Capacitive buttons can get pressed by drops of water, so I personally think the physical buttons are better for a safety system.

It probably depends on what the robot is, but it might not be something you need to hold down for very long, just at the same time as a joystick movement.

With the one-way latency being upwards of 20 minutes, I wonder what the software looks like… They’ve basically got to queue up an hour worth of actions before they’d even see the first response from the rover.

I’m guessing that’s in the Tom Scott video, so I’ll have to watch that later.

Edit: It looks like they’re only using it for manual driving of the test rover on Earth. The autonomous driving is a whole separate system.

It’s basically just all the same info that’s printed on the front. You can scan it yourself with certain barcode reader apps.

A permissive license is a specific category of open source license. As I said, all official repository software is free and open source. If the terms of the license did not allow it to be freely distributed or was a revocable license, it would not be allowed in the arch repository.

All packages published to the arch linux repository are free software that can be built from source and have permissive licenses. Non-free software such as nvidia drivers and other things are distributed separately through the AUR, and aren’t managed by the Arch Linux team.

I would consider the collection of software provided to be no different than any other OS with optional packages. There are several forks of Arch Linux like EndeavourOS that have more complete defaults.

Arch Linux has no EULA because it is not sold software like Ubuntu Pro.

They have a Terms of Service for the online services, which you can be banned from for misuse, but this does not prevent you from using the software. You can legally obtain it through any of the hundreds of mirrors, or even compile it from scratch.

https://terms.archlinux.org/

How to remove all entropy from a sentence…

I had to look these up, and it looks like it’s Adaptive Cruise Control (regular distance based), and Intelligent Cruise Control (ACC + lane keeping). Most likely it switches between if it can’t find the road lines.

I only have ACC in mine, and there’s still situations where I turn it off like in heavier traffic. It reacts too slowly so it ends up braking harder than it needs to, and then waits to accelerate and leaves a huge gap. It effectively makes traffic worse than if I drive manually.

I’ve also turned the lane departure system to warning-only because I’ve had it try to push me off the road before while I was towing a trailer. On narrow windy roads, it’s basically required to hug the lines so I don’t end up clipping trees with the trailer.

I’ve actually seen that video before, and the main limitation is still the pixel transition time. It’s something like 200ms, so any sort of motion will have heavy ghosting. It’s definitely a huge improvement over most controllers though.

If all you do is read/look at static content, e-ink is great, and that’s why it’s used on all the e-readers. I wouldn’t want to be taking a video call on one though.

Tandem oled screens are also starting to become more of a thing, and that improves the usability in direct sun. But I don’t have any problem using my phone outside as it is. I’d rather just have a bigger battery than a worse looking screen for what I do.

e-ink has terrible refresh rates and can barely handle color. It’s not really a comparable display tech to oled.

Just quoting from the introduction:

hive is a formalisation, extension and optimization of what is typically known as a ‘bucket array’ container in game programming

The concept of a bucket array is: you have multiple memory blocks of elements, and a boolean token for each element which denotes whether or not that element is ‘active’ or ‘erased’, commonly known as a skipfield. If it is ‘erased’, it is skipped over during iteration. When all elements in a block are erased, the block is removed, so that iteration does not lose performance by having to skip empty blocks. If an insertion occurs when all the blocks are full, a new memory block is allocated.

It seems a little similar to a std::deque with the addition of flags to mark elements as erased to avoid reallocating blocks.
The main advantage is stable references and iterators with minimal extra allocations/deallocations.

I’ve built similar structures in my own game engine where I keep a sorted list of free indexes, so new entries fill up the earliest free index, and on average the empty entries stay at the end with valid entires staying packed together in contiguous memory at the beginning.

Neat, I might have to set this up for myself. It sounds like it could get around some of the VPN blocking I’ve seen while traveling through airports.

It’s honestly impressive what you can do with under 3W of power these days

To be clear, this is the theoretical minimum amount of energy to lift a mass straight up. I’d be surprised if an airplane moving sideways would hit even 30% efficiency in comparison.

How have I not heard this before… It seems more relevant then ever these days

Now that you mention it, they specify a maximum total weight of 25,000 lbs, so at 100% efficiency it would take 339MJ of energy to lift the fully loaded plane to 10,000 ft. That can be converted as 94 kWh, and at the current cheapest electricity price in the US of $0.1235/kWh, that’s $11.63

Therefore, it is literally impossible for this plane to reach 10,000 ft for $5 fully loaded.

I find the easiest way to think about it is kind of like email. You can have an account @gmail.com and still send mail to other servers @outlook.com. Your posts and comments will be stored on your local instance, as well as any instances that are subscribed to the community. When you post to a community on another instance, I think the text will be stored on both instances, but image uploads point back to the copy on your local instance.

I ignored all the email setup because I disabled signups anyway on my instance.

For DNS on my home server I use ddclient. It supports a ton of different DNS providers, so you can have your server automatically update its DNS record as soon as your IP changes.

I guarantee you the microcontroller that handles the wifi is 10x more powerful than what would be required to just run an offline schedule. There’s no processing involved… It’s not like this is using a camera + AI to see if your cat is too fat.

Edit: I just looked up this model and its actual features… it’s not that far off, since it seems to run image recognition to tell when the cat is eating. If they’re streaming video back to the cloud for processing, that sounds even worse than what I was expecting. It would be entirely possible to run a small image recognition model locally in a product like this.

There are much simpler solutions like the ones that detect a chip in the pets collar (or the embedded id chip most pets have)

‘a mark of belonging within the Cult of Razer’

For anyone interested in this, they sell red flags at Home Depot for a lot cheaper.