more advanced and precise, allowing you to precisely and purposefully influence the codec, instead of randomly breaking it and seeing what sticks. However its downside is that you have to have a solid grasp of the inner workings of those codecs to achieve anything.
Finally, for anyone looking for a good resource to get an understanding of Video Encoding, I highly recommend this[3] github repo as a starting point. It contains a good description of the workings, as well as an amazing references list.
I've spent the last four years now trying to start a business on the Internet (working on ideas roughly two hours per day before work).
Each project had valuable lessons (both business and code/architecture) that carried over into the next one. So far I've built:
- bill splitting service (2017)
I spent days agonising over which new framework to use (this was late 2017 after all), eventually learned to just use the tools I knew (react, node, postgres). I didn't really care about the problem space, so shut it down.
- jobs aggregator (2017/2018)
I learned that implementation is meaningless to the user if you're not delivering value. I wanted to copy remoteok.io and make it serverless (effectively free to serve traffic), I didn't realise existing sites provided value via their traffic. The reason StackOverflow can charge as much as it did is the millions of page views per month it receives, creating value for its job posters.
- appointment scheduler (2018)
I built an appointment scheduler, had no real means of attracting users, shut it down.
- room booking service (2018)
Spin-off of the previous idea, but for meeting rooms. Tried to build the whole thing using Google APIs, eventually got stung by API limitations, gave up (learned not to rely on other's APIs without understanding their limits first).
- graphql API monitoring service (2018/2019)
Traction again, couldn't find users (tried in-person sales for the first time, too).
- site speed monitoring service (2019/2020)
Essentially running google lighthouse as a service. had some users, but fixing all the edge cases around chrome/puppeteer/lighthouse across super slow websites was a total pain.
- uptime monitoring service (2021-current)
Doesn't seem to be as useless as the other projects. Has bought me the MacBook Air M1 I'm typing this comment on now.
Rewrote my old graphql API monitoring service from scratch to monitor APIs, websites and web apps, seems to be going well so far. Planning on adding features for incident management. Content marketing/word of mouth brings in the users.
If you want the long story across several articles:
I trained an ML model on 15K headlines from the local (Dutch) populist newsletter and use that model to tweet a new headline every 30 minutes, for the past three years. To make it look like tweets from the actual newsletter, I take the first few words, run those through google translate, and then use that to find an image on Unsplash to add to the tweet.
The headlines frequently make no sense and often have grammatical issues. Yet a couple of times a month someone (often a boomer) interacts with the tweets.
Naturally, I then released these two albums to the world, where they have sold roughly 0 copies. This doesn't bother me, though; that was never the point. It is, in essence, a long-term elaborate "performance art" piece, that is intended for my own amusement, and possibly my daughter's amusement, once she gets old enough to "get" the joke. In the immediate time, I get value out of seeing her enjoy playing instruments and making sounds. And I'm giving her the experience of making music, in whatever way she sees fit, before she ever gets taught a bunch of nonsense about what ways are OK and not OK to make music through more formal training.
Any project that naturally holds your interest, is probably worth pursuing, regardless of the overall value to humanity. At least for a little bit. Because you're generally going to walk away with knowledge you didn't have prior to starting it, and benefits you wouldn't have predicted had you not gone down the path at all.
Oh yeah, also; don't take advice from me. I make horrible choices, then I commit to them.