I got mine to make a backup copy of the remote controls that I'm worried about losing, which happen to be sub-GHz and infrared.
I got mine to make a backup copy of the remote controls that I'm worried about losing, which happen to be sub-GHz and infrared.
Great statement in theory - but in practice, the whole people-as-a-service industry for AI data generation is IMO more damaging to the knowledge ecosystem than open data. e.g. companies like pareto.ai
"Proprietary data for pennies on the dollar" is the late-stage capitalism equivalent of the postdoctoral research trap.
The document says manual updates are included but not automatic updates (which is just a setting in the App Store that I personally disable).
Whether there will be update notifications is unclear. Is that what you meant by "app notification delivery", or something else?
As an App Store developer myself, I would love to have Tier 1 in the United States, mainly due to no user ratings and reviews. I hate them, and I hate trying to solicit them. As far as I'm concerned, ditching ratings & reviews would be a bonus!
Apple will do whatever they can to ensure that developers that don't pay will suffer the costs.
One game they play is luanti (formerly minetest). I gave them instructions on how to clone the git repository for the game, run the build script, and then start the game. They've probably forgotten the build instructions, but they know to play, they have to type `cd code/minetest` and then `bin/luanti`. Occasionally they have to run `git pull` to update the code on their computers. I handle all the game server administration.
I initially blocked all internet access with a kill switch, but this quickly became an issue because you need to be able to run `apt update` and a few other commands to keep the system up to date. So now I run a proxy server called e2guardian that lets them access sites that I choose.
Later, I introduced them to scratch, and I downloaded the entirely of griffpatch's youtube library with yt-dlp and organized it into folders on each of their computers. I've done the same thing with other tutorial style videos. They don't have access to youtube, and I don't really think it makes sense to give them access at this age.
They run scratch locally, as opposed to using scratch.mit.edu. I enabled the scratch website for a couple of days as a treat for them, and as expected, they spent most of the time exploring and playing others' games, but very little time building their own. I sort of expected this to happen, so we closed off access a few days later, and they took some of the ideas they saw online and started playing with them locally.
So my experience is:
a) linux makes a great platform for kids, since it's very easy to tweak things to stop behaviors that you don't want to reinforce. e.g. `sudo chmod 400 /usr/local/games` turns off all the games, `sudo killall kidname` will close the desktop session if your kid isn't listening when it's time for bed, you can set up time-based login policies with pam_time, you can install your own root certificates for SSL MITM, etc.
b) games reinforce that "computers are fun", and games like luanti are free, open-source, and hackable.
c) interest in games naturally spills over into interest in making your own games.
If you want to try luanti/minetest, I recently cleaned up/released a mod that I built for the kids last year called turtlebots. It's a visual programming tool that lets you program little turtle-shaped robots that can navigate in a minecraft-style world and build things as they move around. Source is here: https://github.com/jmole/turtlebots
In theory, it seems like you could instrument a photovoltaic array to carry some "inertia" with the right control system.
If you need to feed power, you run some power point tracking algorithm, and if you need to consume power, you just overbias the cells and heat them up.
The Gemini name is most likely one of the blandest names picked out of a selection of similarly tasteless possible choices.
It reminds me of "Germany". Google Germany. Can't shake that impression out of my head.
It seems like you could do lots of useful things without having a name attached to any particular sample. There must be some kind of differential privacy approach here that would work well.