▲ 410 points
Ask HN: What are you working on? (April 2025)
What are you working on? Any new ideas that you're thinking about?
back
1214 comments
A tree cutting tool.
Take photos of the tree from 6 different angles, feed into a 3D model generator, erode the model and generate a 3D graph representation of the tree.
The tool suggests which cuts to make and where, given a restricted fall path (e.g. constrained by a neighbors yard on one side).
I create the fallen branches in their final state along the fall plane, and create individual correction vectors mapping them back to their original state, but in an order that does not intersect other branch vectors.
The idea came to me as a particularly difficult tree needed to come down in my friends yard, and we spent hours planning it out. I've already gotten some interest from the tree-surgeon community, I just need to appify it.
Second rendition will treat the problem more as a physics one than a graph one, with some energy-minimisation methods for solving.
This is the kind of thing that makes me love HN. An idea I would never have thought of, with an immediately obvious use in multiple ways (fall path plus ideal lumber cutting?), probably very difficult, yet being tackled with one implementation already... and spoken of quite humbly.
Having joined may father and his friend during the process of cutting down big trees in the village neighborhood I can personally vouch that this indeed a cumbersome and very complex task for both the planning and the execution phases. However, for us the tasks are made easier by the trimming of the tree's branches since my father's friend is the expert tree climber.
From your descriptions, it seems that your tree cutting procedures do not involved precut of the tree's branches before cutting the tree down.
I've got the feeling that this cutting tree problem can be solved by constraint programming techniques [1],[2]. Alternatively generic tools for constraint programming, for example OR-Tools and MiniZinc can probably do the same if not better [3],[4].
[1] Logic, Optimization, and Constraint Programming: A Fruitful Collaboration - John Hooker - CMU (2023) [video]:
https://www.youtube.com/live/TknN8fCQvRk
[2] "We Really Don't Know How to Compute!" - Gerald Sussman - MIT (2011) [video]:
https://youtube.com/watch?v=HB5TrK7A4pI
[3] Google OR-Tools:
https://developers.google.com/optimization
[4] MiniZinc:
Testing this is real pain in the ass, you gotta cut real tree to see if it works in various situations :(
Funny, one of mine also involves trees -- but is mostly outdoor cleanup. The kind that involves decades' worth of it, thanks to what I'll just say is a lot of maintenance that wasn't done over a long time. There's an extensive amount of brush, leaves, etc of varying ages that could maybe be shredded up into something useful, invasive vines I'm still trying to deal with, and more old trash than I've fully figured out how to properly dispose of.
It's turning into various DIY rabbit holes, actually, with the next one (outside of various related landscaping stuff) being to gut a basement.
I would love to have such a model tell me how to prune my fruit trees as they grow up. Should be a fairly straightforward supervised problem with the right front end for the graph generation.
I was imagining something like this for pruning fruit trees — something to help noobs like me see how to put pruning guidelines into practice on a real, overgrown tree. Good luck!
That’s a great idea, but so much liability if the user is an amateur and follows steps incorrectly
I've been thinking for years about a safer alternative to chain saws. Something along the lines of a carbide coated wire driven by an electric motor and battery. Strap it to the tree, turn it on, walk away and some minutes later the tree falls down. The main difficulty is in how to drive the wire. Using friction would create fast wearing parts. Maybe a chain could be used instead of a wire. It could oscillate back and forth, instead of having to be wrapped and spliced to form a circle around the tree. It seems really strange that no one has come up with an alternative to chain saws for decades (except for large scale trucks that can process whole trees.) For small trees and branches even a sawz-all is safer than a chain saw. Inspired by spending some time sharing a hospital room with a guy who had a chain saw accident, but I still haven't come up with a workable idea. Maybe someone else can.
Forgive my ignorance but all the tree cutting I've observed has been based on climbing and cutting in segments from the top rather than letting the tree fall. Under what circumstances is it better or necessary to actually let the tree fall?
The first ever SQL debugger – runs & visualizes your query step-by-step, every clause, condition, expression, incl. GROUP BY, aggregates / windows, DISTINCT (ON), subqueries (even correlated ones!), CTEs, you name it.
You can search for full or partial rows and see the whole query lineage – which intermediate rows from which CTEs/subqueries contributed to the result you're searching for.
Entirely offline & no usage of AI. Free in-browser version (using PGLite WASM), paid desktop version.
No website yet, here's a 5 minute showcase (skip to middle): https://www.loom.com/share/c03b57fa61fc4c509b1e2134e53b70dd
I’m working on Popgot (https://popgot.com), a tool that tracks unit prices (cost per ounce, sheet, pound) across Costco, Walmart, Target, and Amazon. It normalizes confusing listings (“family size”, “mega pack”, etc.) to surface the actual cheapest option for daily essentials.
On top of that, it uses a lightweight AI model to read product descriptions and filter based on things like ingredients (e.g., flagging peanut butter with BPA by checking every photograph of the plastic or avoiding palm oil by reading the nutrition facts) or brand lists (e.g., only showing WSAVA-compliant dog foods). Still reviewing results manually to catch bad extractions.
Started this to replace a spreadsheet I was keeping for bulk purchases. Slowly adding more automation like alerting on price drops or restocking when under a threshold.
I am about to begin a PhD in astronomy. Until last month I was working at Caltech for 3 years on code which calculates orbits of asteroids to high precision. This code is being used on several NASA telescopes now to predict when they will image known asteroids (NEO Surveyor, SphereX, maybe Roman eventually). I was allowed to open source it and I am planning on making it the basis of my PhD research:
https://github.com/dahlend/kete
It can predict the location of the entire catalog of known asteroids to generally within the uncertainty of our knowledge of the orbits. Its core is written in rust, with a python frontend.
https://github.com/colibri-hq/colibri/
Colibri—a self-hostable web application to manage your (and your family's) ebook library, intended as a companion to Calibre. I want it to be a friendly, simple, capable, opinionated app to review your books, add metadata to them, get them onto your reader, share them with family and (few) friends, create a public shelf for bragging, connect with Goodreads etc., and exchange comments and reviews on books.
This is explicitly not intended to ever be monetised, and I enjoy all the implications that has on the design. Colibri is as much a tool I personally want to use, as it is a study in small-audience user interfaces, and the quest to build the perfect book catalog schema.
I'm looking for fellow book-loving people to work on Colibri, to create the best personal digital library possible. If you're interested, feel free to reach out via email (in bio), or on GitHub.
https://mysukari.com - A Diabetes management platform
I got diagnosed with type 1 diabetes in Feb (technically LADA as it's late onset). I'm the first in my family with it so I had zero info on it. I tried getting some CGMs to use but most don't work in Kenya as they are geo-locked, and even apps for measuring carbs like CalorieKing are not available in my region. I was really frustrated with the tech ecosystem, and started working on My Sukari as a platform of free tools for diabetics.
I mostly get time to work on it on the weekends, so it's not yet ready for public use, but I've fully fleshed out one of the main features: Sugar Dashboard - A dashboard that visualises your Glucose data and helps you easier analyse it.
To help with demos, I've shared my Sugar Dashboard here: https://mysukari.com/tools/sugar-dashboard/peter
I'm really passionate about this and getting as much free, practical tools in the hands of patients (it honestly shouldn't be this hard to manage a disease)
So many cool technical projects here. But I am doing something completely different - masonry. Repairing walls in 3 rooms. It includes reinstalling dozens of falling off bricks, installing 30 or so power outlets, replacing old windows with bigger modern ones, fixing openings for the doors and plastering everything afterwards. On one hand it’s interesting, because it’s very different from the dayjob. But doing it by myself pays my newish car in cash immediately. However I wouldn’t do it for money somewhere else, it’s really really hard work.
Instead of masonry I would like to work on time of flight cameras. But the day has only 24 hours :-(
I am working on the sunflower plant density estimation problem. The goal is to be able to estimate the germination rate as early as possible. Farmers benefit from such information, because:
- there are lots of expenses still to be made (fertilizer, pesticide, salaries), which may not be worth it if germination is under certain threshold
- if detected early, there is still time to plant another grain or to fill up the missing plants (requires precision seeders and seeding maps)
- is a very good proxy for yield estimation (farmers often trade futures even before they have harvested)
For the purpose I have created a dataset (a collaboration between my employer and Sofia University) and published it in order to enable scientific collaboration with other interested parties. Still working on the dataset annotations.
https://huggingface.co/datasets/su-fmi/sunflower-density-est...
Working on a browser extension to make it easier to find content on YouTube that fits your interests.
I watch a lot of YouTube videos and have found it very annoying that YouTube latches onto one or two topics that you've watched and only recommends that type of content over and over again. Even if you use their "Not Interested" tool, not a whole lot changes in your recommendations.
At the end of last year I launched Relevant - a crowdsourcing website where users can categorize the channels they watch into a defined hierarchy of categories ranging from broad topics like "Science" and "Gaming" to more specific ones like "Phone Reviews" or "Speedrunning".
Although I've had good feedback on the website, engagement has been relatively low and I think that's because it's a big ask to have someone navigate to the website to find the content. This year I decided that I'd bring the content to them by making a Chrome extension that lets users interact with Relevant directly from within YouTube.
It's still a work in progress but I'd love to get a first version out within a month or so to start spreading the idea and gathering feedback. If this is of any interest to the people here on HN then please let me know what you'd like to see most on your feeds.
I have been working on http://phrasing.app - a language learning & acquisition tool for polyglots. I’ve been using it to study ~12 languages (5 on maintaince, 2 seriously studying, 5 casually “studying”) and it’s starting to feel really good. If anyone is learning/maintaining several languages, please reach out! I’m looking for beta testers in as many languages as possible (it supports 120+).
In what I believe is still the spirit of the question though, I discovered Maltese these week and have added it to my casual study. It’s a Semitic language (closely related to Arabic), written in the latin script, with about 40-50% of its vocabulary being Italian/Sicilian based. It’s become my new obsession
It's niche and boring, but I help immigrants pick German health insurance.
The existing information is mostly blogspam from non-experts who try to make a quick buck. They only recommend the two brands with an affiliate program.
I wrote a better guide with help from competing insurance experts. The information is clear and succinct without oversimplifying things. It addresses the specific needs of immigrants.
Then I turned the advice into an interactive recommendation tool. People get clear, specific advice in a few seconds.
The best advice is "don't choose yourself, talk to a broker". The problem is finding a honest one. It took me years to vet a good one. After testing him for a year, I have set up an affiliate partnership from scratch with him. The partnership incentivises honesty and neutrality, because he has a lot of skin in the game.
I'm super excited about it. I can't overstate how much of an improvement it will be. Readers get far better advice and easy access to an expert. The broker gets a steady stream of well-informed leads. I get a commission for my trouble. It's a win-win-win situation.
I'm almost finished with a project that has taken up quite a lot of my life for the past few years, a book on Estonia! After working for their government and returning to the U.S. I wanted to better understand how the country modernized so effectively after gaining their independence from the Soviet Union so I did a fair amount of research into the topic and many interviews which culminated in a book, Rebooting a Nation.
If you're interested in Estonia, e-government, building tech hubs, and the future of the nation state I'd love if you take a look (and let me know what you think). It's available on Kindle now but Oxford University Press will be shipping out physical copies May 15 and buying from a smaller press is always appreciated!
https://global.oup.com/academic/product/rebooting-a-nation-9...
Myself.
Been a freelance dev for years, now going on "sabbatical" (love that word) imminently. Just moved to reduced hours, still in the transition and unwinding phase.
Planning to do a lot of learning, self-improvement, and projects. Tech-related and not. Preparing for the next volume (not chapter) of life. Refactoring, if you like, among other things.
I'm excited.
I hacked my old kindle and turned it into a eink dashboard for my daughter’s school! Now planning to enahace it a bit further and make it easy to customise.
Here’s a a detailed write up of the process: https://samkhawase.com/blog/hacking-kindle/
Reverse typesetting: reflowing page layouts where you don't have knowledge of the typesetting structure, i.e. a scanned physical book or PDF paper. Naive rules-based heuristics based on the dimensions of bounding boxes and gaps. Point is to reflow things for resizing to eink readers. (Specifically the size that fits in my pocket which I carry around. User #1 is me). Building in Common Lisp and targeting an Emacs mode for interactive execution with manual feedback.
A pentagonal geospatial indexing system: https://a5geo.org/
If you’ve used H3 or S2 it should be familiar, the major difference (apart from the fact it uses pentagons) is that the cell areas are practically uniform, whereas alternative systems have a variance of around 2 between the largest and smallest cells, making them less useful for aggregation. The site has many visual demos, e.g. https://a5geo.org/examples/area
The code is open source: https://github.com/felixpalmer/a5
PodSnacks (https://podsnacks.org) — Personalized AI summaries of top podcasts, delivered to your inbox.
I've been building PodSnacks because I found it overwhelming to keep up with podcasts across tech, business, and science. PodSnacks uses LLMs to summarize the most popular episodes from shows like Lex Fridman, Acquired, All-In, Invest Like the Best, and more.
You choose your favorite shows, and we email you short, high-signal summaries — no audio to skim through, no endless backlog guilt.
So far:
126K+ episode summaries generated 92K+ hours of podcasts processed 48–50% open rates 2,900+ early users
Still iterating and adding features like bundling by theme, language translations, and audio feeds.
If you're the kind of person who wants more inputs without more noise — would love for you to check it out.
Always open to feedback from HN!
I've recently re-watched John Carmack's 2004 Quakecon keynote (https://www.youtube.com/watch?time_continue=602&v=OOVUZZr655...), which happened during the Doom 3 launch. In that keynote, he mentions that he didn't see any differences between stencil and shadow maps. He mentions that he experimented a lot with both approaches, before settling for stencil shadows which were much faster.
I got curious about that statement, since shadow maps tend to look much different. I also knew that he left part of the experimental renderer in the GPLed code. So, I've decided to go on that rabbit hole of Doom 3 graphics, specifically Carmack's experimental renderer and ended up implementing his approach, as well as adding some poisson disc sampling and fixed the peter panning: https://github.com/klaussilveira/exp-dhewm3
I also spend a lot of time on archaic game engines. I like to call it software archeology.
I launched my web-based notepad calculator, https://numpad.io/, a few years ago.
Right now I'm working on a version 2 that has user accounts, multiple documents, markdown support, and document exports. Everything is local-first and it uses CRDTs to sync documents.
It looks like this: https://i.imgur.com/Plk1DQ4.png the calculator is mostly the same for now, with a few improvements. It's unstable right now, so I don't want to publicise the dev url, but if you'd like to become a beta tester email me at contact@numpad.io
I just put live a daily logic puzzle, Clues by Sam. I've been working on it for a while now, mainly on the level generator. It was tricky to generate levels that are solvable using logic (no guessing needed), and also fun to solve (no crazy long deduction chains, but also not just obvious things). I'm sure the implementation still has some quirks on some devices, so would love to hear if you encounter issues!
I'm currently working through Frank Sikora's "Jazz Harmony" book [1] to learn to play Jazz piano.
In parallel, I'm building an exercise generator "Jazzln" [2] to help me practice.
[1]: https://www.goodreads.com/book/show/54391815-jazz-harmony
Creating stealth group in a huge Fortune 500 company with the blessing of my immediate boss but no other higher-ups. Trying to productize critical consulting tool sets in the utility industry so we can stop repeating ourselves for the 100th consulting engagement.
Yes, customer is a special snowflake but they still need 90% or whatever every other client in this industry needs.
Feeling increasingly like this is a fools errand.
Even though we've proved this out with tool sets strung together with duct tape and safety pins, and are therefore the most profitable group within our department, we still need to be 100% billable.
It's only because we're the most profitable group that we can pretend we're all billable while I work with two other people to bootstrap this crazy project
Edit: anyone hiring? Just found out my boss is quitting.
It tried out vibe coding this weekend. I've been using AI to get smaller examples and ask questions and its been great but past attempts to have it do everything for me have produced code that still needed a lot of changes.
I hit an OpenSearch bug this week where you can't get any browser based requests to work. Its due to zstd becoming a standard part of Accept-Encoding and OpenSearch not correctly supporting it so I wanted to install a browser plugin that modified the browser HTTP request headers to my servers.
I don't know about everyone else but I love that browser plugins are possible but I hate having to find them. Its mostly due to never knowing if you can trust a plugin and even if you find one, you have to worry about it being bought out in the future. With vibe coding I was able to build a browser extension in 45 minutes that had more features than I originally planned for.
I spend more time documenting the experience than building which is wild. If you are interesting you can look at the README in https://github.com/mattheimer/vibe-headers
But I left the experience with two thoughts.
Even seasoned developers will be using vibe coding in the future.
I think in the near future the browser plugin market will partially collapse because eventually browsers will build extensions themselves on the fly using natural language.
A note taking app - I know, I know, been done a billion times. But really, no software I’ve found works the way I want it to, so I’m just trying to write my own, starting with a front-end agnostic core and a TUI.
If it strikes a chord with anyone, I’d love to collaborate! The concept is centered around organization bubbling up naturally from dumping info in with tags, and “typing” your tags so that when you go to a tag’s page, the layout is customized based on what it is - a project, person, etc. A project could have all relevant tasks and notes listed, whereas a person might have name, contact info, etc.
Working on a database compatibility tool. So your app works on SQL Server 2022 on-prem, but does it work on Azure SQL DB, AWS RDS, GCP CloudSQL, etc?
These cloud flavours have a compatible SQL dialect, but it's often details like missing features (CDC and Auditing on RDS are good examples) or differences in system objects that make it difficult to support your app on these platforms.
I capture all sql statements, run them through multiple SQL parsers to find all the system objects your app is using (tables, functions, stored procedures, etc). I then check them all against a catalog I have built of all system objects for every version of SQL Server on every platform.
I then give a report to see which platforms your app will work on, which ones it wont work on and which system objects are the problem.
Other database engines will be added once I get it working end to end (almost there).
I'm working on a new open source CAD program for 3d printing based on Signed Distance Functions.
Benefits of SDFs over the standard Boundary Representation (used in Freecad and similar) are that you can do "pattern" operations with domain repetition, which means making N copies of a feature is O(1), vs O(N^2), you can deform objects with domain deformation, which means if you have a closed-form representation of how you want to deform space you can basically directly apply that to your object, procedural surface texturing is easy, CSG operations are easy.
The big drawback is that it is hard to provide any workflow based around "selecting" faces, edges, or vertices, because you don't naturally have any representation for these things, they are emergent from the model's SDF.
I have some blog posts on my progress: https://incoherency.co.uk/blog/stories/sdf-thoughts.html and https://incoherency.co.uk/blog/stories/frep-cad-building-blo...
I am solving the "selecting faces" problem by having the SDF propagate surface ids as well as distances. So the result of the evaluation is not just the distance to the nearest point on the surface, but the id of the specific surface that is nearest.
My next big frontier is reliably providing fillets and chamfers between arbitrary surfaces. I have a handful of partial solutions but nothing complete yet.
The most promising idea is one that o3 came up with called "masked clones", the idea is roughly to make a clone of the 2 surfaces you want to blend, mask them by intersecting with an object that is like a "pipe" along the intersection of the 2 surfaces, apply the blend within the pipe, and then add this "blend pipe" as another child of the lowest common ancestor of the 2 blended surfaces.
And after that I need to work on more standard CAD stuff like constraint solving in the 2d sketch editor.
https://undetectag.com I developed a device that turns an Airtag on and off at specific intervals.
Current Airtags are detectable right away and cannot be used to track stolen property.
That device allows you to hide an Airtag in your car, for example, and someone that steals your car will not be able to use some app to detect it. The Airtag will also not warn the thief of its presence. After some hours, the Airtag turns on again and you can find out its location.
It’s not foolproof, as the timing has to be right, but still useful.
I am working on a super mario land re-implementation for game boy using C. It's made with the GBDK toolkit so the ROM can be run on a gameboy.
Currently the scrolling and background collision has been implemented, I am working on drawing the enemies... their is still a long way to go.
As for why I am doing this is nostalgia and fulfilled a childhood dream
repo at https://github.com/odrevet/marioland-gbdk/
I'm working on an escape room! Initially I was working on a software/hardware bundle that I was planning to market to other escape rooms but I think that is the wrong approach. So I am going to build a bunch of modular stuff in my garage and eventually start my own, its been an awesome project so far! I want a more dynamic and action oriented experience so it might not really be an escape room anymore but I don't know what to call it yet.
Escape rooms are honestly... almost always a let down but the concept has a lot of potential and there are some really neat ones that standout like this local one where you pilot an airship https://www.portlandescaperooms.com/steampunk-airship
Once I build the best escape room on the planet, I can consider selling the tools.
I'm hacking on Trailbound, a hiking and backpacking planning tool. The goal is combining a light version of the routing capability of Caltopo or GaiaGPS with more granular "leg" mapping based on how I learned to plan off trail trips and sharing functions that make it easy for my wife to know where I'm going to be, who I'm with, and who to contact if I'm not back in touch when I said I would be.
Its been a blast to build. At one point I was hosting my own ORS server but that's extremely silly to do when Mapbox has a very generous free tier. Learning about all of the open source tooling and open data available in the mapping world has been incredible.
The cost of the Hetzner box it runs on isn't much more than a Caltopo pro subscription with the added bonus of being much easier to share with non-hikers.
A quick demo: https://www.loom.com/share/a5f7a7c23457400aa92b3f0f71a0008f
The app: https://go.trailboundapp.com
No marketing site, no onboarding, or any real UX attention paid to the app at this point, it is mostly Just For Me and will probably remain that way once I land a job again.
I'm starting to chip away at an iOS app so I can get offline access to maps and routes, but I'm not a Swift dev so the going is slow.
As a follow up to my relatively successful series in x86 Assembly of last year[1], I started making an OS that fits in a bootloader.
I am purposefully not doing chain loading or multi-stage to see how much I can squeeze out of 510bytes.
It comes with a file system, a shell, and a simple process management. Enough to write non-trivial guest applications, like a text editor. It's a lot of fun!
Not quite done with it yet, but you can see the progress here https://github.com/shikaan/OSle and even test it out in the browser https://shikaan.github.io/OSle/
[1] https://shikaan.github.io/assembly/x86/guide/2024/09/08/x86-...
I am working on an app for rhythm training for dancing.
The first step is beat detection in music - here is my code for generating videos with beat counting: https://github.com/zby/beat_counter (vibe coding warning). Beat detection seems to work OK, downbeats are a bit more tricky and I have found not good solutions for further structuring of music (for dancing they usually count to 8 - that is two 4 beat measures - or 6 - two 3 beat measures for Waltz for example, I am not sure about 2/2 measures and other). Seeing all the LLMs generating music I thought there should also be LLMs interpreting music - but so far I have found no such model the available algos seem to be from a previous decade.
My current plan is to test the counting with people with good rhythm sense and once I find a good algo for beat detection I'll proceed with writing the app.
We’re building a new social-enabled git collaboration platform on top of Bluesky’s AT Protocol: https://tangled.sh
You can read an intro here: https://blog.tangled.sh/intro (it’s publicly available now, not invite-only).
In short, at the core of Tangled is what we call “knots”; they’re lightweight, headless servers that serve up your git repository, and the contents viewed and collaborated upon via the “app view” at tangled.sh. All social data (issues, comments, PRs, other repo metadata) is stored “on-proto”—in your AT Protocol PDS.
We just shipped our pull requests feature (read more here: https://blog.tangled.sh/pulls) along with interdiffs and format-patch support! https://bsky.app/profile/tangled.sh/post/3lne7a4eb522g
We’ve also got a Discord now: https://chat.tangled.sh — come hang!
I've always been into MUDs since the mid-90s and every now and then I get this idea to write one from scratch (again) and I get pretty far and then I lose interest.
The new one is browser based.
The mud engine runs in a web worker and takes advantage of some of the modern web tricks to do stuff. For instance, data files (think: area files that don't change often) can be stored remotely and then cached with a service worker. This allows the MUD to run offline. But that's only fun if you're playing solo.
IO between the UI and worker is handled by message passing.
Multiplayer is handled by the MUD opening an outbound connection (probably websocket) to a connection collector host. Other players would then connect to that host and have their IO routed appropriately. The host can even be smarter: it could be a specialized Discord client, allowing users to play from there. Firebase may also be involved. I don't know.
The important bit is that this is still basically message passing, so the engine won't need to know the difference between the local user and a remote user.
The MUD database would be an IndexedDB. Probably. I haven't thought as much about that yet.
I am sure all of this is theoretically possible, at least.
Creating a modern development environment for win98. Currently I’m working on a git client. I have no patience for learning C. So backporting git is not an option. I also don’t want to use cygwin. So I’m using a server to expose git as http endpoints and coding a git client in php to use in msdos. I have Vim 7.3 and and gnu coreutils working in msdos already. So soon I will have a very nice dev environment. I want to create a one click installer which gives you xampp, vim and then all the tooling I’ve created. I’m also interested in creating SPA that works in IE5.5. But I’ll do that when my tooling is ready.
I'm building a platform for small-city business directories called TownWire: https://townwire.com
We launched our first location last week and have had a great response so far from business owners and residents alike: https://canandaigua.com With the current tech climate of AI and big tech, things have gotten so impersonal for the majority of actual people. I'm betting on communities and the small businesses that serve those communities in the medium and long term.
I'm working on my second self-published technical book related to Elixir. The first one has been a surprising success, but since it was about learning to use Ecto on the basis of an old pedagogical toy database (and tiny dataset), the new book is about the development of a production-grade REST API with Elixir and Phoenix.
I find that many books out there are focused on documenting the "happy path" for rather small and simplistic applications without boundary conditions or business thinking behind them.
So, I thought (as with the first books) that it mix things up by also documenting the business context, the questions, decisions, and the decision-making process itself, as well as all the gotchas and "side-quests", rather than showing "here's how you do it" and then expecting the reader to suddenly make the jump from tutorial hell to actually software engineering.
Overall, it's enormously enjoyable, and I hope it goes as far as the first book, and possibly even farther.
https://easel.games/about
A programming language for teenagers to learn to code by making multiplayer games. I've spent 3 years making the multiplayer completely automatic so you can just code it like a singleplayer game, then flick a switch and it just works. My hope is that teenagers will find it more engaging if they can play their games with their friends. A bit like a combination of Roblox and Scratch.
Currently trying to implement some region affinity so it doesn't just put everyone in the world in the same game. It's rollback netcode so the latency is very good even across continents, but it can't overcome the fact that the world is just too large.
The Goodreads of music. I'm a lifelong music nerd and have developed habits for finding and curating good music, both new and old. I'm codifying my habits and heuristics for discovering new music into an AI-powered social platform that can be used by anybody to widen their horizons in terms of music they listen to and to share their taste with others. https://waxnerd.com/adnan
> This weekend I’m working on mobile app that allows you to upload photos and it turns everything into a stitched anime (ghibli or not) with movements and eventually sound and script. Great to make mini animes from your day or travels or anything.
I vibe coded the above including everything: code, design, logos. Just did it solo. It has all error handling, video generation notifications (it takes a while) and credit system. I myself can't believe it's been done in a month with AI. It's already in closed beta in iOS and android app stores. Let me know if you want to try it out before public release.
My quoted comment above was 28 days ago. This is working on this part time and with a family.
EDIT: Added context.