the:chris:walker ↩

It's been 3 years...

It’s my daughter’s birthday today. She is now 3 years old.

Un - Be - Lieveable.

So Happy Birthday Florence! It’s crazy you are 3 already and such a gorgeous little girl. I love you.

But I realised today that I started OPFS 3 years ago too, just before you was born.

OPFS is my vision for archiving my photos/videos in a cloud-independent way.

I love Google Photos and have all my photos there. But I don’t want that as the source of truth. I want to backup a copy. Of course this can be done easily, but then I just have a big disk full of photos. Again not too much of an issue, unless you want to find anything.

So enter OPFS, with it the vision that I will be able to have all my photos in google photos, but also keep synchronized a local copy indexed by OPFS, content-addressable so it’s dedup’d (and I can re-import with no fear of duplication). Format is less condusive to searching by hand, but the machine readable indexes (that is, JSON files) could be indexed/searched easily by many FOSS.

OPFS stores its data (the images/videos) in a content-addressable blob store, the index data in another (or the same one). Both of these store can be sync’d with any tool, FOSS or otherwise (I recommend syncthing).

Because I use Google Photos I intend to implement a sync daemon that downloads/imports new stuff and also checks/creates Collections - read Albums that match the Google Photos albums.

I also want to be able to connect multiple google accounts, because Rosie’s photos should be there too.

I have implemented Google Account OAuth for it, so Photos are private by default but can be made public if wanted, or shared with specific people via a shared collection.

Most of the backend functionality is built, the UI is a major WIP, because I keep getting distracted by new tech…

But shit. 3 years I have been working on this. I should have something to show for it by now.

Maybe in another year or two.