If you've made a backup of all your files, you can browse/filter them with filterous (disclaimer: I made it). It's a shell tool, but it's a lot faster than delicious.com ever was, and can do some searches that Delicious never could. With >13,000 bookmarks, Delicious+filterous have been my most useful knowledge management tools in the last five years.

Now how to get as much as possible of the Flickr metadata out?

Unfortunately, those who have the power to suspend the law are rarely the ones who have to be worried of it turning against them. At least until and unless there's a political takeover by the people.

I hope they didn't patent that, because I've been doing that ever since I realized it was the only sane use of Caps Lock. Not that anyone thinks I'm sane to use Dvorak, but hey...

Analog of course. They've got this new qbits that can have literally any value between 0 and 1!

Show, don't tell. git-gui, TortoiseSVN, gitk, Trac+Git plugin, git interactive rebase. Then demonstrate the awesome speed of a DVCS to someone who does hundreds of commands per day - It'll blow their minds.

the most useful languages I've learned are Bash and Python. Bash for anything shuffling files around a lot, changing access rights, running Git commands in several repositories, and creating symlinks. Python was the first language where implementing ideas outside of Bash's scope (vCard validator, Delicious filtering) didn't feel like a chore. List and text handling is especially beautiful, as long as you don't have to deal with Unicode (I did, and it took quite a while to get reasonable unit tests that didn't fail). 3.0 should fix that though. In short, Python is the most fun language I've ever used.

For everything that doesn't need to be on your own machine, find web equivalents that let you download regular backups: Bookmarks on Delicious, photos on Picasa, blog on WordPress, books on LibraryThing, development projects on GitHub, feeds on Google Reader, and CAD drawings on Thingiverse.

The ultimate tool at home has gone from CVS via Subversion to Git. The learning curve is steep, but it's liberating at the end to know that all the data, in all its versions, are on all my machines and will not get lost bar some really serious happenings. This is for the personal documents, application settings (useful to have the same everywhere) and of course development projects. If you want to forget old stuff, a git rebase --interactive is just the thing. To handle multiple projects which mostly just need to be pulled from a different machine, I've developed fgit, a simple script to run a git command on all repositories below the specified directory (or the current one, by default). Thus, to update everything when moving to a new machine, it's simply fgit pull -- ~.

Yeah, but have you seen the executable?

This is a grave day for democracy, but instead of bemoaning the known failure of the US to uphold democracy, and the known failure of WL being perfect in all that they do, what should be done on a financial level? The first obvious thing would be to spread out the finances - Send some of it to Norway, Finland, Switzerland, New Zealand, and others that have more trusted governments. An investor knows to spread the risk, and unfortunately it looks like Wikileaks will have to learn as well.

More like "There will not be humans then."

His conclusion from the graph is not sound at all - Only a small percentage of the centrifuges went offline at the time. At the same time, they received lots of new centrifuges for installation, so it seems to me that work was simply diverted to installing those.

That's a Chernobyl chicken!

Sorry, hard to measure. Which apparently lots of people think is the same as impossible to measure, or fuzzy (== social scienc-y == boring).

While working at a major research institute library, the head librarian asked a few of us whether we could burn a book. Any book, for example some Office 95 manual that nobody has checked out since way into the last millennium. There were a few muffled answers, which turned into more of a philosophical question: Would you willingly destroy knowledge (or art, as with the stories of religions), knowing that on one side it would make space for new books which we would otherwise have to pass up, and on the other side someone might turn up, wanting to use it for some purpose we didn't imagine at the time of destroying it?