On our product pages we like most have facebook "like" buttons. It's worked great for a couple of years now... visitor clicks like and it posts to their FB page.
Facebook would take the link and when you clicked on it on a persons page it passed through their own php parser page and then would redirect them to our site without the garbage.
Now i'm noticing that there's a bunch of their extra information being passed to our pages. We have our pages set up to 404 any link that has extra stuff to combat confusion with google and to stop some of the spam links we were getting. Seriously we were getting some with querystrings that were pornographic from international locations
What Facebook "like" links look like now is something like this:
A request will only be accepted if it maps to a literal path that exists. Therefore a request with trailing pathname information after the true filename such as /test/here.html/more in the above example will return a 404 NOT FOUND error.
You can only do this if there is no php or similar on your site, because dynamic pages use the extra part. The default setting-- which is officially called Default-- tells the server to ignore anything after static extensions like "html", but to retain it after dynamic extensions like "php".
In any case there's absolutely no reason for them to be putting in that extra stuff. Do they think you have special php routines that do nice stuff for requests that come in with fb_blahblah queries? Are there CMSs that really do do nice stuff? Hm. Oh yes. Option C, if the above isn't possible:
It's not php, it's actually asp but we map everything to .htm files since there were old static pages with that extension. 99% of the files really don't exist... they're all done through IIS and ISAPI rewrite.
Side note: Just checked FB and it appears it's only on FB pages that the link is like that. If you have one of the activity boxes on your site it's not linked that way and they resolve fine. Wonder if this is a FB "error".