Photos and videos temporarily gone¶

Tags: personal, python

My smugmug account was up for renewal. I used smugmug for my photos and videos. At a very good price of 60 dollars per year. But I stopped my account anyway.

The good thing about smugmug is that you get very good advertising-free quality for a very good price. And I’ve always felt that my photos were safe. The addition of the FileVault (optional amazon S3 storage for your video originals) two or three years ago was very valuable to me: I’ve still got my original movie files this way (I downloaded them all last week).

For me there are three main reasons to leave:

• I’m a programmer and I’d like full control over my photos and videos. More control than I’m now having with an iphoto collection + uploading to smugmug. So I’m currently moving everything over to something home-brewn self-programmed on my own server. This isn’t something that’s got anything to do with what smugmug did good or bad, this is purely a personal preference.

• Definitively smugmug related: I cannot easily say that my photos are creative commons licensed. I could put something in a footer or in a photo description, but it won’t ever be integrated like in flickr. I love to see my photos being used in wikipedia and smugmug doesn’t make them easily discoverable as creative commons. Smugmug looks too much at professional photographers that way (at least for my particular usecase).

• Vimeo’s one-movie-per-page layout beats smugmug. I can send around a link that immediately opens the full-screen movie, but closing that gets me my collection view again instead of the individual movie. Harder to pass links around that way.

So… I’m making something myself. Only, it isn’t finished yet. So I’m temporarily without photos and videos. I’ve got some code on https://github.com/reinout/media-manager . Basically some python code that ultimately will do the following:

• Maintain a structure like ~/media/videos/yyyy/video-title.mp4 and ~/media/photos/yyyy/nice-photo-title.jpg inside a git annex directory to manage the big files.

• Not only the files are maintained, I also have a metadata.json file with the modification dates, titles and so of the videos/photos. Plain json. This also contains album and event information for grouping.

• Allow by-hand addition of video files. This one is already done. It asks for a title, a year and an optional category.

• TODO: a script that goes through my iPhoto’s Albumdata.xml plist file and syncs photos (the ones with a title, a good score, an x/y location) to my media directory. Most of the photos can then be removed from iphoto.

• TODO: a script that generates pages on my server. I’ll use videojs to show the videos. I’ll have to convert them so that I have h264 mp4 files for flash and mac stuff and the webm format for the rest. Doable.

• TODO: security for the non-public files.

The drawback: I have to do it all myself.

The bonus: I can do it all myself. It’ll be forever under my own control. And that’s something I’d like to see this year: relying less on external services. I’ll still use some of them, of course. I’d not go without github for the social coding aspect!

And I’m even thinking about moving some of my public photos back to http://flickr.com because of their great creative commons findability. It will be a selection only: photos which I really want to make findable and useable as creative commons photos.

We’ll see… Until I’ve finished this photo/video project, the website will be missing its photos…