vonnieda

It's hard to do, but ignore the trolls. This is a lesson we had to learn many years ago and it applies today. Do good things that you and others find interesting and ignore the trolls.

If he is 75 and still has to work to get by he has already been robbed of his livelihood. The average lifespan for males in the US is 76ish.

Maybe he just does it for the love of the game, but there are so many old, sick, people in pain who have to labor like this just to get by every day.

Universal basic income is a simple and obvious answer. In a few years we will have zillion dollar companies being run entirely by robots making enormous profits. Share the fucking profits with everyone so that everyone can at least live with dignity.

You know, I hope to hell a robot replaces his job. No 75 year old man, let alone Veteran, should have to do that. A robot with replaceable knee joints is a way better fit.

My sincere hope is that robots can take these painful, laborious jobs and free us mushy humans from them.

I hope they can also take the jobs of the snipers and the sniped.

Psyched to see this! I've been working on my own similar thing that sucks and I don't want to - I want to write a local first music app that syncs across all my devices. This might get me there!

> I think its fascinating just how much of a gap there is between what's being claimed, and the verifiable observable data of the open source world.

According to this list of top Github repos by stars[1], of the top 10, nine of them are informational and one is OpenClaw, which certainly has plenty of AI generated code.

The next 10 include react which has a CLAUDE.md and a .claude, and lots of landed PRs that look like AI. And linux, which, we all read Linus' stance. He sees the value. And superpowers, ECC, and hermes-agent. All of which are AI stuff.

I think it would be reasonable to ask which major open source projects aren't using AI assistance?

> Major open source projects are by and large starting to ban LLMs now, because the contributions made by LLM users have been universally terrible and unhelpful.

That is not why. Or at least not the majority of it. The biggest part is that maintainers just can't keep up. They don't have time to do in depth reviews for the number of PRs coming in to find out if it: works, does what it says it does, meets a need, has docs, is well written, meets criteria, etc. That does not mean all those PRs are trash. Some might be trash, some might be gold. Just like human generated ones.

What it comes down to is vibes. Anti-AI folk don't like AI for reasons (that are valid) and they don't want to find any value in it. So they don't. But lots of people are. Lots of really cool, interesting, clever, new software is being written and getting used and building communities but people see CLAUDE.md and go "slop, no thanks" and won't even give it a try.

[1]: https://github.com/EvanLi/Github-Ranking/blob/master/Top100/...

This feels a lot like what I've been wanting for sketching hardware ideas. Gonna spend some time with this. Nice work!

This is really clever and is taking a direction I hope to see more software taking! Totally self contained is awesome. Well done!

Try FreeCAD one more time, if you haven't tried 1.0+, and it might stick. I've finally, in the past 6 months moved all my work to FreeCAD and KiCad after trying both many times over the past decades.

I highly recommend watching one of MangoJelly's beginner videos for FreeCAD, even if you have CAD experience. It made it very clear how to adapt my Fusion360 skills.

Thank you! The README in the repo was a real joy to read.

What plant software is being shown in the animation at the top of the page? I've been hunting for something to simulate plant growth and that looks interesting!

This looks cool. I'm looking forward to seeing the progress on the GPU acceleration.

I'm a maintainer for openpnp-capture and I've always had the thought of hardware acceleration in the back of my mind, but I didn't really know how I would go about it. Reading the comments on using shaders for color space conversion really opened my eyes.

I like that this is in Rust. It should make it very portable over time and Rust's dependency management and builds should make it a heck of a lot easier to distribute on multiple architectures (if you intend to).

Looks awesome, good luck!

Seconding ILDA. It's been the standard for decades and it is well documented and easy to interface with either at the software/library level, or even at the signal level. [1]

You can get a cheap "RGB" laser projector on eBay for ~200 [2]. It will have red, green, and blue lasers, and the interface will be willing to combine them 7 ways. A more expensive laser will have "analog" mixing for closer to true RGB. It will also likely have no documentation, and might be weird. I have one of these.

For something with better documentation, supported API, and lots of active development, check out Lasercube [3]. It has WiFi, a documented API, a Github repo with a client, and the base software is also really cool. This is my next planned upgrade.

Be careful: Most of these lasers put out 500+ mW and will instantly destroy your eyes if you make a mistake and stop scanning. It's not an exaggeration. Find out what wavelengths your laser outputs and get certified glasses to wear when working on it.

[1]: https://www.ilda.com/resources/StandardsDocs/ILDA_ISP99_rev0...

[2]: https://www.ebay.com/sch/i.html?_from=R40&_nkw=ilda+rgb&_sac...

[3]: https://www.laseros.com/

I am working on exactly this in a music player I am writing. Specifically, you can use any S3 compatible storage to sync your music and metadata between devices, and it's end-to-end encrypted.

I plan to offer single click resale of storage directly in the app for users who don't want to deal with access keys, secret keys, ACLs, and the like.

I made a clock based on an IV-18 vacuum florescent display tube. In the process I learned CAD, CNC, aluminum anodizing, metal turning, SMT soldering, laser cutting, box making, power supply design, and a dozen other skills I use every day today.

Here's the clock's homepage, which has some pictures: https://vonnieda.org/tc18, and more pictures on Flickr: https://www.flickr.com/photos/vonnieda/sets/7215762293577593...

And a blog post that goes into excruciating detail about the entire process: https://vonnieda.org/archives/1406

I built a total of five of them. I sent one to the person who designed the enclosure, one to my brother, one to a friend, one to eBay for sale (which ended up in Germany), and kept one for myself. It still sits on my desk ticking along quite happily. I've changed the backup battery once in 14 years and have made little tweaks to the firmware over the years.

When you "buy" a Kindle book, for instance, you click "Buy It Now" not "Rent It Now". The implication is that you are buying it to own, not renting it. Not a good comparison to renting a car.

"You based your resume tips on a sexual grooming chart?"

"As a joke!"

"I'm gonna go work somewhere else..."

I know cultures are different in different places, but in corporate America this post is all red flags. Always Sunny is hilarious, but I sure wouldn't want to be sitting in HR explaining "the implication" to someone who has never seen the show.

Just FYI what you’ve described is exactly what the GPs video shows. I only mention it cause that knob and video are absolutely amazing and deserves a watch.

I wonder what an LLM would say if it had, say, 2 orders of magnitude more parameters and 46 years of training like I do. And if it was trained up to the millisecond with new inputs from millions of individual sensors. It might say:

> I wonder what an LLM would say if it had...

Wow, I've been looking for something like this for years! Thanks for posting. I fired it up on an M1 Mac (Rosetta 2) and it runs quite nicely. Watching through the screencast now, and looking forward to using this.

I've had an interest in writing an open source 3D CAD for some time, but I would far rather help with an existing project. Are you interested in PRs, and are there specific things you'd like help with?

One thing right away, I'd be happy to help with Mac M1 builds if that would be helpful. I maintain several open source projects and provide binaries for many platforms.

Great work!

I don't think this is true. I see "Boosted" posts in my feed from people I don't follow. Maybe this is a server specific feature? I am using Hachyderm.io.

Ditto - worth every penny!

TL;DR: Zwave and Zigbee are (almost) always local first. Home Assistant is great. Almost all CO2 sensors are bad.

I have the same goals with my setup, and I've been tinkering with home automation for a couple decades now, so I'll share what I've learned so far.

First, specifically for indoor air quality, and especially for real CO2, the options are terrible. eCO2 sensors are a dime a dozen, but they are awful. For real CO2 everything seems to be super industrial or cloud junk.

I'm using a variation on https://hackaday.io/project/186682-co2-buddy for CO2. It uses a high quality sensor and is really just three parts, or two if you don't want LEDs. You flash the ESP32 with esphome and then Home Assistant does the rest. Someone, maybe me, should build and sell these.

For PM2.5 I use a bunch of modified Vindriktning from IKEA. I added a tiny ESP32 board, soldered four wires, and installed esphome. I modified four in about an hour one evening. Someone, maybe me, should build and sell these.

For temperature and humidity I love the "Aqara Temperature and Humidity Sensor". I sprinkle these liberally around the house. They are Zigbee, so local by default. In fact, when shopping for home automation stuff, if you see something that "requires Hub" it usually means its either Zwave or ZigBee and will work just fine directly.

Home Assistant for control, data storage, graphing, dashboards, etc. It's free, open source, gets high quality monthly updates, good looking and responsive iOS and Android apps, great web interface, and most important to your question: has extremely good documentation when it comes to what is supported.

Take a look at the Integration page for Phillips Hue, for instance: https://www.home-assistant.io/integrations/hue/. It's marked as "Local Push" which is the keyword you want for stuff that doesn't need someone else's cloud.

https://www.home-assistant.io/blog/2016/02/12/classifying-th...

So, when I am shopping for a new class of device, I always first check the Home Assistant integration page for it and see what the support is like. There will often be helpful comments and FAQs on the page.

In general, if something is Zwave or Zigbee it will be trivial to get it working local first. If it's WiFi there's probably going to at least be an app and a cloud that you have to use to setup and it's a toss up whether you can then use it without a constant cloud connection.

If you've made it this far, here's some other recommendations for solid local first Home Assistant compatible goodies:

- Hue lights. I've tried Feit, Sengled, LiFX, Hue, and various big box store white label junk. Hue just works. They are expensive, but they work great and I never, ever have a single problem with them. Phillips doesn't make it obvious but Hue's hub is just Zigbee and you can talk Zigbee directly from Home Assistant to Hue lights. You can do the initial setup over Bluetooth using their app and then never use it again. LiFX are WiFi and are a solid runner up, but basically the same price.

- Zooz wall switches and dimmers.

- Zooz 4-in-1 Sensor is a great temperature, humidity, illuminance, and motion sensor. I use these in places where I want motion sensing in place of the Aqara temperature and humidity.

- Zooz Water Leak Sensors.

- Zooz window and door open sensors.

- Bond Bridge for 400MHz RF remotes - ceiling fans in my case. Like Hue, set it up using the app, then just run it with Home Assistant.

- Minoston Z-Wave Plug with Energy Monitor for plug in switches with energy monitoring.

I'd love to see an answer to this, too. Just to learn a little bit more about the field!

> One other detail to note is that we built our software architecture in such a way that we can stop traffic to our PostgreSQL databases, queue the transactions, and run a PostgreSQL update without affecting payments acceptance.

I agree completely. I am completely willing to spend real money on an iOS app that is as intuitive and powerful as a TI-89 but it just doesn’t seem to exist.

The author is referring to writing a UI library, not writing an application or UI using an existing UI library.

Are you saying you would hit the physical switch and it wouldn't activate? I don't think I've ever experienced that.

Early on, I would have command failures when the dimmers were powerline only, but with the dual band ones (RF mesh + powerline) that they have been selling for the last 5+ years I don't think I've ever had an error.

I use Insteon dimmers and switches in my house. These replace the wall switch but look basically like a normal wall switch. They use a combination of power line networking and RF mesh networking for automation. This is the second house I've had entirely outfitted with them and I love it.

IMHO it's critical to maintain control at the wall switch - if I am walking into the kitchen and want to turn the lights on, I just hit the switch like any other house. But when I'm in bed and ready to go to sleep I can press one button in the app and all the lights turn off.

Anyone with basic electrical skills can replace a switch in about 15 minutes. I did them all myself in my first house over the course of a few months, and when I moved I hired an electrician for a few hours to just do them all.

The switches and dimmers are about $50 each, but they are highly reliable. I've had one failure out of about 40 total installed over the course of the 12 years I've been using it.

That's such a strange coincidence! Do you still use it?

It still uses Mongo on Heroku for now - the Postgres changes haven't been pushed yet.

And yes, good point on the README not explaining what it does. Thank you. I'll fix that.