* Improving the largest bilingual Manx <-> English Corpus + Search Engine. It's recently become the #1 point on the Isle of Man's 10 year language plan, and will probably end up informing a new dictionary for the language
* Generating a machine-translation system using the above, probably also adding Manx to FLORES-200
* (maybe) Generating a Text to Speech engine for another language
* Moving my book over to my new laptop, and grinding it out to completion
* Scraping websockets to improve the accuracy of bus timetabling. Eventual aim to improve the experience of using the bus to reduce the number of cars on the road. Probably some F# webservice handling most of it, with React Native (web)app, distributed via NFC tags/QR code.
I have started planting local fruit trees with an aim to plant 120,000 local fruit trees in the next 3-5 years. Currently I have reached 3000 in the last 5 months, with another 1000 in polybags.
Leaving programming after 25 years of it, I wanted to do something that will help world be a better place even after I am gone. Trees have always appealed to me. Hence local fruit trees -- sell fruits to run farm once fruiting happens in a few years. Local plants are localized to climatic and soil conditions of the area thereby requiring the least nutrition, water and care.
Its going to be a bit slow from here for few reasons, I have to learn to grow saplings faster and out of season, make fertilizer year round and figure out how to ensure plants are growing. I have figured out some of it, but financial resource crunch will hit me in next 6-8 months. So now trying to come up with ways to handle that.
Help in any of the above is extremely welcome. Does anyone feel social impact funds will want to look at me?
I just restored some backups from 1995, and am digging through that archive to see if I can pull out the multi-tasking libraries I wrote for Turbo Pascal under MS-DOS. No practical use, mostly nostalgia.
I didn't document things as well is I should have, of course I was the only user, and had no thought that future me would be looking at it 27 years later.
I’ve been working on needle[1], a CLI tool (and associated library) that can detect openings/intros and endings/credits across TV or anime episodes. It decodes audio, fingerprints it in chunks, and then compares chunks across files to find common sequences.
The end goal is to use the library to build a Jellyfin plugin (C#) that handles skipping intros. I think I’ve figured how to call a C library from C#, but there is a lot of work to do to actually get a functional plugin.
I'm still working on collAnon[1], a private discussions platform(behind a PWA) with bias avoiding intent through temporary anonymity of the answers(which also means that discussions have an end date).
In the latest development I decided to change route and make it freemium, with all the previous premium features now available to everyone.
This is to make it more appealing to try this kind of discussions that are not your usual way and give ample breath to use all the features until this platform finds its audience I hope ¯\_(ツ)_/¯
A handheld, esp32-based implementation of the pico-8[0] console. Basics work already, sidetracked into writing a terrible Lua to C++ compiler to squeeze some extra performance out of some pathological-case games.
A lots of fun things, feel free to connect with me on Github if anyone would like to chip in and help:
- An Electron template with built-in secure features. I regularly maintain and update the template so it's supported with newer versions of packages.
- A collection of free, public-domain recipes. I've wanted to do some version of this for _years_, but am finally getting around to do it again. This time, I'm paying chefs to create recipes for me that will be licensed as public domain.
A new paradigm for building web apps: a framework integrated with a runtime integrated with an IDE! https://gadget.dev
We think that so much of software development is still the same stuff repeated over and over: auth, hosting, CRUD, search, tables, forms, etc etc. Each app always has some juicy special something about it, but that core is wrapped in layers of stuff you don’t need to redo every time. Our mission is to make the first and only lines of code you write super pertinent to the specific problem you’re solving.
We’re starting with Shopify apps cause we can give developers a one click, fully managed, code-extensible API integration which is a lot of work with the Shopify API otherwise. Would love to know what y’all think!
I just finished grinding to meet the submission deadline for a contest Webtoons is running. The chance at 50k and a development deal sure made it worth pretending I had a work ethic for a couple weeks when my Patron campaign was gonna pay me for it regardless.
I'm gonna take a break for a few days, maybe draw some self-indulgent porn, then clear out my commission queue and start working on the second half of that story.
There are always plenty of ideas, the hard part is figuring out which ones are worth any of my time. Right now I'm working on a spreadsheet that turns everyday spreadsheet users into programmers. The absurdity of the ambition is not lost on me.
The startup I was working for went under at the end of May, so I've been putting my focus into a new UI component library I started at the beginning of the year. It's called Skeleton [1] and utilizes Svelte + Tailwind. Makes building UI for SvelteKit apps a breeze.
Reception has been great so far, just trying to get the word out and gather feedback and additional contributors.
I launched a Discord [2] today. Feel free to stop by and say hello!
I’m building a tool to automate billing audits for physicians.
Doctors work their butts off for the benefit of others with the expectation that they will be compensated accordingly. Unfortunately hospitals aren’t incentivized to make sure all of their procedures are correctly accounted for, and as a result are doctors are often unpaid for their work.
Currently their only recourse is manually going through the EMR and tallying their billables themselves, but they don’t do this because it takes time which they don’t have.
Electronic Medical Record software is built for administrators and doesn’t make it easy to automate, so I’m taking a Robotic Process Automation approach.
At the moment I'm working on FastHash[1], a pet project of mine to port a few high-performance non-cryptographic hash functions to C#.
I'm also trying to build FastLinq, a value-by-reference Language Integrated Query (LINQ) optimized for high-performance scenarios. It is kind of a weird mix as LINQ in .NET is known for its high overhead.
Finally, I'm working on an Office setting synchronization application. I heard a podcast with Paul Thurrott complaining about the lack of sync solutions, so I thought I would do one for fun.
Currently working on https://kubirds.com , a Cloud-Native Supervision Engine. Think alerting similar to Nagios/Icinga/Centreon but as a Kubernetes Operator which can run almost any Docker image.
We are also building on top of this a trading bot with a FinTech company, and a SaaS website monitoring service for non-tech people.
We are also considering building, still on top of Kubirds, a No-Code notification automation/aggregation platform -> https://1-click-notify.carrd.co/
It will sound like an Show HN, but if you want to use a microblog that you can follow more regularly, instead of opening a thread on Ask HN every time for this question, here you go. Theme was especially likened to HackerNews.
Hacker's Work: What are you working on? A microblog for your works (https://hackerswork.com)
The idea is to make apartment living less lonely, and generally increase goodwill between apartment neighbors.
Non-apartment people are not being excluded, but I think apartment living has lots of unique issues that homeowners - even, say, townhome and condo owners - do not have. And, realistically, you have to start with a focus on something -- some one thing -- in our case, it's apartment-living.
I'm still working on Fresh Cards[1], a flashcard app I've written for iOS and Mac.
I'll admit that it started out as "yet another spaced repetition flashcard app" but over the past few months, with constant UX refinement, I've forged a unique approach (IMHO) to how you review your cards and browse your deck. I've received a lot of good feedback from users in my Discord, so I'm constantly changing things up to improve it.
I will say that the Anki import isn't 100% at the moment, but works reasonably well for simpler decks (ones that don't use HTML or javascript). I'd like to get to 100% Anki compatibility, but it will require some additional features to be implemented first, like a templating system.
I’ve been working on a podcast app that skips ads.
After a couple years of being in over my head with ML, I’ve finally got an audio pipeline that consistently detects obvious ads. Have been building the app all spring and hoping to release it soon.
If you’re interested in trying it out or have advice, let me know!
adblockpodcast.com
Harvard CS50, I was coding a bit as a kid in the late 90's in BASIC, and have been in tech most of the past decade, but never went beyond HTML, CSS and grabbing a basic script off Stack Overflow since then, so now I'm trying to learn the foundations of CS and eventually teach myself a language or two so I can evolve from building mini projects in no-code tool to making more meaningful things and more powerful apps.
I am trying to create a small web app to be able to quickly check grammar rules for my language.
Nothing super exciting, but gives me a chance to work on the app itself (next.js), database (postgres with full text search), and some parsing magic to get the raw data I need.
I'm working on Cling, a training community for rock climbers.
Rock climbing is a complex, metric-driven, and growing sport that doesn't have good-enough tooling in tech. Outdoor vs indoor is very different, so we're focused on significantly improving the climbing gym experience for now.
What made you choose to build an iPhone app before an Android app? I'm curious where that debate stands in the last few years as I haven't seen the arguments for a long time.
Main reasons for me are speed, quality, and showing proof of market.
I have worked on Android previously, but I'm much more familiar with iOS. In a space with competition, product quality is one of the important factors to gain user interest. That's also the reason I didn't go with React Native, since the tech needs will grow to a point where I won't be able to provide a good enough UX. Once I've proven value, I can hire an Android dev.
Marketing tech app -- https://datacx.ai/ It's still early days but after being away from the marketing space for a few years, I'm so surprised at how little has changed. I want to help promote a better way to collaborate on analytics, starting with ecommerce and moving to broader CDPs.
Also currently thinking about building a Python 3 search library to be used against Google App Engine. Google deprecated the one from Python 2 and recommends using Elastic Search (which means you're no longer dealing with a Serverless solution)
I’ve been working on a music visualizer that runs in the browser by listening to the microphone on your device. Right now it looks just like the bars on Apple Music but I’ll be adding more visualizations like the ones from cool old stereos from the 90’s.
I have been wanting to dive deep into a Rust project and the challenge of implementing the MongoDB protocol and then translating it into some sort of SQL counterpart was the first thing that really clicked and got me excited enough to get me working on it nonstop for 3 weeks now.
I have created a product that relies on MongoDB for a document store but doesn’t really need any of the distributed features to really justify having a hosted MongoDB or DocumentDB instance. Now that we’re trying to turn this into a product, we’re seeing that some companies have a little bit of resistance around managing yet another database. Most of our clients already have and manage PostgreSQL in one form or another. I knew that PostgreSQL already offered first class JSON support, but I didn’t want to rewrite the application data layer from scratch if I could avoid it. That’s when I started researching if there was a “proxy” that would translate the MongoDB protocol - that I was completely ignorant about - into PostgreSQL. To my surprise there was nothing ready for production use but I found MangoDB that later on became FerretDB. I delved into the code and was in love with the idea. The team around is really nice, but I found that they had greater ambitions - they basically wanted to offer multiple backends, namely Tigris, on top of PostgreSQL.
On the other hand, I have been waiting to find an excuse to delve deeply into the rust ecosystem but never really found something I was passionate about until I had the idea of challenging myself to see if I could learn about the protocol that MongoDB uses by relying on their public documentation and the hints I found on FerretDB.
Another thing I added to my toolbelt while developing this was about creating parsers. In order to transform MongoDB JSON to SQL queries, I ported an existing library from the MongoDB team from PEG.js to pest.rs!
It’s in very early stages, and it’s work from someone that is not yet super comfortable with the stack so keep in mind this is the beginning of a journey for me that I embarked out of pure joy on getting a tiny bit better on rust and making things click internally.
thinking about converting my new tab extension --
https://new-tab.vlad.studio/ -- into a real website, accessible from any browser, not just Chrome. As always, it is harder than it seems.
Learning Loop helps users spend 10x less time searching how to learn a skill, by showcasing how other people mastered it and offering competence-based and demographic filters on top of it.
I'm really interested in exploring some AI ideas related to emobidment. To that end, I needed a "body" to put an "AI" in. So I started out by gutting out an old boombox, and putting a Raspberry Pi, a USB powerbank, and a USB charger in there. I also built a power supply multiplexer board that can allow switching from the battery to the charger, and back, without causing the Pi to reboot. And that stage is pretty much where I am. Getting that multiplexer thing built proved more challenging that I expected, but that's a story for another day.
Next steps are to start adding sensors to allow the "AI" to experience the physical world. Plans include:
1. GPS module, so it can experience motion and sense its velocity and location in space, etc. The same board also happens to be a cellular data board that will allow the device to upload data to the cloud.
2. One or more accelerometers for more localized perception of motion / orientation in space.
3. One or more cameras for vision. Ideally two so it can do stereo vision and have depth perception, etc.
4. One or more microphones to receive audio.
5. Anything beyond that is getting into pretty speculative area, but I can imagine adding sensors for temperature, barometric pressure, and other environmental factors, and even get into "senses" that humans don't really have, like sensing IR, ultraviolet light, electromagnetic radiation, magnetic flux density, etc.
Now obviously a Raspberry Pi doesn't have enough computing power to do a whole lot from an AI perspective, so the basic concept is to have the data being received by the box pushed to a remote server, where complex training and reasoning stuff will happen. Modules can then be pushed back down to the box so that some operations can be executed locally. That's where both Wifi and cellular data come into play. One of my "next steps" right now is to get the cellular data board configured, and figure out how to get NetworkManager to seamlessly switch between that and the various Wifi AP's that will be available depending on where I take this thing.
Anyway... there's a lot of speculation to all of this, and a lot of it is still very fuzzy in my mind. But the basic concept is simply to focus on the idea of how an AI can learn from physically experiencing the world in a way that's loosely analogous to the way a human does. If you've read any of Ben Goertzel's stuff and are familiar with his whole "baby AI" notion, then you may recognize some spiritual kinship in terms of thinking. I've been heavily influenced by Ben's work, and by OpenCog[1]. In fact, OpenCog (or some parts of it) may become part of this project before all is said and done.
rasp pi cluster could add more computing power for cheaper.
Well, I don't have room in the physical case for more Pi's as it stands, and TBH, I'm not terribly interested in running a cluster at home. It might mean spending some money, but I'd rather just do it all in the cloud. One advantage to that is I can easily scale things up and down dynamically, and I can use a mix of CPU and GPU (or even FPGA) instances without having to go out and buy a lot of expensive hardware.
Not to say that I would never bring part of it "on prem" and run out of my home, but right now the cloud approach feels more manageable.
* Fixing AnkiDroid to migrate to Google's Scoped Storage model without trashing user data: https://play.google.com/store/apps/details?id=com.ichi2.anki
* Improving the largest bilingual Manx <-> English Corpus + Search Engine. It's recently become the #1 point on the Isle of Man's 10 year language plan, and will probably end up informing a new dictionary for the language
* Generating a machine-translation system using the above, probably also adding Manx to FLORES-200
* (maybe) Generating a Text to Speech engine for another language
* Moving my book over to my new laptop, and grinding it out to completion
* Scraping websockets to improve the accuracy of bus timetabling. Eventual aim to improve the experience of using the bus to reduce the number of cars on the road. Probably some F# webservice handling most of it, with React Native (web)app, distributed via NFC tags/QR code.