I have at least 50 internal links from my own search results. I'm seeing a giant list in the webmaster console of things like:

• http:// blog.forrestcroce.com/index.php?s=abundant
• http:// blog.forrestcroce.com/index.php?s=aperture
• http:// blog.forrestcroce.com/index.php?s=arboretum
• http:// blog.forrestcroce.com/index.php?s=ascent
• http:// blog.forrestcroce.com/index.php?s=autumn
• http:// blog.forrestcroce.com/index.php?s=avery
• http:// blog.forrestcroce.com/index.php?s=avoiding
• http:// blog.forrestcroce.com/index.php?s=basin

And so it goes through the alphabet. These aren't links; I've never linked to any of my own search results, and as far as I know, nobody else has, either. It's possible someone liked a post of mine and linked to a search for the broad subject ... I could see one or two searches being indexed this way. But not 50 or more.

All of these are single-word searches, which seems like the kind of thing a bot would do. I've spot checked a few in the list, and Google doesn't show any matches, at least for any of what I've tried.

If search engines can't post to forms, how would they know any of these pages exist? The last found dates range from Dec 9 to the 21.

Your search form doesn't explicitly have the "post" attribute set to it. As a result, when I do a search for "totally random picture and some junk", I get this in my URL:

http://blog.forrestcroce.com/index.p...+and+some+junk

You could either fix the form so that it POSTs instead of GETs, which probably makes things messier somewhere else thanks to WordPress (I haven't tried this myself), or you could add a line like this to your robots.txt file. Personally, I'd rock the line science. It's much easier:

Disallow:  *index.php?s=

http://www.mattcutts.com/blog/googlebot-keep-out/ <-- for more info.

It looks like you're right. I never stopped to consider a form could be coded this way and work. Still ... how does Google find out about a big list of searches? They're not submitting the form ... if somebody decided to post a bunch of links to meaningless searches, it's plausible, I guess. Maybe if the anchor text were just the search word, it wouldn't show up in a search. But if that's the most likely answer ... how does it make sense to link to a bunch of valueless searches?

there's probably a scraper/splog/MFA piece of crap somewhere that does random word searches on wordpress driven blogs and publishes the results.

No one ever said spammer behavior had to make sense, Forrest.

Like Hirst said, there's probably some piece of crap "software" out there that likes to go around exploiting search forms, finding results, merging them together, and creating a "Search Results" page on some random domain name like abc123wxyandz.com .

Hi Forrest -

Are you using WordPress? If so, you might consider the selective No-Follow plugin.

Yours -
Scott