Since we have an article about having fun by creating at #1 right now, I thought this would be a good time to ask: What cool project are you tinkering with this week? Please limit it to things that aren't seeking profit.
It's a a simple HTTP-based pub-sub notification service. It allows you to send notifications to your phone or desktop via scripts from any computer, entirely without signup or cost. It's also open source if you want to run your own.
ntfy looks great! I love these "do one small thing right" tools in the OSS ecosystem!
A few months back, I was looking for the opposite. A way to bring notifications from my phone onto my laptop. I tend to usually keep my phone in a different room during the workday, and have to keep jumping in everytime I need something like a SMS 2FA or to just check if I have any new notifications. Whatsapp Web+TOTP take a lot of the pain away, but it would still be great to have a way to seamlessly get notifs from phone (Android if it makes building it easier) to laptop (Ideally cross-platform, but Linux). Any idea of any tools that do that?
I just helped a friend connect his iPhone availability polling script to ntfy.sh so it could be run remotely and still notify him when the latest iPhone was available at a nearby Apple store. It worked wonderfully, and now my friend has a new iPhone!
Thanks for making and maintaining such an amazing tool!
Thanks for the kind words, especially since your friend is using the iOS app, and that's super buggy and neglected. I wish I had more experience with iOS development (especially the Core Data part) to fix those annoying bugs that make messages not show up in the app after they arrive. One of these days I'll spent more time with iOS again.
If ntfy.sh is used (not your selfhosted server), then messages are also forwarded to FCM and APNS (unless the X-Firebase header is set to "no"). For selfhosted servers, Firebase/APNS are not used. Since Firebase is pretty slow in delivering messages instantly, the Android app also has an "instant delivery" option which keeps a JSON stream or WebSocket connection open at all times.
With Firebase, the reported battery consumption is 0% obviously. Without Firebase (i.e. with the instant delivery option) it varies depending on phone and usage. On my phone it is usually around 0-1%. Others that use it to deliver their Matrix chat messages have reported higher numbers up to 12% (though that is rare).
If you "allow background services" for the app, Android will kill it less often. Though some manufacturers are quite aggressive even with that option.
As for who pays the bills: The public server is run in EC2 ($25/month, though it could run on a much smaller instance), and the Apple developer license is $100/year. Other than that, there are no costs. Up until a month ago, i was carrying all the costs. I have recently started accepting donations, and now all the costs are carried by my awesome sponsors (see https://github.com/sponsors/binwiederhier)
I'm interested in making a "chat with yourself" program I'm calling Soliloquy [1].
The idea is to let you explore different sides of an argument, or different sides of your psyche. For example, you might choose the three characters, "optimist", "pessimist" and "judge" to hold a group conversation that looks like a mobile phone chat, as a way to work through a difficult challenge in life that needs reflection and long-form thought.
The reason I'm building this is I find that I often don't complete a thought before negating myself--I cut short embarrassing or superficially trivial feelings, but I believe they sometimes deserve more stage time. Sometimes it's exploring thoughts that relate to uncomfortable feelings that yield the highest return on time spent.
Soliloquy is being built with neutralinojs, so it will work on any desktop OS (Mac/Linux/Window). It is "local only"--no network connection, so you can rest assured your private conversations are your own. I intend to publish it with MIT license but haven't got around to that yet :)
I would be interested in something where I'm given a topic and position to argue for, and I argue against another person who has the opposite side, and then other people vote on who did it best. Obviously a lot of details there to work out to avoid sybil attacks but I think that would be stimulating.
I think you're really on to something there. It's quite along the lines of something I've been practicing the last 18 months or so, particularly not stopping short when uncomfortable feelings come up as I examine an issue or feeling and try gain a proper perspective. It's something I consider part of my own mindfulness techniques, which have really been powerful and profound mental exercises for me. I think something like Soliloquy could really not only help people with the intended goal of the project but even to just expose more people to the very idea of truly exploring things from different vantage points.
How cool, I’ve been wanting to try just such a conversation between two of my characters. I might give this a try!
That conversation will relate to my own project, in which I’ve lost the fun aspect of working on it. There’s something important about doing a thing purely for fun.
https://letsblock.it has been on a slow but steady growth trajectory since it's January ShowHN[1]. The public instance served 290 customized lists this week, and operations / project upkeep are a pretty light time commitment.
It's a companion project to uBlockOrigin, working towards improving the internet's signal/noise ratio by filtering out low-value content and nags. While cosmetic filtering is very powerful, its learning curve is steep, so the project provides customizable templates maintained by the community.
I'll be giving a talk about the project, titled "Designing an Open-Source project for low maintenance" this Thursday at Ory Summit[2] and will share the recording on HN once it's public. You can still watch the conference's live-stream for free!
After that, I'll need to update the website's copy and simplify some UX, to prepare for another visibility push later this year.
I'm interested to understand why you find this provocative.
My ambition is to build a dependable service that will be up for several years to come. For that to happen, I need to limit myself to a scope I am comfortable maintaining in the long run, instead of burning out under scope creep.
Until the maintainer team grows, I would be on the hook for updates and fixes, and I refuse to create an account on any of that company's sites.
FYI, I have such a call to maintainer [1] for a "Twitter cleanup" template similar to the "Google search cleanup" one [2]. There is definitely a need, but it seems over my current carrying capacity.
For Meta's properties, I would wait for the maintainer to have a track record of contributions before being comfortable opening that door.
Site is intended to be like the book Hacker's Delight, but recast into a game.
Or maybe like professional programming, where you're mostly trying to understand/modify other people's code.
Or maybe like programming in a post-GPT3 world where you're checking/fixing a transformer language model's plagiarized/regurgitated code. Our dystopian future.
Later this week I'll add a Hash Treap puzzle (the fastest and simplest balanced binary tree) following up on the reroot and remove-root puzzles (amazing little algorithms that allow treap insertion and removal, top-down, no rotations).
I'll launch the site properly once I have enough puzzles, maybe early next year.
It's a personal project, but completely free and without ads, (in the future I want to open source parts of it!).
The idea is to solve the problem that boards have out there that just puts everything as "worldwide" and in the end, it's not available for me as a Brazilian.
The time spent researching job boards it was easier to make a crawler. In the end I got a job.
Right now just want to focus on it being complete as possible, so there are a few bugs and some listing that slips.
It's a free tool to read & review research papers together with your colleagues. You can annotate, draw and comment on papers you upload. It's a tool that I wish I had when I was doing my research and super excited to finally work on it!
Please don't follow what Polar did to their project. I hadn't used it for a year or so, went back to use it, they had upgraded to a SaaS model (online account required) so I couldn't access my highlighted notes. I swear they were open source from the get-go, but the community has seemed to die down now.
Piano Gym is a learning and practice ecosystem focused on prioritizing music theory and performance skills acquisition through the use of flash cards. We use flash cards in order to pair them with modern learning techniques like spaced repetition, graded feedback, and progress tracking so that you can practice material and work through content that is managed by Piano Gym, and all you have to do is enroll in a school/course/lesson and do your reps! Just show up every day and do 15 minutes of reviews. You're going to make progress.
The website uses the Piano to navigate exercises as well as regular keyboard/mouse input. It works on browser technology and I'm looking to eventually make it mobile devices.
It provides content creation for everyone so that anyone can make their own schools/courses/lessons and the best part is each school gets its own landing page.
My goal is to do this for free. I believe that no one should be blocked from learning. And one of the issues with this at the moment is that it's just me currently working nights and weekends to make this happen.
This looks incredible and very similar to an idea that I was kicking around a while back. Please email me (see profile). I'd love to contribute to this.
In particular, something that would widen access to this platform is a microphone-based input method for acoustic pianos. There are only a couple of commercial platforms that have this and AFAIK, no completely free ones have it. There are some challenges of course -- out of tune pianos, different types and makes and models sound different, variations in how the piano is mic'd, etc., but I believe those can be worked around and have some ideas about how to make it happen.
I spent some of my free time learning how to work with Emscripten. So far, as an output, I have tiny sample files that can be used to start new projects:
Stictionary[1] is my free, no-ads, offline dictionary that remembers the words you look up. It gives words of the day and flash cards for words you have looked up before.
I built it after years of manually tracking words i looked up in hopes that i'd retain them better. After having it on my phone for months, i can say it works great.
I'm a C++ dev, but wanted to learn js/typescript/tailwind in case I'll want to switch job/stack.
Started with the simplest project possible that I wanted to do https://www.jjson.me - just a json formatter as a static website hosted on Github & Cloudflare
Learned to tinker div blocks, callbacks, css classes)))
Now planning to create something simple but with a server and database involved, hope it'll work out
I'm working on the second part of my Nerf Dart Missile Defense system project [0].
Building a robot that can track nerf darts and shoot them out of the air has a lot of interesting technical challenges so it's a fun project :) I also get to learn a lot about the process of making videos.
The second part was almost ready a few months ago, but then I had to redo a lot of stuff and I lost steam for a bit.
Hopefully I'll have a second (more well-put-together) video out soon!
I am publishing a newsletter about Ruby https://newsletter.shortruby.com, and I am trying to automate part of my process to focus more on curating the content.
My process looks like this:
1. I have a list of tweets and Reddit posts that I gather during the week. This list is sending self-DMs on Twitter or Telegram
2. I have a Ruby on Rails app that will read the tweets and posts and will save them in the local DB
3. I browse the list of things, and I have a simple button, "Hide" that I press to remove what I don't want to include
4. Now comes the more manual part
4.1 I click to open each Tweet (mostly) or Reddit post and then I take a screenshot of that Tweet
4.2 Then I transform the screenshot (like resize) to fit the Substack (my newsletter is hosted there)
4.3 Then I drag and drop the screenshot in Substack, and I add a caption (where I specify the source) and alt for the image
So far, I managed to write an apple script that automated 4.2, but I am trying to find a way to take the screenshot automatically so I am playing a bit with headless browsers.
Enjoy the time with your little one. They’ll soon grow up before you know it and then you’ll get time back for tinkering, but by then will miss the time you had with your first born.
1) I used to be an energy engineer, so I'm taking some of the most common energy saving calculations I used to do and making a python library of them. And then maybe a quick flask calculator to show the energy saving methods and how much energy they save
2) Trying to learn the basics of font making so I can make a font based on the golden ratio, mostly for personal use
Baby Buddy[0] just turned five and I’m still hacking away on it! Well mostly reviewing PRs and helping contributors. I get a great deal of enjoyment out of interacting with its users and contributors.
I'm working on improving the https://UnifiedPush.org explainer with more fancy CSS animations this weekend. Also improving the ntfy integration in matrix-docker-ansible-deploy (an ansible playbook to easily host a Matrix server).
Currently trying to work out why glyph heights are the same on all platforms except MacOS in LibreOffice.
I suspect I will be writing a utility to look at each font's glyphs and metrics. Funnily enough, MacOS is literally off by 1 pixel from every other platform in terms of glyph bounding box height.
I’ve been having tons of fun messing around using the depth sensor data from a Kinect for Xbox to create procedural visuals.
It’s been growing and growing to the point of now having 4 or 5 different input methods and accidentally discovering new emergent effects based on different combinations of parameters.
Also threw in intel’s 3D rendering engine Embree because why not.
Anyways, this playlist documents the progress in chronological order...
I rebuilt the iOS version of OneBusAway, an open source, real-time bus tracking app used heavily in Seattle, Washington DC, Tampa, and San Diego, entirely in Swift between 2018 and 2021. I have a few bug fixes I'm hoping to take care of for it this week: https://github.com/onebusAway/onebusaway-ios/
(n.b. I'm the current maintainer of the iOS app, and also a member of the board of the non-profit that technically owns OneBusAway.)
This week, as most weeks, I split my time outside of the day job and my other hobbies and obligations between a few projects:
https://github.com/bri3d/VW_Flash - Flashing tools for select control modules in VW MQB and now PQ35 platform cars. This week I'm working on old stuff: a simpler exploit chain for older Simos ECUs, as well as tweaks to expand support to older DSG control units used in PQ35 platform vehicles.
https://github.com/fpv-wtf/msp-osd - I pushed a rearchitect of this on-screen-display overlay system for DJI FPV Goggles last week that seems to have sorted out a lot of issues - I switched from just passing through the OSD drawing messages from the Flight Controller to a system where the video transmitter maintains the OSD character buffer and sends a compressed representation of the screen state. This makes the system much more robust to packet loss in situations where the Flight Controller sends delta updates rather than frame-at-a-time.
I only really started publishing Open Source projects a year or two ago, and while they're pretty much my worst code by any objective measure, I've met some great people and really enjoy working on these. It's fun making things that achieve a goal without so much pressure of deadlines, stakeholders, and competing priorities.
I've been doing a ton more work on Radiant -> https://getradiant.app/ this week (Takes Spotify and turns it into a personal live Radio station complete with an artificial DJ).
Currently adding more hyperlocal live music capabilities related to your music preferences to the presenter's (Rad) skillset.
It's not open source but I'm very transparent about what stack it's using and how it works if anyone has any questions about it!
It's impressive, I tried in French (my native tongue) and it seems "it" understand I was mocking it, then I tried in English on a serious topic and 'it" didn't understand what I was saying (talking about ALS and a protein) as if "it" were a layman, non specialist of the topic.
I'm working on fine tuning a very fast online search algorithm I invented, provisionally called HashChain.
It's already faster than all other algorithms for most data, pattern lengths and alphabets, but it also has a lot of different ways to parameterise it. So I'm exploring the parameter space, so the algorithm can auto tune itself.
Not sure if anyone really cares about this sort of thing anymore. Most people still seem to think Boyer Moore is the best, but that is positively ancient and long superceded by others.
It's not public yet. I intend to write a paper on it and hopefully get it published in some relevant journal. I'm not an academic though, so no idea how to go about that!
I will also publish code and the algorithm on the web somewhere, once I've nailed down the parameterisation.
I needed to refresh my skills in Go, so I wrote a retro lunar lander side-scroller game (using ebitengine). Still a work in progress, but it has been a refreshing change of pace from my day job as an embedded/firmware developer. I haven’t written a proper game in decades.
Outside, I am putting the finishing touches on several new ponds, each around an acre in size and 7-10 feet deep. That was a huge multi-year undertaking, but I hope to stock them with fish next year.
A database with chemical properties of compounds and mixtures.
It is a non profit side project of my company. I created it because I was not able to get access to such data during my PhD.
I used it as a playground to learn assembly, CI/CD, etc.
For me, this is yhe best part of the non-profit, you do not try to make money, this is a license to experiment, try new stuff and do the right way without time constraints.
I'm creating yet another recipe hosting website https://faluszelen.hu (just in Hungarian).
I'm thinking of it as a replacement for my physical recipe notebook. Loads insanely fast, have no tracking, ads, story or anything distracting, just the ingredients, steps and a way to easily edit it on the fly.
Also it's free to host, except for the domain name fee.
I've been working on a custom ML pipeline to create novel Gameboy music by training on existing songs through the CPU instructions that produced them.
The project involved writing a custom emulator (Rust) and an awful lot of data cleaning and preparation software, with the ML training and inference written in Python.
Setting up and playing around with Omeka, a brilliant document publishing system, to help publish an archive of digitised physical books and documents.
I am writing spooky short stories and submitting them to various publications. The publications technically pay, but they pay so little (far less than minimum wage) that it’s more about feeling valued than about the money.
Also, if I were offered a full time writing job with good pay, I would decline: writing is how I escape my work, and if it became my work, I would lose it as a fun and critical hobby
That argument is why I initially didn't become a software engineer. I was doing business thingy jobs because writing code was what I loved and i didn't want to ruin it.
But then I realized doing what I love and getting paid for it is even better. Have been doing that for many years and never regretted it.
That said, it's a personal choice and different for everyone, so what was right for me doesn't need to be right for you.
Btw I love how random this hobby of yours is. Usually all you see here is tech projects.
I used to do game design as my hobby. Now I do game design as part of my job (programming and design), and that’s fun, but not as fun as it was before. Now, peoples lives are affected by whether I do a good job now (Eg, bad design -> failed game -> my friends at work lose their jobs). I also have to do it when I’m not in the mood as well, with strict deadlines.
I don’t want that to happen to my writing, which is purely enjoyable right now :)
I spent the weekend writing some code to generate pdf templates to send to my laser cutter for making custom boxes for minis. Downside - someone saw them and wanted some too and offered to pay. Upside - no way it generates profit once I account for dev time, material costs, and the cost of the laser cutter.
Other than that - working on some designs for a few small robots. One combat related the other education.
Been slowly making progress on my 6502 homebrew computer, with things like a keyboard I took out of an old Tandy word processor, reading its keyboard matrix with an ATMega, and sending ASCII to the 6502. Writing an OS for it in assembly, with an 8 bit FAT file system (using serial EEPROMs for storage) has also been a very rewarding learning experience.
I got a 3d printer this weekend. I'm starting by just printing DnD minis, but I've got schemes to use it to make a cleared game of my own design. Something almost like, chess crossed with Magic the Gathering.
For years now I've kinda just dismissed 3d printing - it was cool, sure, and had it's uses. But it always seemed like something that I'd be looking for a real reason to use it but never find one. Then I learned about resin printers, and how much more fine detailed those are, and it all came together. I've had it running basically non stop since I've got it. I'm first just starting by printing entirely too many minis that I will probably never need. It's been a real delight to just watch this little army of characters come out of the goo like uruk hai.
I'm not in it too make money, I just want to know if I can at this point. It'll probably up on the pile of failed game prototypes, but at least I'm having fun.
You never know where something will take you. I’ve had a 3d printer for the past 5 years, but connected a few dots earlier this year and have been paying all my bills with a small business built around a very niche item I design and sell.
Recently bought a used 10 year old MacMini to use as a home server behind Tailscale. So now writing small apps here and there for myself. Latest is bland, it's like del.icio.us but you can host it yourself: https://github.com/valueof/bland.
If you want to FOSS it up as an alternative to Tailscale, I’ve had pretty good experiences with Netmaker. Though Headscale seems to also be gaining traction as well
I'd love to see this. I'm a huge fan of the information availability the internet has brought and a firm believer that education is transforming into something we do of our own volition and in our own way, aided by freely available information online, and what you're working on sounds like potentially an excellent tool congruent with these ideas. Do you have something you could show yet?
Trying to add a shoutbox this month, to get feedback more easily than the survey. Happy to pay for one. I haven't found one I really like. Some feel a little glitchy/spammy.
I feel like a shoutbox is the wrong way to get feedback for this. Maybe adding a little more of a personal touch to your ask about the survey would help. Something like "I want to make this site awesome. Filling out this survey would help me do that. I read every response."
I've done shoutboxes in some other projects and they've worked amazingly. But in those cases, they were manually built and a lot of effort to maintain.
The main issue with the survey is that it's incredibly one way too. The most frequent feedback is that outputs are conflicting. They're designed not to conflict. Some of it might be poorly explained (e.g. tsundere, mean on the outside, secretly nice when you get to know them). Most are edge cases, and it would be good to know more details.
On (some) weekends I work on Megadetector GUI [0]. Megadetector [1] is an object detection model trained on millions of camera trap images and is widely used by conservationists. The issue is that it's rather technical to set up, so I made a GUI for it.
Currently working on a brand new version (not public just yet) that will use the latest MD version (v5 is way faster), better UI and most importantly GPU support out of the box.
An accurate recreation of the mmorpg final fantasy 11 when it first released - emulation software (https://github.com/AirSkyBoat/AirSkyBoat) and the upcoming private server horizonxi.
The README explains all, with words and animated GIFs, but to summarise:
`shite` is a low-performance way to turn heredocs, org-mode text (or md or html), and inotify events into my website (https://evalapply.org). It is about 350 LoC of FP-style Bash. The hotreload does not use javascript. shite is terrible and I love it :)
Rendering heartfelt follicle trimmage to that glorious animal has been an unmitigated delight (except for templating, with sucks universally).
Hey, this is one of my projects too! I use a DJI Mini 2 along with the Litchi flight control app, which allows doing things like building up waypoint missions for covering a large area. I tried doing post-processing myself with OpenCV image stitching, but found that OpenDroneMap's NodeODM program was really nice to use and produced better results. I then wrote a program[1] to parse the output of the NodeODM results package, and upload it onto my website[2] that displays them using MapLibre.
At this point, I can lay out my waypoint missions ahead of time, grab my drone bag and some coffee, walk out the site, and get the vehicle up within a few minutes. I have three batteries, so that usually translates to somewhere around an hour of flight time with reserves, and over 100 images. When I'm done, I can pull out my computer, upload the day's images to the locally-run Docker container instance of NodeODM, wait for the results to get spit out a bit later, then run my post-process script, scp them up to the VPS hosting my site, and there's the map! The automation has made the process a lot more enjoyable. I'm still hoping to tinker with things making the images separate layers so I can do a time-series of orthophotos made in the same location, but it's still been fun in the interim. Feel free to reach out if you'd like to chat--I don't see many other people that are also interested in the area!
Nice! I use Dronelink (I'm flying an Air 2 S), and open drone map as well. But I'm mapping hundreds of acres, so I end up throwing my images into S3 and processing them on a spot instance.
I'm also working on a time series. I'll check out your projects, thanks for sharing.
Oh, interesting! I've thought about going for larger areas, but most of the places I've been interested in looking at so far (kelp forests, wetlands) are fairly small, and it's lower anxiety to do low and slow flights where I don't have to worry about local air traffic. As a result, I've tried mostly aiming at high-resolution rather than large area. Of course, that helps with the size overhead as well--my existing computing setup is almost never the bottleneck. If you end up making any of your work public, I'd definitely be interested in taking a look--I've been starting to think about larger area surveys as well.
Gotcha--I'll have to run the numbers on pixel sizes between the Mini 2 and Air 2S at various heights; I'm sure that the Air has a much nicer camera, but it would be interesting to see how big the difference is.
I've actually got my commercial license test scheduled for the end of the month! I'm hoping that it might help me get permission to run surveys in areas like national forests or parks that are otherwise pretty much off-limits to recreational flyers.
I am working on something super basic, that I believe is still worth my time. It is just for me and my friends at the moment. I am creating a new flashcard app (how original :))
It is a combination of Anki, Quizlet and some other bigger apps. Taking their best features and turning it into one free app. I do not like to pay for a flashcard app, but Anki has really bad UX / UI and cloud features. On the other hand, the Quizlet algorithm is not great.
Therefore I am creating my own free app. Let's see where it goes. But one thing is clear. I do not want to make money of education.
I'm tinkering with The Tüül (https://github.com/incidentist/the_tuul), a site for making a decent karaoke video from any song in about 10 minutes. The members of my monthly karaoke group have been using it for the last year and it's been great to see what they've made. There's a lot of potential for automating some of the markup (such as count-ins, and INTRO/OUTRO/INSTRUMENTAL segments) but I need to redo the data model to make it easier to work with first.
FYI, since people seem to be actually playing with this, some of the sounds are very loud, others contain profanity, and the best ones meet both of the aforementioned criteria.
Sound suggestions and other feedback welcome. PRs also more or less welcome, although I have a pretty picky vision for the site (aka whatever makes me and my friends laugh). https://github.com/hfuller/omg.such.press
I actually have one week break from my businesses so I am diving into some coding again.
Thinking of building a social media app focused on voice, but constrained in some way so people use it very less and only connect to people close to them.
I made some toolbox organizer trays out of 2x12s, using a homemade CNC router to carve out the pockets. when the creative wells run dry, it's nice to just make functional, non aesthetic, basic projects to get organized.
Working on a keyboard-based inertial cursor navigation script[0] inspired by rvaiya/warpd[1]. I have just finished the inertia system yesterday but trying to figure out which default key combos to use as to avoid colliding with common OS keybinds in most situations
Having a new born baby. Also, I finally got suspend working reliably in Arch Linux. I also started writing a self-modifying bash script as a key-value store because I thought it would be dumb enough to be cool.
I recently implemented a feature for a client that I believe would be a nice little open source tool for Android apps that use the Google Maps SDK. If they give me permission I will hopefully build it, and I’m optimistically sketching it out.
The feature is a way to move info windows into view when they are rendered partially offscreen. Default behavior is to move the marker that was clicked to open the info window to the center but that can be jarring IMO. I liked the iOS Maps SDK behavior that just pans enough to bring the info window into view, so I duplicated that.
I'm still hacking away at LLDAP (https://github.com/nitnelave/lldap), a light and secure mini-LDAP server made to be easy to configure. It's targeted at self-hosters, potentially on small hosts.
We have a growing community that keeps pestering me with bug reports for the integrations with the myriads of services that can talk to LDAP, so things are going well. I'm struggling a bit to find the time to work on bigger features, though.
I'm working on a non-profit search engine (https://ask.moe) where the plan is to donate 100% of our profits to other non-profit organizations and charities (that are much better equipped to put the money to good use, and have a proven track record).
I believe it will be possible to make the first donation next month as we will finally reach the AdSense threshold of 600 DKK, and since October has been really good month then we will likely be able to donate 3000 DKK.
I’ve fallen deep into the mechanical keyboard hobby so I’ll likely be tinkering with qmk or 3D printing a case or researching how to route out a solid piece of wood to build a custom wooden case.
And in case somebody is curious, no, is not about functionality, although I do love using my custom keyboards! But it’s more about building something that include manual work, embedded development tinkering, aesthetics, acoustics and a fun community :)
I'm working on an open source terminal based flashcard app called Flashdown which uses a simple plain file format both for storing both the card data and your learning progress.
I've been working on an incredibly over engineered platform to run my home lab and to help learn about some lower levels of the stack I don't get to play with much in my day job as a consultant.
It's based on the kubernetes api server so while it doesn't understand "pods" or even proper "namespaces", it uses the same YAML resource model and api server code.
I'm working on and off on calendago: https://github.com/zikani03/calendago. Partly using it to help a friend get better acquainted with Go as well as getting familiar with graphics/image processing libraries in Go.
The goal is to run a free service to generate high-quality printable personalized calendars (as PDF).
A supabase alternative. Partly non-profit. There's just a handful of edge cases they prefer not to deal with. Working on it for weeks now and it's kinda fun.
Cool project. A fork or what? I run the full Supabase FOSS stack on my own stuff and I’m kinda curious what a new stack would offer above and beyond what Supabase already does.
Some parts were taken from their repo. Others are added like app-level authorization, a realtime engine using a c++ lib used by trading platforms, an option to sync some tables to a search database. edge functions are way down the roadmap but i plan to replace it with a js/ts one instead of a deno one.
I want it to be easily deployable to other cloud providers. Initially as monoliths which is ideal for small projects, then eventually as microservices that scale well for larger projects.
Ive started working on a supabase spring boot starter on the weekend.
https://github.com/tschuehly/supabase-spring-boot-starter/
I want to make it as seemless as possible without needing the Javascript Client.
HTMX made it possible to work around the Oauth Authenticating Challenges.
My phish streamer! I built a little web streamer for phish for all their taped shows from 1983-present, so I’ve always got music available to listen to and I never need to worry about music disappearing from availability.
I've been building a basic sqlite like database [1] to improve my understanding of relational dbs. It's still early days but it's been fun using different data structures than I typically do in my day job.
I've been working on a PEG-based Turing-complete language which creates completely standalone and trivially-embedded C. The "selling point" would be "A DSL for creating DSLs".
In the last few weeks I've been working on a libtcc-based add-on in order to build a REPL to help create/explore a desired grammar interactively.
the guiding principle is to give reasons why we should bother with a problem.
I have a couple of feedback that I have received that I need to implemebt.
Slowly working on a way to automate webpage to audio file conversions with "AI" supported TTS. There's a manual way to do it via the AWS interface, but it's a PITA. Want to just click a couple buttons and get it, while still doing my normal job.
I built a privacy preserving counter for my own websites and a dashboard that looks like a Renegade BBS Sysop waiting for caller screen. I’m running it all on a Raspberry Pi 400 in my home office.
I plan to do a write up about it, but thought this was an opportunity to give for broad strokes.
I'm doing some prototyping for a toolbar plugin for Neovim that uses images (via sixel output) for icons, sort of aiming for a VS Code-like toolbar. I'm trying to dynamically generate the sixel palette based on the selected Neovim scheme.
I'll check this out, thx. I think this project was created just to see what hand coding a lexxer and parser would be like. You know... just for fun. I want to go back afterwards and re-implement it with (f)lex and bison. Probably a good excuse to learn antler as well. I'll check out the tree notation stuff too.
I'm working on https://superintendent.app regularly. I use it at my work to process very large CSV files with SQL. Excel can't handle more than 1M rows. I could have used a database, but it is such a hassle.
It is built with Electron + React + Rust + C + Sqlite.
Right now I'm adding the workflow feature where it can chain multiple SQLs together called a workflow. Then, I would be able to handle the repetitive task better at work.
I'm experimenting with a mechanism to replace the selected text. You can select a text (on any app), activate the tooltip, and select one of the options, and that option can replace the selected text. The UX isn't as smooth as I want. There is some sort of modal blocking going on, and I'm still figuring out how to overcome that.
It's a a simple HTTP-based pub-sub notification service. It allows you to send notifications to your phone or desktop via scripts from any computer, entirely without signup or cost. It's also open source if you want to run your own.
You can use it like this (more in the docs: https://ntfy.sh/docs/):
It's 100% not-for-profit and always-just-for-fun.