The author is talking about that they only use Google Maps, Spotify and some podcast app and that those are not available in the car. Interestingly for me only the Google Maps thing is a problem, actually Google Maps doesn't work here in South Korea so I have to use Kakao Maps or T-Map. Anyway music and especially podcast work wonderfully over bluetooth.
I'm going through the list of US presidents https://en.wikipedia.org/wiki/List_of_presidents_of_the_Unit... and take their first names (if there are severeal with the same first name I only take it once) in order from oldest towards today. The latest computer is called Abraham, so there still many left before I have to call one of my computers Donald.
That reminds me of a small project I did to classify drawings of cats and dogs 10 years ago with machine learning https://jeena.net/catdog but no neural networks back then, just things like Canny edge detector, Hough transform, k-nearest neighbor, etc.
Why then is the text2speech in reader mode (which other than that is excellent) on a Linux Firefox so extremely bad? Much worse than Steven Hawkins text2speech.
You can follow other servers and help to spread the load. You can decide how much of your space you want to use for that and which strategy you want to use to cache other peoples servers (new videos, most watched, hot, etc.)
Because if the dev wants to fix the site for your browser, knowing what browser to target is absolutely essential.
In this case, it sounds like the devs specifically targeted Safari to not be supported. You could try to get around this with a user agent spoof, though.
Not that I'm aware of. You'd need to run a reverse proxy somewhere and then point it to your home server as a default but having your other server as a backup running the software at the same time. The big problem is how to store the data on both servers at the same time, especially small texts which are in the database.
Thinking out loud here, but if this is public-facing you can put together a static archive using something like ArchiveBox/SingleFile and have the monitoring server serve the files and update the DNS entry when it detects downtime past a certain threshold.
Sure, it won't offer exactly the same affordance but the OP mentioned rare downtimes like electricity going out (rare here is used generously, obviously some places see this happen often).
Providing a read-only, static version of the services, particularly toots and blog posts, while the origin is unreachable is straightforward and inexpensive.
Somewhat related, the Solar Protocol [0] does something like this to host websites across an array of solar-powered servers across the globe.
http://solarprotocol.net/
Yeah with Lemmy it's pretty extreme. But to be fair, Lemmy is the only fediverse software I run in docker, Mastodon and PeerTube work easy enough without. But Lemmy being written in a compiled language makes it more difficult to get everything installed and set up to compile it. Once it's done it's very lightweight in comparison to especially Mastodon, which is nice.
Yeah my self hosting story started in 2003 and involves moving all of those things to many different servers, etc. I had problems with hacking when I used WordPress, the server has been taken over 2 times. Once I stopped that those problems stopped too. And because I have only 1 person instances I don't get much attention of angry people.
The depricated PHP version is my biggest problem until now, I used a lib written in PHP 4 which are now incompatible with PHP 8 and I can't easily rewrite it to make it compatible. So this one is still running on PHP 7, but I will need to do something about it in the future. The best would be to export everything into static HTML but that's also not quite straight forward.
