“In an ideal world I would have a database with all the tracks I have ever listened”
This struck me. There’s way too little information about the past available on my computer evben though there’s gigabytes of available space that could be filled. Can I please have a history of all the music I ever listened to, all the videos I’ve watched, all web pages I ever visited, all the background images I ever used. Even my whole GConf history would probably only take 10MB.
Why do I want that? Do you remember you visited this webpage with fotos your friend showed you last month? Or the music you listened to when you were a teenager? I don’t, but why can’t my computer?
9 comments ↓
We had a speaker at school once who told us about life in a POW camp. “I expect you can remember what you had for dinner yesterday. What about the meal you ate a year ago today? We used to practise remembering that when we were in the camps.” I can’t actually remember what I ate yesterday.
Well, last.fm will remember all the music you play, and your browser history remembers all the pages you go to, and so does the Google toolbar if you don’t mind installing that (I don’t think it’s free software, but perhaps I’m wrong). And you can get irssi to log everything anyone says to you, and so on. Of course there are then privacy concerns, because with all this information you need to back it up, but anyone who can get hold of it will know a lot about you and can use that for impersonation / identity theft.
Hey, this thought is kinda what beagle is trying to accomplish in the near (ish) future with its enhanced metadata stores. Its not functional yet, but input about features like this that users might want are awesome! If you wanted to drop a line on the dashboard-hackers mailing list, that would be awesome! If you don’t want to, thats fine. When I get some more free time, I hope to put some effort into some more timeline-based stuff like this after the summer. But if users start asking for it now, the features will come! (and much faster ;) )
Sounds cool, but you would need an “Anonymous” checkbox in the panel to avoid recording stuff you don’t want to go in your history ;). (be it pr0n stuff, or adulterous material to hide from your wife in chat logs or visited blogs)
last.fm
last.fm
last.fm, del.icio.us, Google Browser Sync.
Nice that more people than me are thinking along the same lines. When I tell people about it, they look at me confused. I can’t even make them understand by mentioning usecases, very similar to your examples. I clearly remember rough dates when I did something, saw something, listened to something, but sometimes I can’t remember enough text searchable details to find them.
And like yop says, the more you approach this level of history, the more the need for lying arises. :) If you look at what’s going on with twitter and jaiku and those weird “microblog everything you do all the time” systems, the lying system is actually going to be much more complicated, since it needs to carefully look at all the cross referencing you can do. You have this girl in your friends list and she had a party last friday. You twittered every five minute about where you were, except that time, and you were in the vicinity.
I see this even bigger. I can’t believe that in 2007 there is no generally-used versioned file system, there’s no reasonably used desktop environment with versioning on files and preferences turned on by default.
Sure, many applications generate tilde~ files for the next-to-last version of each file, just in case you accidently overwrote it. And there’s the trash for files you accidently removed. But those are just work-arounds for the most problematic problems created by the lack of history. It’s the best we could do a decade ago, when diskspace was a problem. But now diskspace is cheap.
The type of documents that are likely to get updated often and where versioning would be of tremendous help (letters, spreadsheets, memo’s, source code, config files, IRC logs, music listening logs) use usually limited diskspace compared to the files that rarely get updated (music and video, whether self-generated or (legally?) downloaded).
A versioned file system could also help synchronisation between laptop, home desktop, PDA, etc.
The quote alone is misleading, I should have been more precise in that sentence and write:
“In an ideal world I would have a database with references about all the tracks I have ever enjoyed”
Diff:
+ references about
I’m not talking about the thick music files but about the references track/album/artist that are enough for the brain to remember a song and make it play in my mind. This diminishes drastically the amount of disk space needed.
– listened
+ enjoyed
Yes, there is a lot of music that I wish to forget and keep forgotten.