Yahoo doesn't spider.
But, aside from that, historically the reason search engines didn't spider across dynamic content was just that -- it's dynamic. There's no guarantee that the string you found on Jan 1 ('widgets' for example) will be there March 15, or even Jan 2.
Some sites made this determination of "scan/not-scan" based on the URL containing a ? in it or not. Others, however, were much smarter, and realized that people were using mod_rewrite, et al, to get around that, and starting looking for Content-Length headers. Dynamic content can't give a content-length header because by the time the script runs to completion, the headers are already long-since-sent. (I say "can't", but there are workarounds for it, which involve sending NO data to the user until the entire page is generated in the server's memory first, but that's cumbersome, slow, and bad user-experience).
Nowadays, though, some sites (google, IIRC, comes to mind) do spider dynamic content. I don't think AV does though. Dunno.
Cheers,
D
