I have never understood the furore that greeted Google’s introduction of the rel=”nofollow” attribute. It seemed to me to be a perfectly sensible way of specifying the provenance of hyperlinks on a Web page. However, many people took it to be Google’s way of penalising contributors to blogs and other interactive websites. Some even thought that Google was guilty of a blanket attack on all blogs. These accusations aren’t true, but the reasons they came up are interesting: Google gave a spurious justification for nofollow; and the name “nofollow” itself is misleading. Actually there are good reasons to use nofollow, and if it had a better name it would be more widely used and less controversial.
Google’s first mistake was that they presented nofollow as an anti-spam measure. Since it obviously would do little to discourage spammers, it seemed as if Google were hiding the “real” reason. Many people noted the downward effect that nofollow would have on blog PageRank and decided that Google were trying to kill blogging. Of course, a moment’s reflection shows that the real reason for nofollow is that Google was merely trying to do what they always do: improve the quality of their search results, which ensures their reputation remains strong and they continue to make lots of money.
The second mistake was more simple: the name. “nofollow” emphasises the fact that Google would not follow such links when calculating PageRank. This gives the impression that Google would not follow the links at all, and would somehow marginalise the linked websites.
What nofollow really is is simply a way of identifying links that were not created by a website’s editors. Google may choose to use this information in its PageRank calculations, but that’s not the only possible use. For example, Google have said that they apply ranking penalties to websites that link to known “bad neighbourhoods”. If a spammer adds a link to his bad neighbourhood on my blog, then the link will be marked with nofollow. This may not deter the spammer, but it will prevent Google from penalising me for linking to that bad neighbourhood.
Another example is that a website backend or fancy CSS could present nofollow links in a special colour, serving to show that they were not created by the website editors, but instead contributed by a reader. This gives a clue to what might be a better name for nofollow.
“Contrib” would be a good name for the attribute. The meaning is clear: a link with rel=”contrib” was contributed to the website by an external party. The abbreviation “contrib” also has a long history of use in computing, so it has the weight of tradition behind it. Unfortunately the people at Google, who probably spent many weeks or months fine-tuning their algorithms to handle their new attribute, seem only to have spent about two minutes deciding what to call it. Names are powerful things, and bad names are just as powerful as good ones.
I agree “nofollow” is terribly open to mis-interpretation. I initially started looking into it because I wanted to link to a couple of demonstration pages that I didn’t want indexed.
I like the idea of “contrib” or similar, it adds symantic value to the relationship, rather than creating an instruction for search engines.
I agree with the points you have raised in your post, but I feel you have missed the big picture. Off all the different kinds of pages on the net, the only population that consistently uses nofollow for (nearly) every link on every page are blogs. Nofollow’s defacto use on all blogs as a deterrent to comment spam has not only failed to stop comment spam, but it has also decreased the pagerank love spread out between blogs interlinking effectively giving all the other websites who don’t use nofollow the edge in SERP game.
Not only blogs, but also forums, etc. use nofollow only for externally-contributed content like blog comments and forum posts. Obviously I can still link to whatever I like from my own blog and nofollow will not be applied, so there is still “pagerank love spread out between blogs interlinking”, but it’s now only controllable by the actual author of a blog, not by any random person who happens to leave a comment. There is no reason why me leaving a comment on your blog should somehow make my blog more relevant for search queries. Nofollow helps to prevent this anomaly. It doesn’t disadvantage blogs; rather, it simply removes the (unfair?) advantage they used to have.
As I have argued ad nauseum, it is plainly obvious that nofollow was never intended to stop comment spam; it was intended to negate comment spam’s effect on SERPs.
Finally, when I hear people say things like “pagerank love” it makes me think that perhaps they are missing the big picture. PageRank isn’t some kind of “warm fuzzy” that bloggers are supposed to bestow upon each other; it’s simply one component of one company’s proprietary search and retrieval technology. These things change all the time — nofollow was just one of the more public changes. I bet the PageRank calculations two years from now will be very different from two years ago.
“There is no reason why me leaving a comment on your blog should somehow make my blog more relevant for search queries.”
I can see your point, so what if I left a comment here — the keyword ‘maxpower’ doesn’t need a SERP boost (or perhaps even deserve one) because I made a comment on this blog. I agree, so what? But what if I left a comment on your blog, as I do now, that mentions that there is a great resource called NoNofollow and it’s all about NoFollow, a subject discussed most prominently on this very page. I think its not only very informative, but also highly relevant to this discussion. Doesn’t that page deserve some pagerank love (a term created to explain the complex calculation that is pagerank and not out of ignorance — as in missing some kind of undefined picture) because it’s mentioned on a page all about NoFollow?
But, because this is a blog and not another kind of website — it gets no boost as NoFollow is applied by default. That doesn’t seem right.
“…it’s [pagerank] now only controllable by the actual author of a blog, not by any random person who happens to leave a comment.”
I have a blog and I control everything. Every detail. I decide (as you do) what comments get (or stay) published, what those comments say, and where they link to. My blog is not a democracy. Therefore, I argue that to claim that “any random person who happens to leave a comment” is somehow usurping your position and taking pagerank (or using pagerank without your permission) is false. A blog owner controls all facets of their online website including other people’s comments — that makes him/ her responsible for those comments. NoFollow robs blog owners of this responsibility.
The basic premise of your argument (if I may, and I could be wrong) is that nofollow is a good way of linking to stuff your site doesn’t want to be associated with, negates comment spam, and therefore it is good.
Akismet provides some spam statistics. These stats show that spam is increasing. Spam hasn’t dropped since nofollow was introduced, so the only good use for nofollow is its use to link to undesirable or unreviewed (by webmasters) websites. In your mind does this lone good use make it a good idea to apply nofollow to all blogs? In mine, it does not.