I assume the real router OS is extremely neutered to basically only route traffic and filter inbound with everything else being removed? But yeah I can definitely see that.
Built something similar almost a year ago during the holidays [1]. Open-source if you want to check it out [2]. I use the mobile app version from time to time when I'm going on walking adventures around the city.
Funny coincidence, I just published an offline infinity-scroll notes app[0] today to replace my long txt file. Desktop version probably in a couple of days. Last time I published an app for myself, my friends (and ~1k others!) loved it so trying doing it again.
I've used a .txt pretty much my whole life from my old Vaio running Ubuntu to my Mac books after, especially as a heavy terminal/nano enjoyer.
I always saved it as do.txt in my base dir. Thousands of lines which was always nice to look at and more importantly easy to reference links I used during debugging or troubleshooting from months ago.
It's a weird mix of a bookmarking list, daily to dos and quickly jotting down phone numbers or details while on the phone with someone (if I'm not on my personal laptop, I usually type the thing while I'm on the phone in the PC browser address bar then copy it over which is not ideal because auto-search).
Another strategy I've used is iMessaging myself with links or notes, which in my opinion is the best way since it auto syncs AND you can pin yourself in the iMessage app for quick access.
Yeah, here's one[0].
Obviously long ways out but using speech to run through scenarios of different simulations from pre-defined parameters would be great. You're basically applying formulas to already designed objects. If the model could run the scenario and spit out a report when it's done with summaries, it would be great since some of these simulations take hours. Then adjust the model for better sim results and iterate accordingly.
Somewhat related, but I'd love GPT enabled multi-physics simulations on objects. Designing in CAD, especially for intricate objects seems (currently) to be better when done by "hand", but I'd absolutely love to use speech-to-text to run different simulations on said objects.
I made something[0] last year to have something very consumer friendly.
Unbox->connect->run.
First iteration is purely to test out the concept and is pretty low power, currently working on a GPU version for bigger models and launching Q4 this year.
I have similar goals, get enough land to do hobby rocketry and get back to long range shooting. Both require a lot of land haha. Bumming it out at publicly available land at the moment. Awesome to hear you're enjoying the hobby with your son!
Always love seeing Ben Eater’s “build [X] from scratch”
His YouTube is one of the best channels to understand how stuff works from fundamental building blocks.
Weird quirk I inherited from Dreamweaver: I used a bootlegged/trial version in high school to learn coding and it would close itself every few minutes so I would Ctrl+S every few seconds to save my progress. I still do that to this day even though all my Jetbrains IDEs auto-save.
I think I still have my webassist.com PHP extensions for Dreamweaver somewhere on my hard drive from 2008.
Inference speed is heavily dependent on memory read/write speed versus size. As long as you can fit the model in memory, what’ll determine functionality is the mem bandwidth.
I mean, not really? It's not like I make money from it, it's completely open-source and free to use. I gain nothing besides sharing similar work to OPs which I found interesting.