One of the minor annoyances of using Google to search for scientific papers is the habit of some academic publishers of 'cloaking' their PDF files. This means that in the search results these files are listed as if they were freely accessible but when you click through you immediately go to a page demanding payment. You are not even presented with an abstract of the paper so you can judge whether or not you might be interested in its contents. I would have normally expected Google to have removed such cloaked PDFs from their search results, but it seems as if they have made some deal with the publishers which allows this to continue in return for access to the publisher's papers. It all seems a bit clumsily done and smacks of deception and spamming.
Well it seems that quite a few proper scientists are sufficiently annoyed by this cloaking to discuss doing something about it. See The n-Category Cafe, BlogSci, and Science After Sunclipse for more details.