In this particular case, the web pages include a script to return 403 to normal browsers but not for search engines. I guess search engines are supposed to read the page and process the links. I don't know if they do that for text pages. The instigator has also tried to hide himself from the wiki admins, although it would have been smarter to select more innocuous names for his files.
The IE thing, I think, is different. For that to work, you would want to add the text file to a page and have IE using visitors click the file to view it. I've added a bug report (
FS#1077)