What Part Of “Server Too Busy” Didn’t You Understand?

I had a weird experience this morning in browsing Nasawatch (a recommended site, by the way, for those interested in space policy and doings–Keith doesn’t always get it right, but it’s a good source for scuttlebutt that often turns out to be quite correct). I clicked on a link to a story at Spaceref, and got a long delay as it tried to access the server. Like watched pots boiling, watched browser windows never load, so I gave up and switched to a different one momentarily (I run Opera, which allows one to have multiple windows open simultaneously). When I went back to check progress, I saw a simple message–“HTTP/1.1 Server Too Busy.”

I backed up to the main page, and tried the link again. This time, without delay, I got the same message. Theorizing that I was possibly looking at a cached page, I hit the reload button. This time, it came back with the same message, but in a font at least ten points larger–“HTTP/1.1 SERVER TOO BUSY.” I could almost have sworn that it was shouting at me.

OK, OK, I’m sorry. I’ll try later.

Did someone really anticipate that someone might do a reload in such a situation, and deliberately write code that would up the font size? Or has the combination of Moore’s Law and the interconnectivity of the net finally hit critical mass in servers, and intelligence (and irritation with us lower, less-intelligent life forms) is becoming an emergent property?

Theories welcome.