The persistence of chain letters fascinates me. We all get them, I guess, and I’m sure quite a few of us never pass them on, despite their promises of bad luck if you break the chain or impossibly good luck if you don’t. Nevertheless enough people do pass them on to keep them alive. I’ve blogged twice about the “Chinese Proverb” chain letter that has been circulating most recently as a powerpoint file:

I also blogged about Pearlas Sandborn:

  • Who is Pearlas Sandborn, For That Matter? which points to mentions of Pearlas in a quite different chain letter, which promises that Bill Gates will pay you for simply forwarding the chain letter in order to help him test Microsoft email software. You might conclude that this email will die now that Bill has retired, but I suspect it will live on.

Take a look at the traffic graphs for those blog postings and they reveal a pattern of persistence. Normal postings (i.e technology postings) to this blog get between 300 and 700 hits unless they’re unusually popular when it can spiral into the thousands. However the pattern of traffic is very distinct, with the posting getting hit for about 3 weeks and then the traffic dying away to a sporadic hit every now and then.

The chain letter articles are completely different to this. First of all they are not particularly popular when first posted, but the traffic just goes on and on.

pd007chainletgraph.gif
You can see this from the illustration of traffic patterns for all three postings. Nearly all the traffic comes directly from Google searches searching on “Chinese Proverb”, “Anthony De Croud” and “Pearlas Sandborn.”

The spikes occur mid-week, with the dips happening on Saturday or Sunday when fewer people care to make inquiries about chain letters. But also notice the long term cycles. The first two graphs show a small peak of activity in January (when I was first sent the Chinese Proverb chain letter) and one which grew through May and June and is now starting to tail off. By contrast the Bill Gates chain letter peaked in February and grew slightly more active in June. I bet these patterns are similar to the patterns of infection of common infectious diseases.

I intend to keep on watching these graphs as they almost certainly correlate to the volume of each chain email in circulation. In the event of new large spikes I’ll let you know.