Yes, but self-hosting does whatever the HOWTO, YouTube vid or AI slop the user follows tells them to do. If the user doesn’t know the basics, how could they know what an instruction for activating UPnP does or opening a NAT port does and why that might expose their data? Laymen don’t even understand what making theie stuff publicly accessible means. It might simply mean “Yay I can access my stuff on the go.” 😄

If on the other had the user learns the basics, they can tell when a doc instructs them to do something dangerous and they can do something about it to avoid disaster.

You bury your lede by mentioning “your project” at the end.

Basically means the user has to trust that project to do the right thing and be maintained to keep their setup secure.

Same. No WAN ingress without VPN.

“If you can’t configure Docker, reverse proxies, and Yaml files, you shouldn’t be self-hosting.”

If this is an example of gatekeeping, I think you are misjudging.

Whenever self-hosting there’s a very real risk of exposing your private data to the internet. Potentially a lot more private data than you’d otherwise expose via cloud providers. This risk necessitates a basic understanding of some of the importand bits and how to operate them securely. If not for that, then anything would go.

Understanding docker, reverse proxy, and YAML which is used to configure those is part of probably the simplest way to get to secure self-hosting. I’d add a self-hosted VPN to access local resources. I’m not aware of a magic UI solution that does it all and securely. Docker compose files are very accessible. A couple of those followed by docker compose up -d and you have a basic env up and running.

Generally the lack of knowledge in X or Y doesn’t mean there’s necessarily an easier path than learning X and Y and that you’re being gatekept by being told you have to learn X and Y. Some things are harder than others. Buying Apple Cloud and setting it up is easier than self-hosting Nextcloud. I don’t think that should be the case, but today it is as far as I’m aware.

Nice.

Internally, and for their shareholders’ benefit alone.

What’s that?

The former.

One of these approaches is much better than the other.

It’s what I did. Or you could keep it on a fixed version.

That’s what I’m waiting for as well. Online meme services are failing.

Always and forever!

😆

Somewhat. I’m guessing this would be proper full OCR, perhaps even selectable. Still scanning.

I don’t see an ability to rotate in the web app. It’s been available for a long time in the mobile app though.

Oh it’s not a problem on ongoing basis. Just the initial scan is gonna take forever here. 150K photos…

The OCR job is running rather slow though. It does about 3 photos per second with 8 jobs concurrency on a Ryzen 7900.

I can still see it in-stock here and there.