Good morning!

Most of you have probably heard about this, but I haven't found a thread concerning this here yet even though I searched a lot.
Google introduced a new way to prevent link-spamming.
By link-spamming they mean people or bots who purposefully link their websites in dozens of comment sites, wikis, guestbooks etc to increase their page rank. It's common among those who also spam emails so you can guess who: penis pumps, discount drugs and ways to become a millionaire.
Google has had trouble with these fellows because this is hard to control.
Now they decided to implement a new standard (very interesting that Google is setting the standard but Yahoo Search and MSN are just going to obey it.), that enables webmasters to mark links as worthless.
This way the links don't get an increased page rank because of the link (it doesn't decrease either though).
This topic has been discussed a lot lately especially on Wikipedia, another standard-setter on the internet, (The Wikipedia:Nofollow Page is a good example). Right now all links on the Wikimedia projects are marked no-pagerank except the links on the English Wikipedia (with the explanation that link-spamming would be noticed quite fast on such an active project).

The way it works:
The attribute is standard in Strict, Transitional and Frameset.
It works in the following elements as far as it is compatible with the W3C specifications: <a> (normal link), <link> (Multiply applicable header tag)
That leaves the tag <area> which specifies links in imagemaps and the tag <base> (well it would not be a of use there anyway) out, because they may not have the rel-attribute. I don't know if Google parses them all though, if anyone knows: tell me! The way I understood they just use it for simple links à la <a> only, because it was invented to prevent spamming blogs.
Let's get to it: Blog owners and others are invited to implement user inserted links in the comments sections etc. this way from now on:

So you can just put in in all link tags that you wish to not gain any page rank. A quite powerful instrument.
According to Google this is to prevent comment-spam in Blogs and the following Big ones have already signed on:

The nofollow value for the rel attribute is also recognised if seperated from other values by punctuation (preferably standard-conform spaces).
The rel-attribute was originally meant to show logical relationships between pages (together with its little buddy rev. rel was for logical forward-relations and rev for logical backward-relations.)
The following values were possible

But the nofollow tag is still valid. This excerpt was taken from the W3C specifications. Also interesting: How they express the use of rel/rev

Well, so Google has been screwing with standards, just gave webmasters a powerful weapon to kill others pageranks and screw with Google's system and nobody is talking about it here? WHAT'S UP?
I hope this incites a discussion.

Best regards,
Ruben

This post has been edited by ruben: Mar 30 2006, 10:07 AM

I have been shouting for long enough to all the idiotic Google lovers that the moment someone gets power, it tries to exercise Monopoly.

Although yes, this tag can create a problem for the "legitimate" websites, but i think in case the rel="nofollow" will be checked by the google spider to be linking to a page from the same relative path, then it wont be that much of a problem. In short, the rel tag should be read only for a link to the webmaster's own page.

For example, I can block my Guestbook entry links from being spidered, but i can't stop a link to, say "http://someone.com/show.php" being linked via "http://mypage.com/article.php". Google can compare the relative domains in this case. In other cases, relative paths can be compared.

I support this technique, provided the rel tag is checked against the link it is used for.

I really don't get what you are trying to say. The rel="nofollow" can be applied to any <a href> link on any website. It does not mean that Google won't spider that site, it just means, that it doesn't increase its pagerank.
You can stop all or some of your own pages from being spidered by using the robot.txt, but that's really old news. If you stop your guestbook from being spidered, that's a way too, but it's not the same thing. You would never want to do that for Blogs with comments on them for example, because if you don't let Google spider the comments on the page, using robot.txt you can only exclude the whole entry. That's what the rel="nofollow" is made for, but it can be abused too, to make legitimate links not heighten their pagerank accordingly!

Uh... sorry for the garbled post. Initially i got the concept wrong. Now I have it straight.

Like you said, it can be abused for the decrediting of legitimate links, and I find your argument correct. And for that reason I suggest, let Google observe a rel="nofollow" link for a length of time before giving due credits. What I mean by this is, rel="nofollow" shouldn't mean that the credits are not given at all, rather suspended. By doing that, it can wait for the website owner to remove spam links, and also finally give due credits to a legitimate link after a period of time when it notices the link is still existing.

I have made a similar suggestion at the wikipedia discussion. See the tech is needed for sure.... the question is how to prevent its misuse.

Yeah, in my eyes Google isn't there to give a tool of judgement for link-legitimacy. I mean this is definitively going to be used to cheat on link exchanges and ****. In Wikipedia for example, why shouldn't you get credit if your link is good enough to stay in place in an article? Google should show me the popular pages. And pages with links in popular places are popular! That's how the pagerank works.