uncSoft
uncSoft on Hacker News
Statistically 0
Functionally, a literal handful per season which are typically caught and prosecuted.
The right leaning Brookings and Heritage foundations dug into the numbers and found this
"Across the entire country, and utilizing a maximally inclusive estimate, between six and 46 cases of mail voting fraud were identified in each general election"
Table:
Year Total mail voting fraud percentage (out of total mail ballots cast)
2016 0.000089%
2018 0.000024%
2020 0.000051%
2022 0.000010%
Average 0.000043%
https://www.brookings.edu/articles/mail-voting-in-the-us-dat...
Posted by
uncSoft on Wednesday September 02, 2026 @04:38PM
from the show-hn dept.
https://github.com/uncSoft/Mousetrapped
JT from uncSoft here, dev of cyberWriter and Anubis-OSSThere is an annoying as heck bug in macOS when you are using a secondary monitor and have other macs around - if you have push through edge to control another mac on and/or screen connect active, often times your mouse will get eaten and disappear completely. Especially if you have a VM running in one of them using Screen Connect. Hotkey or...
Posted by
uncSoft on Sunday June 07, 2026 @01:58AM
from the github.com dept.
I was "bullied" into making a light theme for my app as well after refusing for months - I didn't realize how many people need a light mode.
That's why I wrote cyberwriter, my company needs sandboxed apps because we work on sensitive data. Community plugins are too risky running with full system access
Obsidian is great- it motivated me to make cyberwriter, where the app is sandboxed and the top community plugins are baked in. I work in healthcare and we can't have unvetted community plugin updates running in our network with full system access and with sensitive data.
exactly, but even moreso that Apple exposed the API in a fairly complete way. I integrated their local model in less than a day - creating embeddings and RAG essentially on an entire document vault in about 2 seconds- for people who dont have local AI configured, is pretty crazy
thanks, I am probably the worst designer and didn't want to use AI, but you're not wrong
Posted by
uncSoft on Monday April 20, 2026 @09:07AM
from the show-hn dept.
https://cyberwriter.app
Apple has quietly shipped a pretty complete on-device AI stack into macOS, with these features first getting API access in MacOS 26. There are multiple components in the foundation model, but the skills it shipped with actually make this ~3b parameter model useful. The API to hit the model is super easy, and no one is really wiring them together yet.- Foundation Models (macOS 26) - a...
They just need to call it GW Classic apparently and it will sell
Thanks for the heads up! I just reached out to see if I can contribute, thank you
I build anubis-oss because my friends and I are always trying to see how local LLMs run on our different Macs with different configs. So I build a benchmarker for us with a bunch of tools like exportable benchmark reports and arena mode, then I said hey, I can add public leaderboards with the full dataset of submissions open sourced, and give that to my ML and model tuning friends to gauge performance across broad configs. Yes it cost me money, but it's a fun project and I'm trying to get it into homebrew. Just need a few more stars.
https://devpadapp.com/anubis-oss.html
https://github.com/uncSoft/anubis-oss
https://imgur.com/a/X64WsWY
Ha, wasn't it windows vista that allowed you to plug an SD card to use for swap space/fake ram?
Addendum-
Anubis OSS is GPL-3.0 licensed. Built fully in Swift and dev cert signed for safety (if you don't want to clone the source and compile yourself), no external dependencies except Sparkle for autoupdates if you want them, privacy-first - benchmark data is submitted voluntarily and never includes anything beyond hardware specs and model performance metrics.
Posted by
uncSoft on Wednesday March 04, 2026 @09:44PM
from the show-hn dept.
https://devpadapp.com/anubis-oss.html
Why open source local AI benchmarking on Apple Silicon matters - and why your benchmark submission is more valuable than you think.The narrative around AI has been almost entirely cloud-centric. You send a prompt to a data center, tokens come back, and you try not to think about the latency, cost, or privacy implications. For a long time, that was the only game in town.Apple Silicon - from M1...