The approach was wrong. This big, bloated monolith is unpleasant to setup, host and maintain. Django, Postgred, a REST API, OAuth, all wrapped in a lot of docker containers... it's just too much to reason about.
I'm rebuilding the same thing from scratch, with much simpler tech and dramatically reduced scope. Now it's basically a static site generator. Point it at directories, run the command, get a timeline as a static website. The filesystem - my personal files - is the single source of truth. The timeline is just a visualisation of those files.
Best of all, it ships as a Python package, not a set of heavy docker images.
As always, OnlineOrNot (https://onlineornot.com) - I recently integrated with a bunch of on-call schedule providers, so I'm going to spend a few weeks marketing that so my existing users find out about it.
On top of that, looking at adding mutual TLS as a means of letting folks allow-list OnlineOrNot without punching a hole through their firewall.
A tool for scraping websites. Not intended for parasitic web scraping but could probably be used that way. My interest is in stealing video links from sites that try to hide them. I like playing movies and other content on a video player that is external to the web browser. Getting video links is not always easy (Rumble, for example, doesn't try to hide it) but it can be deviously hard to grab from pirate web sites which use many tactics to block access (obfuscation, encryption, special headers, etc.). Sometimes this requires mocking the website (to get past the bot challenge, for example) and possibly inserting code into the site JS to pass information across a websocket to my app.
It's an ongoing project that started with a simple browser-based video player and has expanded to a full-featured proxy server which works with the browser to bypass cross-origin restrictions and other techniques to prevent unauthorized access to a website.
Planning to release the new updates soon (which I have not done for several months).
I'm building an online tool to help folks applying for jobs manage their applications, interview stages, communications, etc. It's pretty common nowadays to apply for many positions, so the goal is to reduce some of the mental overhead and help you focus on what matters - getting that job!
I'm making adjustments to my 6-bit CPU and (re)designing the board that implements the 6-bit instruction set using SOT416's. I worked out a way to map 6-bit sextets to Unicode (jokingly called UTF-6) -- I probably need to document that.
oh. someone upvoted this. i think that means i should document my 6-bit instruction set along with an open license so RISC-V isn't the only open instruction set around.
I would be very curious to look at your 6-bit instruction set. 64 instructions spread across how many general-purpose registers? Maybe 4 at the most plus you may need specialized instructions for process control and housekeeping tasks.
Yeah. The project currently has a CoC, so I'm not sure how interesting it would be to the 8Chan crowd. You might want to look at Chuck Moore's F21 & F32 CPUs. They were stack-based machines, so you didn't get any GP registers, just instructions that operated on the top of the stack. I got a kick out of his moniker: "Minimum Instruction Set Computer (MISC)"
I'm working on a web-based text editor with extensiblity in mind. The source code or website isn't available yet, but I plan on releasing it by the end of this month.
I was working on a BitGrid simulator[1], but I got stuck. I got the emulation running, and at a reasonable speed too. (I can emulate a 1024x1024 bitgrid at 34 Hz on my desktop machine).
I'm stumped as to how I should do I/O. The primary aspect of the bitgrid is that it's an FPGA with zero routing fabric, but clocked to prevent race conditions. This means that results could be skewed. I could either force the outputs in parallel before output, or have the output handler deskew them externally.
It's silly that I'm stumped at this point.
If this idea actually works, you could get Exaflops of performance out of it, but it's not a CPU based system, so no software will port to it.
Airdropped glider delivery drones. Idea is that you can use large carrier aircraft (manned or unmanned) to cover the big middle mile distances and land packages right where they’re needed. The focus is on providing reasonable cost one-ish day delivery to really rural or remote or low infra places, rather than sub-day delivery to urban cores.
I had to design the airframe custom and am almost to test flight since it needs to be ok getting yeeted out of a plane and not endanger the plane. Luckily pixhawk has me covered in terms of a great extensible autopilot so the software side isn’t actually all that daunting.
I’m tweaking a workshop I give to beginners who want to learn React & Next.js. I like thinking about the things I can improve to make the learning experience as good as possivle!
Personal project: I'm writing a web server in C++ as a MIT-licensed library.
Familiy: My dad is a salesman (insurance and investments). A really, really good salesman! He's getting older and wanting to slow down, so I'm helping him put together his own "curriculum" about his non-traditional prospecting techniques that he has perfected and used for years.
I'm studying AI atm, so for my final project I'm writing a booklet/resource about mental health and AI. I'll be illustrating and designing it too. Love this sort of work, because I get to flex multiple skills and use my brain for planning, creativity, research, deep focus and presenting ideas through my personality.
https://nicolasbouliane.com/projects/timeline
The approach was wrong. This big, bloated monolith is unpleasant to setup, host and maintain. Django, Postgred, a REST API, OAuth, all wrapped in a lot of docker containers... it's just too much to reason about.
I'm rebuilding the same thing from scratch, with much simpler tech and dramatically reduced scope. Now it's basically a static site generator. Point it at directories, run the command, get a timeline as a static website. The filesystem - my personal files - is the single source of truth. The timeline is just a visualisation of those files.
Best of all, it ships as a Python package, not a set of heavy docker images.