I think looking at the inbound links for deep category pages is a good proxy for how a directory develops its traffic and PR. If a directory got its PR by buying links to its top level, chances are that google can detect that. Matt Cutts has been quoted as saying that google doesn't give much weight to most directory links.
I agree that links from crappy directories probably don't hurt you. However, I think another thing to look at is what the directory links to. If the category page has links to shady stuff, then a link from that page probably isn't work much.
Lets say for example, that I created luggagespamshack.com and did a bunch cloaking, blog spam, link farms, etc, then google would probably detect me as an "unreputable" site. Now lets say genericdirectory.com puts me in their luggage category. I believe that google would then discount the value of the links for everyone on that category page. In other words, if genericdirectory.com links to known garbage, I think google is smart enough to realize there is a lack of editorial control and discount those links, on a page by page basis.
Additionally, I think google is smart enough to weight links based on the context in a page. For example, a link inside a verbose blog post would be worth more than a link on a page full of links with nothing but short descriptions and navigation. Especially if the description that accompanies the link is the same on a bunch of different sites. I think a link inside a navigation section would be weighted even less.
Maybe I give Google too much credit with my crazy theory of relative link value.![]()
Bookmarks