More on the Google nofollow tag

A bit of a backlash this morning on the announcement of the new rel=”nofollow” tag that Google et al. are going to be implementing. Most of the critiques are referenced in this Register article.

While acknowledging that the new nofollow tag will not do anything — immediately — to stop comment spam, I think most of the criticisms are off-base in one way or another.

1. The Internet is becoming balkanized, where people only accept content from people they know.

   This may come as a shock to some, but the Net is not a big wiki. Most web content is blocked against content from folks other than the creator. That blogs — and wikis — have the capacity to allow people other than a site owner/administrator to contribute is fairly amazing, and putting some restrictions on that is neither novel nor alarming.

   The “democratic nature of the web” doesn’t mean that everyone gets to post whatever they want on my site. It means that anyone can open their own site, and the web will “vote” on it by who links to it or visits it (and Google will index it regardless).

   I’ll say it again: comments are not the keys to blog content. Blog entries are. Of the blogs I visit, very few comments have links in them to go elsewhere — and, frankly, since those comments will still work, the damage to the “web” seems minimal.

   Now, if someone put the nofollow tag in their blog entries (which (a) nobody is proposing, and (b) the MT plugin doesn’t do), that would be a problem. But why would anyone actually do that, since the blog entry content is already protected? Again, this is not “invalidating” all links, just the Googling of those put into comments. Huge difference.

2. The nofollow tag is “effectively declaring PageRank dead for weblogs.”

   Huh? I don’t know about most blogs, but for mine, there are rarely links in comments (except from spammers) and rarely is it key to the actual post content. Comments in a blog are not really the same as discussions on a discussion board, in a variety of ways. If links coming out of comments are no longer considered in PageRanks or somehow indexed in Google, I don’t think that’s a huge problem.

   And if there’s a really key link that I think, from a comment, should be highlighted and PageRanked and all that, I can (as the original post owner) put it up in the actual post itself as an “UPDATE.” (Conversely, if I want to reference a site that I think is particularly despicable, I can now do so as a link, manually, without worrying that I’m boosting their Google ranking. That’s not necessarily a bad thing.)

   Now, what this does mean is that if I’ve written about subject X, and I comment on someone’s blog (that’s implemented this solution) and say, “I’ve written in more detail about this at [link back to my blog entry],” that cross-link won’t get indexed, my blog page won’t get a PageRank boost, and a search in Google for pages that mention my pages won’t see it. Boo-hoo. Folks can still click on the link. And since I wasn’t sobbing that such a comment didn’t generate a Trackback (in MT, at least), I don’t see why I should be sobbing that it doesn’t boost my PageRank — unless PageRank is what I’m looking for.

3. This does nothing to stop comment spam, but other things like Captcha does.

   There are two ways to disincent behavior. One is to make the behavior more difficult or painful or costly to perform, and the other is to make the payoff for the behavior less attractive. Captchas and blacklists and the like are part of the former, creating armor and fortifications against folks being able to willy-nilly post content. The nofollow tag takes the latter approach, making the reward for actually getting comment spam posted that much less. The two approaches are complimentary, not swappable.

   I’ve toyed with using Captchas here before, but have refrained for two reasons. The first (and, admittedly, minor) one is that Captchas are unfriendly to those with visual handicaps. The second is that, for determined comment spammers, there are ways to get around them, to essentially recruit humans to solve them for the spammers (either through pittance payments to third-worlders or by tying them into pr0n sites where someone resolves an intercepted Captcha to get in to see something, which then lets the site owner get in to where the Captcha was originally formed). I don’t know how common this bypassing is, but it’s at least conceivably possible, and as long as there is a reward for going to the effort, the effort will be gone to. Hence the long-term potential for the nofollow tag.

4. Spammers will just go to blogs that aren’t using the nofollow tag.

   Ultimately, yes, though it assumes that spammers are really looking at the sites they’re going to (and my experience is that they don’t, much). But if they do, that’s still a victory for me, because it means they’ll leave my site alone.

   The same, though, can be said for any defense, whether it’s blacklisting or moderation or ID verification or whatever. Burglars will hit the obvious targets on a street. Virus writers depend on the folks who have no AV on their system and never download security patches. That’s a separate problem (how to encourage updates, what to do about abandoned sites), but it doesn’t invalidate what the nofollow tag can do in this case.

The Register article, aside from taking a snarky attitude toward blogs in general, is missing the point. This approach seems to me to be a good, long-term way to disincent comment spamming, with minimal effect on the blogs or the Net itself. I’m not sure where the problem is with that.