The --delete flag is already in there so it should empty the camera_backup folder, but the robocopy will copy every single photo back into it the next run.

Ah, I am not familiar at all with Windows, so I honestly wouldn't know. Does the Immich CLI even work on Windows?

But going off by the comments and the parts I understand, I guess it could kinda work, assuming that the arguments are all correct. Although, it will try to upload all photos every time, regardless if Immich already has them or not. Which is not ideal but I believe Immich will filter out duplicated photos. But it's worth checking if it's indeed smart enough to do that.

For the Immich command it will upload the files to an album called "Camera_Backup". Not sure if that's what you want. If not, then remove the -a flag. If you want to upload it to a specific album, then instead add '-A albumname' (edit: I realize I might be wrong here. If you have multiple folders in Syncthing you're backing up it will work differently)

I understand you. I also don't use the auto upload feature as it's creating more problems for me to solve than what it fixes. I already had Syncthing running anyway so I currently use that combo (except I manually sort through my photos on a semi-weekly basis before I upload it to Immich).

If you want, I can cook up a little Python script you could stick into Cron to do all the tasks you described. I haven't worked with the Immich CLI yet but I'm sure I can figure it out. Send me a message if you're interested. I will probably use it myself as well.

I love that idea, and I'd love to implement that. But I honestly can never figure out how people are working with services that enables the user to change settings (for example, to set their location to get their local weather) while still maintaining a read-only system.

Regarding not being able to make new communities: We're all federated. What's wrong with making a community on a different instance with your lemmy.ml account?

How is a websocket being the problem in this?

There's a couple of us!

In practice it can be anything though. It just gets thrown in one of the hashing functions. They way they describe it:

user-key = SCRYPT( user-name, user-secret )

site-key = HMAC-SHA-256( site-name . site-counter, user-key )

site-password = PW( site-template, site-key )

Which is kinda interesting, they mention site-counter and site-template. The counter indicates you can set it to a different value to get a different password for that site. But then obviously every time you want to recall that password, you have to set the site counter correctly. I guess the app will remember this, but the web version obviously doesn't. But the gimmick is that it doesn't store anything, but it seems for the app to work it does need to store the settings to generate your password.

And also there is a site-template, which seems to hold various ways to generate passwords. Long, medium, short, pin, etc etc. With or without special symbols. It even mentions 'saved personal password' so I guess it can save custom passwords? Hopefully encrypted though.

That sorta addresses the concerns I have. But obviously that means you need the app, the website doesn't do all those extra things. And if you loose the app you loose your custom site counter and template.

Also, the CLI version seems to happily store the username and secret on your pc? Or at least lets you read it from a file, so...

(I haven't actually used the app, just going through some docs and source code here)

Same! So much better this way. I wrote my own program so it just automatically downloads new videos from channels I've told it I'm interested in, and then it puts it in the appropriate Jellyfin folder. Way more reliable this way, I don't miss videos. And I don't care if my internet shits the bed, it can take however long it wants to take to download and once it's there I can even watch it even with no internet. I even implemented sponsorblock in it so Jellyfin automatically skips sponsored segments.

That only works if they're actually discoverable when you search for them, so you can pick the correct one. On reddit if you search for a specific thing, you get a list of all possible subreddits in one go. On lemmy... it really depends on the instance you're on, and if anyone else on your instance has discovered the other communities yet. You never know if the search results are complete. Even https://browse.feddit.de is not always showing all the communities I noticed.

And even now I have to keep rechecking communities/interests I have already subscribed too because new duplicates get made all the time with new instances showing up.