Backup ≠ Archive


In late 2001 I created an implementation of constructive reals in Haskell called RealReal.hs. It was on my university account at Berkeley. When doing irregular backups of my data, I never backed up my university account. It was being backed up by the university. I lost my university account in 2004 when I left. I copied all my data to my laptop then. In August 2004 my laptop hard disk died. Recently I have been looking for this file, and I cannot find a copy of it anywhere. I even asked Berkeley if they had a backup; however they don’t keep backups that long.

It’s not so bad. My job is to basically recreate my Haskell library in Coq and prove it correct. Still, it would have been nice to have a copy of my previous work.

I’m trying to figure out what lessons are to be learned here. The advice of regular backups doesn’t seem too be helping people. It’s hard to separate the stuff I create from programs downloaded off the web, and individuals don’t have the money to create duplicates of their entire hard disk.

I did have this file on the web; for years even. But I never linked to it because I was never ready to release it. If I had linked to it, I could have retrieved a copy from The Internet Archive. Maybe I should put anything I want to keep on the web? This seems like it wouldn’t work for emails. On the other hand I seem a bit more successful in archiving my emails.


, ,

Russell O’Connor: contact me