homepage Welcome to WebmasterWorld Guest from 50.17.86.12
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

    
Matt Cutts: Further Clarification on the nofollow Attribute
tedster

WebmasterWorld Senior Member tedster us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 3436630 posted 2:47 pm on Aug 30, 2007 (gmt 0)

Thanks to Rand over at SEOmoz. He managed to tweeze out some precise detail from Matt Cutts about the rel="nofollow" atttribute.

Matt Cutts: "The nofollow attribute is just a mechanism that gives webmasters the ability to modify PageRank flow at link-level granularity. Plenty of other mechanisms would also work (e.g. a link through a page that is robot.txt'ed out), but nofollow on individual links is simpler for some folks to use. There's no stigma to using nofollow, even on your own internal links; for Google, nofollow'ed links are dropped out of our link graph; we don't even use such links for discovery. By the way, the nofollow meta tag does that same thing, but at a page level."

[seomoz.org...]

I've been quite resistive about the use of nofollow on internal links. Just seemed wrong to me. But Matt says it's OK, so who am I to contradict?

And how about not using nofollow links even for discovery! Totally dropped from the link graph - that one surprised me.

 

Reno

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 3:05 pm on Aug 30, 2007 (gmt 0)

Thanks tedster for posting this much needed clarification. Google's usage of nofollow is finally starting to make some sense -- we've come a long way in our understanding in the past month.

.................................

achtungbaby

5+ Year Member



 
Msg#: 3436630 posted 3:09 pm on Aug 30, 2007 (gmt 0)

I was reading Rand's post earlier, and decided to have a scout around to around to see if I could find any expert commentary on the subject.

This has been at the forefront of my mind for a site I've been working on for about a year that will go live in the next couple of months. Most of the commentary pre-Matt's post generally seemed to suggest it's a no-no.

However, it would now appear Matt is basically 'green-lighting' this method...

There's no stigma to using nofollow, even on your own internal links;

which is great news if it indeed does help to control which direction linkjuice is flowing in.

jimbeetle

WebmasterWorld Senior Member jimbeetle us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 3436630 posted 3:59 pm on Aug 30, 2007 (gmt 0)

And how about not using nofollow links even for discovery! Totally dropped from the link graph - that one surprised me.

Yeah, that's a huge surprise to me also. Knowing how in the past G's appetite for links was voracious, this basically puts whatever number of pages it turns out to in a virtual black hole.

Anybody who uses nofollow "to modify PageRank flow at link-level granularity" (luv that engineer talk), must be absolutely sure it doesn't inadvertently orphan any important pages.

Quadrille

WebmasterWorld Senior Member quadrille us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 3436630 posted 4:05 pm on Aug 30, 2007 (gmt 0)

I'm surprised too - I'm not about to reread every page of MC's blog, but I could swear that he once said they [b[]were[/b] followed.

Still, good to get a clear current position :)

Tonearm

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 4:31 pm on Aug 30, 2007 (gmt 0)

I wonder if the <meta name="robots" content="none"> tag is one of the alternatives Matt alludes to.

edit:

I wouldn't want PR to flow to the content="none" page and *then* have it excluded.

[edited by: Tonearm at 4:32 pm (utc) on Aug. 30, 2007]

achtungbaby

5+ Year Member



 
Msg#: 3436630 posted 7:39 pm on Aug 30, 2007 (gmt 0)

Matt Cutt's has done a follow up post to Rand's posting which says...

I think saying people "should be" using nofollow is a bit strong. More like people can use it for internal links if they're power-user-y enough to want to sculpt PageRank flow within their site at the link level. But I'd say that most regular webmasters don't need to worry about link-level PageRank flow within their site. I think saying "power users and webmasters should be employing on their sites" overstates it a little. It's available if you want to get into that much fine-grained control.

Hmmm, so it's not so much as a green light as I stated earlier, more like an amber coloured light!

Oh well, I think I'll stay on the side of caution methinks - it's Find & Replace time on all the no "nofollow's" added in earlier today!

anallawalla

WebmasterWorld Administrator anallawalla us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 3436630 posted 3:50 am on Sep 4, 2007 (gmt 0)

And how about not using nofollow links even for discovery!

One of my sites showed a backlink from my comment in someone's WordPress blog (nofollowed) - but not right now. This will put a dent in those comment spam tools.

soapystar

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 9:10 pm on Sep 4, 2007 (gmt 0)

i wonder if this reflects a change in application rather than a clarification?

i dont really understand how they could green light its use at will given the implications. If it caught on that the way to stop pr leakage was to nofollow on your links and it didnt have a negative impact on your own site, then everyone (not just harden webmasters) will end up using it (overtime) on all external links. And then where will that leave an algo based on link votes?

When this tag was first introduced i only recall the comments being that it was meant for untrusted links where anyone can post a url. It was clearly a signal of untrusted or unvetted, how can it then be applied internally with the same meaning?

CWebguy

5+ Year Member



 
Msg#: 3436630 posted 9:18 pm on Sep 4, 2007 (gmt 0)

I've been quite resistive about the use of nofollow on internal links. Just seemed wrong to me. But Matt says it's OK, so who am I to contradict?

I believe Matt was specifically refering (at least in another interview if not this one) to using these internally to block "pointless" links on your site, links such as "post a comment", "view all comments", etc and never for page rank manipulation which is a Google no no. Blogs are notorious for these kind of links all over the place and can easily be taken care of using the "no follow" tag as they need not be indexed. (Matt also tends to address the blogging community a lot, maybe because he has a blog :-)

Just my take on it!

God Bless!

[edited by: CWebguy at 9:21 pm (utc) on Sep. 4, 2007]

orionsweb

5+ Year Member



 
Msg#: 3436630 posted 9:51 pm on Sep 4, 2007 (gmt 0)

Matt mentions that this can also be managed via a robot.txt file and that is how it should be done.

The nofollow route on individual links ONLY works in Google and none of the other engines / directories, so google will or could end up picking up the links through other listings on other sites. Using the robot.txt route at least MOST of the SE's will follow your requests that way.

europeforvisitors



 
Msg#: 3436630 posted 10:17 pm on Sep 4, 2007 (gmt 0)

i dont really understand how they could green light its use at will given the implications. If it caught on that the way to stop pr leakage was to nofollow on your links and it didnt have a negative impact on your own site, then everyone (not just harden webmasters) will end up using it (overtime) on all external links. And then where will that leave an algo based on link votes?

If the misuse of "nofollow" were to become a problem, then Google could simply adapt and refine its policy to deal with it. For example, a non-blog site that had hundreds of external links, all with "nofollow," might be treated with more skepticism than a site that mostly used normal links. (And why not, with the site owner being stupid enough to offer Google such a blatant indicator of unnatural SEO?)

CWebguy

5+ Year Member



 
Msg#: 3436630 posted 10:20 pm on Sep 4, 2007 (gmt 0)

The nofollow route on individual links ONLY works in Google and none of the other engines / directories, so google will or could end up picking up the links through other listings on other sites. Using the robot.txt route at least MOST of the SE's will follow your requests that way.

Yahoo and MSN say that they have implemented the nofollow tags, but wheter they truly have or not I can't say for sure.

God Bless!

[edited by: CWebguy at 10:21 pm (utc) on Sep. 4, 2007]

Lord Majestic

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 11:07 pm on Sep 4, 2007 (gmt 0)

Yahoo and MSN say that they have implemented the nofollow tags

There is no exact spec on what exactly a search engine should do when finding a link with a nofollow on it - the variety of things that can be done is high and all MSN/Yahoo/others who "support" it actually mean they will do something unspecified but most likely discounting value of PR like algorithms.

whitenight

WebmasterWorld Senior Member 5+ Year Member



 
Msg#: 3436630 posted 11:17 pm on Sep 4, 2007 (gmt 0)

For example, a non-blog site that had hundreds of external links, all with "nofollow...

And why not, with the site owner being stupid enough to offer Google such a blatant indicator of unnatural SEO

Gee! Do you mean like those morons over at Wiki who have ONLY nofollowed external links?

Yea, I can see how Goog has slapped down on them (and countless other sites who do this) by increasing their rankings since it's implementation...

europeforvisitors



 
Msg#: 3436630 posted 12:43 am on Sep 5, 2007 (gmt 0)

Wikipedia is using nofollow in the same way (and for the same reason) that a blog would: to prevent link spam by third-party contributors. But I'll bet you already knew that. :-)

whitenight

WebmasterWorld Senior Member 5+ Year Member



 
Msg#: 3436630 posted 1:06 am on Sep 5, 2007 (gmt 0)

Just to clarify:

Are you saying i should not listen to MC?

Cause from previous conversations
(especially the one about 8 months ago about this subject in which i said exactly then what MC is saying now),
I assumed I should always listen to MC, yes?

Just clarifying...

Reno

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 1:12 am on Sep 5, 2007 (gmt 0)

If the misuse of "nofollow" were to become a problem, then Google could simply adapt and refine its policy to deal with it ... And why not

For this reason I will err on the side of extreme caution and only use nofollow for specific kinds of external links, such as those that are paid. Otherwise, I'm sticking to the tried and true, because for one thing, I don't want to have to go back and modify them all again when the refinement takes place (which it inevitably will when the abuse becomes blatant).

.................

whitenight

WebmasterWorld Senior Member 5+ Year Member



 
Msg#: 3436630 posted 1:18 am on Sep 5, 2007 (gmt 0)

Sigh.

Amazing how people listen to MC when they shouldn't and then ignore him when he's being straight-forward.

Here's the aforementioned thread.

Some nonsense about testing and data as opposed to fear-based conjecture.

[webmasterworld.com...]

Miamacs

5+ Year Member



 
Msg#: 3436630 posted 12:59 pm on Sep 5, 2007 (gmt 0)

Yahoo and MSN say that they have implemented the nofollow tags, but wheter they truly have or not I can't say for sure.

...

I did a test when I didn't believe Google would behave as it said it would. I checked it again today. In short, this is a trusted, authority site, where I've put a link onto the homepage itself, pointing to a page created for this test, not linked from elsewhere, using a unique phrase of otherwise common words. Very basic. Since the site is region specific, and the link was well below the fold ( yet came like... the 3rd in the code ) there was very little chance of someone corrupting the experiment. Yeah not much of a test, but it's better than nothing.

And so the results:

MSN / LIVE

- Will not index the page if it can only be found through a nofollowed link.
- Not sure about crawling, but it won't index it.
- Brought up the page with the link on it for an exact phrase query ( between quotes ), the phrase is unique, no other results.
- The source page with the link on it **ranks** for the phrase without quotes ( position 80 out of 40.000 )
- The anchor text is treated as text and is indexed.

Google

- Will not index the page if it can only be found through a nofollowed link.
- Not sure about crawling, but it won't index it.
- Brought up the page with the link on it for an exact phrase query ( between quotes ), the phrase is unique, no other results.
- The source page with the link on it does *not* rank at all for the phrase without quotes
- Neither the target page, nor the link to it is shown in any reports.
- The anchor text is treated as text and is indexed.

Yahoo!

- **Will** index the page even if it can only be found through a nofollowed link.
- Not sure about crawling, should have paid more attention, but it's 99.9% that there were no ways around that link. ( see below )
- Brought up the page with the link **to** it for an exact phrase query ( between quotes ), the phrase is unique, no other results.
- Meaning: the source page with the link on it does **not** come up, not even with the phrase between quotes.
- The source page with the link on it does *not* rank at all for the phrase without quotes either
- But neither does the target page with the link **to** it.
- The link and the target page is featured in reports. No other links point to the target page.
- The anchor text is **not** treated as text and is **not** indexed.

...

I'm sure no one expected results that made sense.

The right blend would be something combining what Google and Yahoo! does.

Huh? That Google got it right already? Well sure.

In case of Google: if it works the same for interdomain linkage, the link text with pills, p0rn, stuff you don't want to rank for is indexed, and casts a faint little relevancy vote for the page which it's on. Or at least, helps it towards the net of semantic filters.

On a sidenote I have some blurry memories on Google initially crawling and indexing the page as a supplemental result, then dropping it entirely. But that was more than half a year ago, and the way Google behaves with nofollow may have changed. Also, I see inbound links in GWT using nofollow all the time. So the way Google treats nofollows within the same domain / to pages that are indexed already, might be different.

...

giuliorapetti

5+ Year Member



 
Msg#: 3436630 posted 1:22 pm on Sep 5, 2007 (gmt 0)

I quote from

[webmasterworld.com...]

" As explained by Tim Converse (Yahoo!) and Brian White (Google) in one of the 2006 PubCon Las Vegas conferences, the name of that directive is unfortunate and deceptive.

nofollow does not actually mean "do not continue your crawling any further from here" but, rather "do not count the links included in this page as *votes* for the linked pages for the effect of link popularity".

Tim Converse "translated" nofollow in "untrusted/non-endorsed links on this page".

Therefore, the crawlers will surely follow the links included in this page. "

I think a lot of this forum's members were in that session at PubCon, and what everyone said was very clear.

So they have changed their mind for sure.

Also, is not only words that have been said, who are contraddicting this new "wave of thought": there is also Wikipedia:

"What nofollow is not for - The nofollow attribute value is not meant for blocking access to content or preventing content to be indexed by search engines."

:)

I guess this news from Tedster deserve a rewording: not "clarification" about nofollow but "change of meaning"...

soapystar

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 1:28 pm on Sep 5, 2007 (gmt 0)

giuliorapetti

your singing my song......

it doesnt matter how Google ACTUALLY treat rel=nofollow..its what the wider world BELIEVES to be the case....if it became accepted thought that using that tag will not hurt you and will stop pr drain then it will be catastrophic in the long run....because just like recip links eventually became common knowledge for every schoolboy and amateur webmaster...if this tag became used for genuine one way outbounds then it will change the whole concept of the original algo...the question would be left..why bother to not use the nofollow on every single outbound...you are still serving your users by proving the links to those sites...your just not giving away any of your own sites value...

now if Matt is truly saying the tag has the same effect as the meta nofollow noindex or using robots...does he just mean from the point of view of not indexing that page?

because i was always of the belief that even though a link was not followed say using the robots....the pr from that link was still given to the linked page..infact you can see that using the toolbar...non-indexed but linked pages get pr...use the rel nofollow stops that flow..but does that flow get redirected to the rest of the page its on or is it just a stop point for its outgoing pr value...

[edited by: soapystar at 1:36 pm (utc) on Sep. 5, 2007]

Lord Majestic

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3436630 posted 1:29 pm on Sep 5, 2007 (gmt 0)

" As explained by Tim Converse (Yahoo!) and Brian White (Google) in one of the 2006 PubCon Las Vegas conferences, the name of that directive is unfortunate and deceptive.

I've been saying this since the day they announced it :(

Miamacs

5+ Year Member



 
Msg#: 3436630 posted 1:33 pm on Sep 5, 2007 (gmt 0)

I know, I've been saying the same thing.

But since it was disputed, I did some tests.

At first, it was somewhere along the lines of the quoted comments.

Not so the past few months.

See test above.

...

StuffOfInterest

5+ Year Member



 
Msg#: 3436630 posted 3:02 pm on Sep 6, 2007 (gmt 0)

If this is the way Google is now handling internal nofollow attributes it could be really good for me. Over a year ago I took a gamble and implemented nofollow on my internal links. My site has a few hundred content pages, pages with original unique text or images, and many thousands of "reference" or "placeholder" pages, pages which point to a place but don't have any details. Being that both types of pages are in the same directory tree I can't use a simple robots.txt directive to tell the crawlers some pages are not worth looking at. What I ended up donig was putting robots meta-tags on the non-content pages and nofollow attributes on any pages pointing to the non-content pages.

Has this helped? I can't really say. I still see a lot of placeholder pages in the search results. As mentioned earlier, Yahoo slurps up anything it can find. Google does seem to be getting more selective on what it goes after on my site. Hopefully, in time, I'll start to see cleaner results. At least it is good to know that I shouldn't be incurring a penalty for setting things up this way.

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved