I bought a Home Assistant Voice Preview Edition to try out. It's surprisingly good, but still falls short when compared to Google Home speakers:
- Wake word detection isn't as good as the Google Homes (more false positives, more false negatives - so I can't just tune sensitivity).
- Mic and speakers are both of poor quality in comparison to Google Home devices.
- Flow is awkward. On a Google Home device, you can say "Okay Google, turn on the lights" with no pause. On the Voice PE, you have to say "Hey Mycroft [awkward pause while you wait for the acknowledgement noise] turn on the lights" - it seems like the Google Home devices start buffering immediately after the wake word, but the Voice PE doesn't.
- Voice fingerprints don't exist, so this prevents the device from figuring out that two separate people are talking, or who is talking to it.
- The device has poor identification of background noise, so if you talk to it while there is a TV playing speech in the background, it will continue to listen to the speech from the TV. It will eventually transcribe everything you said + everything from the TV and get confused. (This probably folds into the voice print thing as well.)
On the upside, though:
- Setting it up was really easy.
- All of the entities I want to control with it are already available, without needing to export them or set them up separately in Google Home.
- Despite all of the above complaints, the device is probably 80-90% of what I realistically need to use it day-to-day. If they throw a better speaker and mic array in, I'd likely be comfortable replacing all of my Google Homes.
Mysa thermostats - don't integrate well with Home Assistant and can't even reliably follow schedule.
Every Samsung phone I've ever owned - great hardware, but the software is a mess, especially with all of the Samsung apps that duplicate the Google apps.
Sony smart TV - was excited about running Android TV apps, but the onboard hardware is so bad that everything lags, and I actually ran out of space after installing 10-20 apps (it only has 4GB of flash storage). Also, its Ethernet adapter has a hardware bug that occasionally freezes up my entire network by spamming it with flow control frames.
I've got this running on several units and it works great. If you buy ESP32 development boards with pre-soldered pins, you can even build the boards without soldering.
Is there any documentation on what Mitsubishi's own MHK2 units can do, how well these libs replicate it, and how third-party thermostats factor into that equation?
I recently had a Mitsubishi heat pump installed, and I'm underwhelmed by the 1999-esque stock thermostat, but the vendor swore on the proverbial stack of Bibles that it was the only one that could properly run the unit as it was designed given all the proprietary control laws and so on.
The problem is, if your landlord ever comes around for inspection, or the bloody thing breaks down due to your installation attempt, you can be held liable up to and including getting evicted.
Where can you get evicted for something like that? The worst case is that they would sue your insurance for damages, or you'd have to pay them out of pocket.
I'm especially curious about which ones are earning "life-changing" money for their authors. I couldn't find an easy way to list Substack newsletters by subscriber count, though. (By revenue would be even better.)
Roughly three categories:
1. Newsletters that cost a lot of money and (claim to) deliver a lot of value, examples being any number of finance/investing newsletters.
2. Massive newsletters with broad appeal. I imagine Scott Alexander, Nate Silver, Matt Yglesias, et al., are doing very well for themselves.
3. Anyone in an industry with extremely lucrative advertising opportunities. And any closely-followed reviewer of very expensive tech products (high sales price * affiliate cut) is probably doing well too.
Roblox | Senior Software Engineer | ONSITE (San Mateo, CA, USA) | Full time
We are hiring systems engineers experienced in C++ to work on the game engine that powers our platform, supporting extensive user-generated content from 2M+ developers. We have three roles we're actively hiring for:
The Engine Networking Team pulls the players together by ensuring the communication of the game state to all. You will help the players experience the game as a nearly synchronous world.
The Foundation team works on the building blocks of Roblox itself -- core C++ libraries. Your work will touch every part of the engine, from threading and memory management to physics and scripting.
The Reliability team works to ensure that the Roblox Engine is as stable, reliable, and debuggable as possible. You will work to ensure that Roblox delivers world-class user experience while also supporting internal developers to root-cause issues and prevent recurrence.
You can install the microcode driver to have Linux provide the updated microcode, if you'd like. (Depending on your distro, it might already be doing it.)
I'm running Proxmox (Debain underneath really) which has a meltdown-spectre checker package available, I followed a debian guide I believe for implementing the microcode and the results are mostly in the green now.
I haven't looked into it but as far as I know I don't think I need to worry about this for the VMs but I could be wrong.
I think the big issue is the different "tuning" for each heat pump system. For example, a heat pump water heater needs a different maximum temperature than an air conditioner, and has much different cycling behavior. This leads to very different design decisions, including refrigerant choice, pressure, compressor type, whether the refrigerant loop is even serviceable, etc. (Some appliances operate at a refrigerant pressure so high that they require being fully assembled at the factory, which would be a non-starter for a manifold system like we're discussing.)
I think this might eventually become a thing once science gets us far enough that there's an "obvious" refrigerant choice for most applications, but we definitely aren't there yet. There are hundreds of different kinds that perform better or worse in different applications.
AFAIK all of the big commercial systems that do multiple different types of heat transfer use water to do it, thus bypassing the entire refrigerant selection issue. Right now the most advanced we can do is VRF ("Variable Refrigerant Flow") systems that can individually select air handlers for cooling or heating (i.e. move the heat from one room to another). These are still commercial units and not really available for residential installs.
- Wake word detection isn't as good as the Google Homes (more false positives, more false negatives - so I can't just tune sensitivity).
- Mic and speakers are both of poor quality in comparison to Google Home devices.
- Flow is awkward. On a Google Home device, you can say "Okay Google, turn on the lights" with no pause. On the Voice PE, you have to say "Hey Mycroft [awkward pause while you wait for the acknowledgement noise] turn on the lights" - it seems like the Google Home devices start buffering immediately after the wake word, but the Voice PE doesn't.
- Voice fingerprints don't exist, so this prevents the device from figuring out that two separate people are talking, or who is talking to it.
- The device has poor identification of background noise, so if you talk to it while there is a TV playing speech in the background, it will continue to listen to the speech from the TV. It will eventually transcribe everything you said + everything from the TV and get confused. (This probably folds into the voice print thing as well.)
On the upside, though:
- Setting it up was really easy.
- All of the entities I want to control with it are already available, without needing to export them or set them up separately in Google Home.
- Despite all of the above complaints, the device is probably 80-90% of what I realistically need to use it day-to-day. If they throw a better speaker and mic array in, I'd likely be comfortable replacing all of my Google Homes.