Inpired by Lars’ excellent photo tagging, I have decided to shine up my photo collection. Since mono is much more integrated in Fedora Core 5 I decided to try and use of the fancy new C# applications I have been reading so much about. I had already read some good things about F-spot so I decided to try that out. I downloaded it easily using yum and started it up. F-spot has a very nice GUI and I had quickly imported about 1800 photos into it. Importing photos however, is a small part of the job and I have now embarked on the enormous task of tagging all the pictures. So far I have 32 different tags, and that number keeps increasing as I tag more and more pictures. When I have tagged all of them I have some more pictures still waiting for processing :). This is tedious work, but it should be easy to keep update when adding new pictures. I really like the concept/idea of tagging pictures since I can use all the tags I want, it is a much more convenient way of working compared to just having the pictures organized in the ever-so-common “albums” where you have to limit the pictures to a single album.
F-spot can export its database to varius web based formats, but do not expect to see all my pictures online anytime soon. First I need a place to put them (it’s more than one gig) and I might need to apply some censorchip as well :)

I have been trying to recover some of the data from my formatted partition these past days. Since the partition has been formatted I have given up on ever recovering the emails and other kinds of data with non-trivial file layouts. I have instead focused on recovering the secret key to my gnupg public key for one of my email addresses. By playing with hexdump on secring.gpg files gnupg generated for me I found that these always started with two octets (0×01, 0×95). I then made a simple scanner scanning the device (/dev/hda6 in this case) for these octets. Each time this header was located it was written out with the following bytes making a bunch of small files of 4000 bytes. I then wrote a bash script using a simple truncator program I wrote to actually run gpg –import on each of these candidates truncated to all lengths between 100 and 4000 bytes.

However, I was unable to recover anything :|. I had already filled around 2 gigs of the 16 gigs making up the partition and by doing that I might have owerwritten the key data, bollocks.

David from my dorm made me aware of the movement to ban DHMO, I am not entirely sure that would be a good idea. I found a FAQ about DHMO.