Maeyanie.com

In the latest development release of OneSwarm (version 0.6.2) they added another way to find peers: community feeds. Basically, a central server takes public key registries and hands out the few “nearest” peers to you (based on binary search of public keys).

The sample community server is in Java only. My VDS is a bit short on memory to be running random Java programs, so I took a little time and hacked together a simple PHP one. It doesn’t (yet) support some of the important features of the mainline one, like registration DoS protection or IP limits, but you can get the source under GPL if anyone wants it.

One of the latest “privacy-enhanced P2P” things I’ve found recently is OneSwarm. The idea behind it is that instead of using trackers to find peers, it uses an anonymous, distributed, friend-to-friend swarm to find things.

This seems at first glance to be a good idea. Only the people you trust know what you’re trying to find, making it very hard for your favourite government/corporate acronym to spy on you, and thanks to the “six degrees of separation” factor it shouldn’t be hard to find anything you’re after.

Of course, you have to find friends who are using it in order for it to work, but it’s not so bad — they have a handy keyserver where you can give it your GMail/GTalk data, and it checks with a server to get public keys for all your contacts.

Hmmmm… a registry which can link uniquely-identifiable public keys with email addresses? I thought this was supposed to make things more anonymous. That just sounds like a disaster waiting to happen. It does let you directly exchange public keys with friends, at least, which seems a safer route to take.

Still not entirely sure I trust it…

Have slow internet? Use TOR, or another slow anonymizing service? Then you probably get annoyed by Firefox 3 timing out page loads after 15 seconds. I sure did.

Read the rest of this entry »

As at least one of you noticed, my HostMonster account got suspended — again.

Read the rest of this entry »

I like being watched as little as the next girl. But so far, the only Linux-based automatic PeerGuardian blocklist updater I’ve seen has been horribly inefficient, using over 200,000 individual iptables rules. It takes forever to parse, and can seriously slow down a low-end system.

I thought to myself, “here has to be a better way.” And, of course… there is. Here’s how.

Read the rest of this entry »