deivid

29 ms·

61.
by deivid 1y ago
Maybe the memory of my Pi2 is better than reality. I have a VisionFive2 as part of my lab (next to some arm and x86 sbc), with an nvme disk, and it's unbearably slow. It's still useful for testing platform compatibility, but I wou
62.
by deivid 1y ago
Can't beat that price, just be warned that it's slower than a Pi 3, probably similar to a Pi 2.
63.
by deivid 1y ago
An offline-only translator app for Android, using Mozilla's translation models (on device) and Tesseract. App is alpha quality but working: https://github.com/davidventura/firefox-translator Trying to figure out F
64.
by deivid 1y ago
Can't wait for a faster risc-v SBC, waiting for rust builds on my VisionFive 2 drives me crazy.
65.
by deivid 1y ago
Interesting! I had the opposite idea: to convert arbitrary SVG to CSS..
66.
by deivid 1y ago
> CosmicPlay: A startup that leverages GPU emulation and graph theory to build a next-generation video game engine that can render celestial bodies with nuclear propulsion and solar sailing capabilities, while generating AI-powered gamep
67.
by deivid 1y ago
Booting Docker images is fairly straightforward, I wrote about how to do this manually some years ago: https://blog.davidv.dev/posts/docker-based-images-on-baremet...
68.
by deivid 1y ago
You can use bergamot ( https://github.com/browsermt/bergamot-translator ) with Mozilla's models ( https://github.com/mozilla/firefox-translations-models ). Not the easiest, but easy enough (r
69.
by deivid 1y ago
You can add '-ai' to your google search to disable the AI summaries. The rest seems here to stay. I've gotten popups offering me to summarize 1-line emails on Gmail..
70.
by deivid 1y ago
Cool project, but I'm missing what does this "remapping" do that the glasses can't do by themselves? I see the glasses are connected (via usb-c?) to the OrangePi. Is this project converting HDMI signal to the Viture prot
71.
by deivid 1y ago
Can you sell these assembled? Using Tang nano for learning right now, but the tooling and docs situation is not great
72.
by deivid 1y ago
> It’s really, really easy to get a better code sample than the one shown here. The problem is that in a team setting, you can't control the type of "code sample" that your teammates send for review. It's frequently t
73.
by deivid 1y ago
I'm using this to run on a risc-v board I have at home, it's pretty nice. My action just calls `make build test` but it is convenient to have it hooked to GH for automated testing on PRs
74.
by deivid 1y ago
Nice! I built the same thing a few years ago: https://github.com/davidventura/on-air It supports mac & Linux, and gets sent to an ESP32
75.
by deivid 1y ago
How big are the models that you use/built? Can't you run them on the browser? Asking because I built a translator app[0] for Android, using marian-nmt (via bergamot), with Mozilla's models, and the performance for on-device i
76.
by deivid 1y ago
SVG+CSS is super powerful, a simple feature that I love is making diagrams respect dark/light mode, without any javascript. Check the diagrams here for example: https://blog.davidv.dev/posts/ipvs-lb/
77.
by deivid 2y ago
For all paid users _in America_. It's not available for me in Europe.
78.
by deivid 2y ago
It's a fun pasttime. I'm rewriting mdadm in rust: https://github.com/DavidVentura/mdadm-rs Mostly, I am tired of tools requiring root access, or a block device, to function, even in read only mode. If you hav
79.
by deivid 2y ago
I've tried goaccess and goatcounter, finally settling for umami.
80.
by deivid 2y ago
You could connect airplay to snapcast[0] then use an embedded snapcast player, like mine[1], I do this on an ESP32 with 320KiB of SRAM [0]: https://github.com/badaix/snapcast [1]: https://github.com/dav
81.
by deivid 2y ago
I struggle a lot with this, I keep a blog and every time I start a new post, I get the feeling that it should be "good" (for some definition of good), which makes the writing of most entries to drag for a week or so. On top of tha
82.
by deivid 2y ago
> discoverability issues Webrings[0] are somewhat being used again. I keep a list of the blogs I follow[1] in OPML & HTML, so that you can either bulk-subscribe, or browse through blogs that you might find relevant; you can do the sa
83.
by deivid 2y ago
Are there any "good" OCR models that run in restricted/small environments? I'm thinking about local models for phone-sized CPUs Obviously these models would have lower accuracy, but running at all would be nice.
84.
by deivid 2y ago
Amazing! I'm interested in models running locally and Kokoro seems amazing. Are you aware of similar models but for Speech to text?
85.
by deivid 2y ago
This is a really nice write up! It's giving me motivation to go back to my WASM implementation
86.
by deivid 2y ago
"feat: Enhance Auto-Commit Bot with new features and improvements" It's definitely _a_ commit message. Not sure it beats "asdasd" or "do stuff" though.
87.
by deivid 2y ago
Excited to try it, but I don't have a goodreads account. Could I import data from my calibre library directly?
88.
by deivid 2y ago
Some jobs at HFT deal with a lot of this fun--I was doing Linux at one for 6 years. A lot of jobs are extremely mundane though, compliance, regulations, legacy code bases, etc. Yes, all mitigations get disabled
89.
by deivid 2y ago
The Meta Rayban glasses have "bone induction headphones", but no screen; so it should be doable
90.
by deivid 2y ago
This looks like a cool approach, but seems like it needs some more research vs the micro OLED devices. I'm super excited for smart glasses, but haven't seen anything yet that offers both a Heads Up Display (aka screen) and audio&#

More ›