"nofollow": Google's got it right

TDavid had two good posts about rel="nofollow": No Google juice for nofollow... and Treating all commenters like spammers is a slippery slope. Now I see that Matt Cutts at Google has replied to these posts (and an e-mail), and tells TDavid:

I couldn't agree with you more. I've been asking folks to move in that direction (untrusted people get nofollow, but anyone who is trusted or authenticated via something like a captcha gets full credit for their links). I think LiveJournal has already implemented this philosophy, and I'd expect many other software makers to do something like this.

TDavid comments:

I continue to feel that making this an option in the blog hosted and blog software arena (wake up Typepad, MSN Spaces) is an important move. Down with the bad guys, yes, but let's be sure not to punish the good guys in the process. It is nice to know that Google feels the same way.

I asked TypePad's help service if there is any way I can turn "nofollow" off for commenters and TrackBack-ers I want to trust and so let their links get full search-engine credit. TypePad told me:

There is currently no way to turn it off. At this point, it's automatic on all links. We decided to get it up and running as quickly as possible to short-circuit the ne'erdowells. I will pass the idea along of possibly making a Safe List, where this attribute isn't used, to our Development Team. Thank you for letting us know it's something you'd like to see.

Characteristically helpful and prompt response from TypePad's helpline — and I hope TypePad users get the option as to whether or not to apply "nofollow" very soon.

Comments

[Yaron]

Dave, after reading this, I get a sense of deja-vu from their response.

Much like the responses I got (which I can shamelessly plug here for your readers, knowing I'm not just exploiting the opportunity to raise my PageRank - http://www.ordinarynothing.org/blog/2005/01/typepad_and_the.html ), this one reads like it's the first time they heard about the idea of not globally implementing it.
And in both cases they make sure to show that they don't promise that an option will be given, or that some control method would be provided.

Of course promising exactly what may be a problem, but if they want to do something, they could at least say for sure that they are indeed working on providing some measure of control... They're quite experienced in the field, it isn't likely they didn't think about this at all before starting to implement.

[Yaron]

Oh, my, sorry for the apparent over-familiarity. I just remembered you name as being Dave instead of David for some reason...

[David]

No problem, Yaron!