I tried using Takeout but the exported file structure and various metadata files were almost incomprehensible. This is likely partly my fault because I didn't download the full 100+ zip files, but I suspect there would be so many files scattered in too many directories to be much use.
As such, Nextcloud doesn't really lock you in; it just provides a framework for the app. You can, theoretically, continue to use Syncthing to sync files while running Nextcloud on top of it (probably not ideal though)
I want to note though, the "no lock-in" philosophy refers more to being able to move out of Nextcloud/Memories at any point if you want. Nextcloud still just stores everything on your disk as folders and files, so you can just decide to nuke it one day and still have everything (not fully true yet, you'll still lose some things like tags and albums; exporting these out too is WIP)
I like the approach taken here. Nextcloud is becoming the defacto open-personal-cloud standard so it makes sense to integrate photos into that. If Nextcloud were getting up to shenanigans in the future I'm confident the project would be forked, and in the meantime I don't expect it would be hard to plug in an alternative backend.
I think for an open-source and/or self-hosted solution to come close to an approximation of google cloud/iCloud/whatever we need projects like this to be able to pick their niche and hyper-focus on it, which leaning on Nextcloud does in this case I feel.
For what it's worth, I think for people like me (who already use Syncthing and Tailscale), all the reasons the FAQ mention for why Nextcloud is really necessary (auth, file upload, etc.) are already covered, which is why I'd be so interested in something a little bit more lightweight.
(As an aside, I am not sure I agree on the "Nextcloud upgrades are seamless" part – every time I've had to upgrade a Nextcloud instance so far I was in for a world of pain.)
Anyway, I wish you tons of success with your project! :) It might be what will push some of my family members to leave Google Photos and/or Dropbox, and that would be a huge win already!
I agree. I used to use Nextcloud and upgrades were a mess. I missed one once and tried to upgrade and the whole thing went nuclear. The upgrade started and failed, breaking everything.
I found have read the note but how can a dev agree to have an upgrade started knowing it will fail and fuck up the db??. Thank you borg for the backup.
It is such a mess in their container that they could not agree on how to get the real IP being looked.
This is really crappy software.
Syncthing is fantastic and just works (right after you get your PhD in syncthingology because, man, it is complicated when you start)
So you saved yourself overhead in lines of code by pushing that overhead to people who don't want Nextcloud platform as slow piece of irreparable bloatware.
Wow, the state of self-hosted photo libraries has gotten way better since I last settled on Photoprism a few years ago. Both Memories and Immich seem very polished. The timeline features look great. I may need to play around with these.
HDR kinda works, depends on the photo. Live photos are fully supported (Apple/Google/Samsung/Xiaomi at least). HDR videos are a pain to transcode, it mostly doesn't work well (but this is not very well tested / worked on). No idea about the spatial stuff.
Only the originals are stored. Transcoding is needed to play any video in browsers, e.g. HEVC isn't supported by most browsers. For this, Memories transcodes your video on the fly and streams it to the browser with HLS. RAW support for photos is easy and already supported, no idea about videos.
As far as "beating" Apple, I'm ready to bet that'll never happen (not just with this project but any really). A small open source project can't really compete with a $2T company
