72 replies to this topic

[Alan Perkins]

@ Alan, I have to disagree slightly. Nofollow has a meaning - don't follow this link.

No, the meaning it had - note, had - was don't trust this link, not don't follow this link. Then it became "This link is paid for". Now it's "Manipulate your PageRank here". That's what I mean by bastardised. Maybe, to Google, it's been like that all along and it has just been a slow process of discovery for the rest of us. But the way Google treats nofollow is not the way it was designed to be used, and is not the way that other engines do (or may).

[Jill]

But the way Google treats nofollow is not the way it was designed to be used, and is not the way that other engines do (or may).

Making it a "use at your own risk" proposition for me.

I personally don't believe in taking the band-aid approach to my site architecture and would not recommend using nofollow on links in my main navigation. Either the links are important (and you have them there) or there not (and you don't).

[Alan Perkins]

Making it a "use at your own risk" proposition for me.

Yep, me too on the whole. In some circumstances it could be useful - precisely to avoid having to use something like JavaScript to stop links being followed "too soon". However, if in so doing you taint your site, then why bother?

[qwerty]

No, the meaning it had - note, had - was don't trust this link, not don't follow this link. Then it became "This link is paid for". Now it's "Manipulate your PageRank here".

Then again, it's named "nofollow" rather than "notrust". "I don't trust this link" is the original use they gave us for it, but its name comes from the robots exclusion protocol, where nofollow means "don't follow this link". So it all makes sense to me linguistically, just not to the extent of the use of it.

[DanThies]

But the way Google treats nofollow is not the way it was designed to be used, and is not the way that other engines do (or may).

Let's see, which other search engines are we worried about exactly? MSN sounded pretty well on board with honoring nofollow as "don't follow this link," so that leaves Yahoo, right? 'Cause Ask isn't exactly crawling up a storm. The nice folks at Yahoo have this to say:

(silence)

When you ask for specifics... but it doesn't really matter if they follow the links or not, does it?

[nethy]

If this new meaning (manipulate link juice here) becomes universalised, it may have some usefulness.
For example URL tagging on your own site to differentiate between identical url links for analytics purposes.

[projectphp]

MSN sounded pretty well on board

That's my bigget beef: the lack of documentation. http://www.php.net/htmlspecialchars is an example of proper documentation. It has the usage, expected inputs, expected in/outputs. Where is a similar such page for nofollow?

I have no way of telling anyone what it does. I can't mjustify my belief, because there is no documentation at all. Heck, we are SEO professionals, and we are pouring over several blogs, conference announcements and individual chats trying to decipher what it all means. It is like a religious class!

Until there is a standardised, central location for documentation, even just on each SE, I can't recommend using nofollow.

For those who think I am dreaming, http://www.sitemaps.org/ is an excellent example of what can (quite trivially) be acheived.

[Alan Perkins]

but it doesn't really matter if they follow the links or not, does it?

No, that wasn't my concern. My concern was that a nofollow'd link on your site to a another page on your site would be a contradiction, given the reason nofollow was invented was to indicate a lack of trust in a link. AFAIK it still means exactly that to all other search engines except Google.

My further concern would be that this contradiction would paint your site as one that had been SEO'd which, in borderline cases (i.e. given other evidence), could be enough to trigger some kind of penalty. That's still a concern.

However, my concern has been massively lessened by the realisation that the Web must be full of nofollow'd links on a site to other pages on the same site; not because everyone is an SEO, but because most blog software would nofollow every link made in a comment ... and some of those links will be to other pages on the same site.

[daniel]

You can always use JavaScript to create a link that the search engines can't follow, and that will be the same as using a nofollow attribute. Of course, not as nice for accessibility and you'll be causing problems for users without JavaScript in their clients (such as screen-readers, etc.).

Isn't the nofollow a much nicer and more user-friendly way of achieving the same thing?

[DanThies]

Yes, Daniel, it is a much cleaner method. That's why I was generally opposed to "dynamic linking" before nofollow came along.

Alan, as you say, any site that allows non-owners to create links (blogs, forums) can also create internal links that are nofollowed. If a search engine wanted to figure out who's doing SEO, they don't need to look at anything more than the use of anchor text (internally and externally), which is a "cheap" calculation since they're already doing text analysis and would likely be able to use the same functions.

[projectphp]

However, my concern has been massively lessened by the realisation that the Web must be full of nofollow'd links on a site to other pages on the same site; not because everyone is an SEO, but because most blog software would nofollow every link made in a comment ... and some of those links will be to other pages on the same site.

Never considered that AT ALL!!! What a hoot! I bet the people making blogging software never thought of that.

Maybe I should petition wordpress to start "whitelisting" sites to avoid nofollow. But, wait, is that the right thing to do? I mean, didn't you pay for the link by "buying" the domain and "paying" for hosting?

I am so confused!!!!

[Randy]

For example URL tagging on your own site to differentiate between identical url links for analytics purposes.

Yahoo has already released a way to deal with these. Hopefully Google etal will follow their lead.

[nethy]

Thanks Randy. Dunno how I missed that thread, it has been an issue forme lately.
You mentioned it also on some other thread recently. But again, what about google? Is there any reason why the NoFollow should not be used to prevent indexation of pages with non-content effecting parameters?

[qwerty]

If there's a page you don't want indexed at all, it makes more sense to block it via robots.txt. Using nofollow on links to the page is more trouble and isn't as reliable: you could forget to nofollow one of your links, or someone on another site could link to the page without putting nofollow on the link.

[Dodito]

OK something else, noone seems to address: no-followed links are not even used for discovery. Can anyone tell me this referred to ON SITE no-followed links only ? I can say from experience that this is not true for external links. We have had pages, NOT linked internally, and the ONLY link being external, nofollowed ranking in the search engine. If the link were not even used for "discovery" how was it discovered, and how did it even end up ranking in Google (based on the content of the page). I do not think things are that simple. But I am not an SEO expert by any means, just trying to get our own website up and running and ranked.