Google Blogoscoped

Thursday, July 24, 2008

Knol’s Nofollowing Of Links

Knol is a publishing tool but unlike some other such tools by Google, this time all your article’s outgoing links will be “nofollowed.” The nofollow attribute is a mechanism to disable the juice a link is sending to another site. So while you can link to Knol pages to give them more authority in the eyes of Googlebot and others, Knol pages do not pass this authority measure back to other sites.

In a Knol article, at first glance, you won’t see the nofollow attribute if you look for it in links. There’s also nothing in the robots.txt which would prevent spidering of Knol content, and in fact, filling knowledge gaps in Google results was one of the proclaimed main aims for Google to build Knol. But you will see the following meta directive on top of the HTML page:

<meta name="robots" content="index,nofollow" />

So nofollow, originally introduced with a focus of preventing comment spam, continues its expansion into other Google areas. Wikipedia already does such disabling of linkjuice for external links since some time. Let’s see how Google originally suggested to use this attribute in their 2005 announcement:

We encourage you to use the rel="nofollow” attribute anywhere that users can add live links by themselves, including within comments, trackbacks, and referrer lists. Comment areas receive the most attention, but securing every location where someone can add a link is the way to keep spammers at bay.

Whether nofollow would be justified in Knol according to this guideline depends on the perspective. From Google’s perspective, users can add links by themselves, so nofollow may be appropriate. But from the author’s perspective, other users can not add links by themselves – if the article is set to moderated or closed – so such usage may be inappropriate. What does Google’s Knol say on the subject? Their content policy seems to be clear on this issue, stating “We respect our users’ ownership of and responsibility for the content they choose to share. “ Their terms of service argue “Google does not monitor or edit the content of knols, and takes no responsibility for such content.”

In a tool like Blogger, Google assumes the perspective of the user, but in Knol, they apparently look at it from a different perspective, despite their policy claims. Perhaps the main aim of this is to prevent a flood of low-quality Knol spam articles just intended to pimp someone’s pages. But as far as the quality of search results is concerned, a blanket disabling of link juice in publishing tools isn’t effective – because both “good” links and “bad” links will be disabled. If a couple of experts link to a great new site in their Knol articles and that site is rather low-ranked in Google so far, then Knol won’t help to push it further to the top. (Provided Google always includes the nofollow declaration in articles, that is. It may also be the directive depends on factors like the article’s age, or average rating, in a kind of guilty until proven innocent approach. If the current high-rated and frontpage-featured article “How to Backpack” is any indicator though, these things currently don’t matter, as that article also includes the “nofollow”.)

And after all, spammers can use any other publishing tool if indeed followed links in themselves were an effective spam tool... but in themselves, they’re not, because first your page needs rank authority in itself to pass authority. Or is Google scared that Knol articles receive a higher ranking on average as they’re so close in the vicinity of well-ranked Google? For instance, the Knol frontpage contains a selection of “followed” links to articles (I’m not sure in how far these are moderated or dependent on factors like ratings). As the Knol homepage itself can be expected to be linked from all kinds of different, high-ranked Google properties, it too may receive a high PageRank in the future. (The current PageRank 9 link from the official Google blog is a start... or just imagine they put a Knol link on their PageRank 10 homepage.)

What does Google think, will Knol articles receive a rather high organic ranking (to leave the issue of potential future special rankings or rank formatting, like through a Google one-box, or the display of eye-catching stars next to a result snippet, aside for the moment)? Danny Sullivan of Search Engine Land in his coverage says, “Google assured me that the authority of Google’s domain wouldn’t give Knol any additional trust. Knol pages will be scored based on the links and PageRank pointing to individual pages.” In Google’s Knol pre-announcement from 2007, Google’s Udi Manber wrote that “participation in knols will be completely open, and we cannot expect that all of them will be of high quality.” He added that Google’s job in Search Quality “will be to rank the knols appropriately when they appear in Google search results.” (Search Engine Land also asked Knol product manager Cedric Dupont “if spam isn’t kept in control, could Knol find itself banned on Google?” Cedric claims yes. It’s hard to imagine though how Google could possibly be completely neutral in the often manual decisions of which sites to ban.)

All of this taken into account brings up the question: if Knols are treated like any other page by Google, why would there be a need for a blanket “nofollow” approach to the content, then? Google can’t be possibly suggesting a general nofollow is the appropriate way to publish content on all web pages out there (which are also all “user generated” content in some way), or can they?

[Thanks Ionut!]


Blog  |  Forum     more >> Archive | Feed | Google's blogs | About


This site unofficially covers Google™ and more with some rights reserved. Join our forum!