PageRank Sculpting Revisited

[Webnauts]

This thread is not for Neophyte SEOs. So think twice before posting, to avoid spoiling the thread with nonsense or irrelevant posts which will be removed without prior notice. Thanks.

This thread is a continuation of a discussion we had at the forums WebProWorld back in 2007 http://www.webproworld.com/webmaster...l=1#post353033 and I still did not get a clarification.

So lets continue the discussion here about Dangling Links (or Nodes) and PageRank Sculpting to come up to date.

So here are some points to get started with.

Eric Enge Interviews Google's Matt Cutts

Eric Enge:

"What we've been doing is working with clients and telling them to take pages like their about us page, and their contact us page, and link to them from the home page normally, without a NoFollow attribute, and then link to them using NoFollow from every other page. It's just a way of lowering the amount of link juice they get. These types of pages are usually the highest PageRank pages on the site, and they are not doing anything for you in terms of search traffic."

Matt Cutts:

"Absolutely. So, we really conceive of NoFollow as a pretty general mechanism. The name, NoFollow, is meant to mirror the fact that it's also a metatag. As a metatag NoFollow means don't crawl any links from this entire page. NoFollow as an individual link attribute means don't follow this particular link, and so it really just extends that granularity down to the link level.
We did an interview with Rand Fishkin over at SEOMoz where we talked about the fact that NoFollow was a perfectly acceptable tool to use in addition to robots.txt. NoIndex and NoFollow as a metatag can change how Googlebot crawls your site. It's important to realize that typically these things are more of a second order effect. What matters the most is to have a great site and to make sure that people know about it, but, once you have a certain amount of PageRank, these tools let you choose how to develop PageRank amongst your pages."

Source: http://www.stonetemple.com/articles/...tt-cutts.shtml

Notice: Lets ignore for now the bolded sentence and stick to the meta robots "noindex" and "nofollow" directives, because the link attribute "nofollow" Google is treating different: http://www.mattcutts.com/blog/pagerank-sculpting/

######

So now lets continue:

Dangling Links & Google's PageRank

A classic example of what a "dangling link" is.

Ammon explains that a page that is that is not in the Google index but the links are picked up to that page, won't count. Why? Well, the original PageRank document has a concept called "dangling links," which reads:

"Dangling links are simply links that point to any page with no outgoing links. They affect the model because it is not clear where their weight should be distributed, and there are a large number of them. Often these dangling links are simply pages that we have not downloaded yet.Because dangling links do not affect the ranking of any other page directly, we simply remove them from the system until all the PageRanks are calculated. After all the PageRanks are calculated they can be added back in without affecting things significantly."

Source: http://www.seroundtable.com/archives/014335.html

Is that really the case?

######

Looking a bit further, what is about the IBM patent posted by our member Bill Slawski:

Abstract:

A dangling web page processing system ranks dangling web pages on the web.

The system ranks dangling web pages of high quality that cannot be crawled by a crawler.
In addition, the system adjusts ranks to penalize dangling web pages that return errors when links on the dangling web pages are crawled.
By providing a rank for dangling web pages, the present system allows the concentration of crawling resources on those dangling web pages that have the highest rank in the uncrawled region.
The system operates locally to the dangling web pages, providing efficient determination of ranks for the dangling web pages. The system explicitly discriminates against web pages on the basis of whether they point to penalty pages, i.e., pages that return an error when a link is followed.
By incorporating more fine-grained information such as this into ranking, the system can improve the quality of individual search results and better manage resources for crawling.

Source: http://www.seobythesea.com/2007/08/s...angling-nodes/

######

Why all this? I think it is time to crack down and figure out if PageRank Sculpting works or not, and not just relying on stories of Google employees or well known SEOs, which many times they turn out to be misleading, even if they were not intentionally.

All above said: Where does the PageRank go in the above situations: To heaven or to hell?

Thanks in advance for your contribution.

[bill slawski]

Last year, when Google acquired a couple of thousand patents from IBM, one of the patents that was included in the first transaction was that very patent involving dangling nodes - http://www.seobythesea.com/2011/07/g...tents-in-july/

The patent is System And Method For Ranking Nodes In A Network

The Google paper that discusses dangling links is:

The PageRank Citation Ranking: Bringing Order to the Web

The use of a nofollow link (rel) value by itself isn't necessarily responsible for creating a dangling link, since it's quite possible that the page being linked to might be linked to by other pages, and can be crawled and indexed by search engines.

On the otherhand, a link to a page that has been disallowed for crawling through the use of robots.txt is a dangling link. A page that has a meta robots nofollow directive in its head section, is a dangling node, and a link to a page like that is a dangling link. A link to a PDF page or a Word document that contain no links or just an image are also dangling links.

The IBM patent does provide for a way for some dangling nodes to be given PageRank scores, and I'm sure most of us have seen PDFs, for instance, that have accumulated PageRank, so Google does go back an score some dangling nodes. From the IBM patent:

The rank redistribution processor alternates forward and backward steps in the random walk when encountering a node that is not a penalty node. The backward step confers score to the node. If the node is a dangling node but not a penalty node, the rank redistribution processor propagates the score associated with the dangling node among all the web pages contributing scores to the dangling node in the same proportion the score was contributed. If the node is a penalty node, the rank redistribution processor jumps to the virtual node, preventing scores associated with the step from being assigned to the node that linked to the penalty node.

I'm not sure that the dangling link/node analysis is helpful in determining what happens to the distribution of PageRank from a page being linked to via that link. We aren't telling search engines not to crawl or index those pages with a rel="nofollow" link value

We've had Matt Cutts tell us that the PageRank that would otherwise go to a page that includes a link rel="nofollow" evaporates. The other links on the same page aren't being given the PageRank that would otherwise go through that link.

Interestingly, if we look at Google's reasonable surfer patent, it tells us that some of the types of links that people were trying to sculpt using rel="nofollow" links are the kinds of pages that Google probably wouldn't pass along much PageRank through anyway:

Systems and methods consistent with the principles of the invention may provide a reasonable surfer model that indicates that when a surfer accesses a document with a set of links, the surfer will follow some of the links with higher probability than others. This reasonable surfer model reflects the fact that not all of the links associated with a document are equally likely to be followed. Examples of unlikely followed links may include "Terms of Service" links, banner advertisements, and links unrelated to the document.

With the reasonable surfer approach, Google may already doing some PageRank sculpting of its own. If Google's removing PageRank completely that would have gone to a page otherwise if not for a rel="nofollow", it might do a reasonable surfer type analysis first. And then just take that PageRank out of the equation.

Google also published a patent application that would enable webmasters to determine a percentage of "link weight" that would pass through links, in Embedded Communication of link information. Matt Cutts is listed as one of the co-inventors on that one. Google hasn't announced the ability for us to do that, and I'm not sure that they every will.

[Webnauts]

I posted on Stackflow an answer to a question of a member there:

Just to clarify some things once and for good.

Implementing this

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

not only tells the bots not to index the pages, but blocks them passing PageRank throughout the links found on the page. But that is not a problem itself. The problem is that such pages become dead end pages, otherwise called "Dangling pages" or "Nodes", which cause a dilute of PageRank, because the juice will be removed from the PR graph.
That said, the optimal meta would be implementing:

<meta name="robots" content="noindex,noarchive,nosnippet,follow" />

If that page is duplicated, it is a good idea to implement a canonical tag too, linking to the page version you want to have indexed. If not, at least to the most relevant page.
Using the canonical tag alone does not prevent indexing. At least not in all cases. For example, if a page have PR5 which has a canonical link pointing to an identical page which has PR 1, Google can ignore the canonical tag and index the page with PR 5.
So to avoid any misconceptions, the canonical tag is not a 301 redirect. That said, pages with a canonical tag can still accumulate PageRank. It depends who much juice reaches that page.

So again another reason that makes sense adding the "follow" meta robots directive as last.

According to Matt Cutts, Lead of Google's Web Spam Team, just using "noindex" or "noindex,noarchive,nosnippet" without adding the "follow" directive at the end, could be that Googlebot can mess up and not follow the links on that page.

Source: http://stackoverflow.com/questions/1...488305#9488305

But after talking with Andy Beard about that, a new question raised:

If the "Page B" has the meta robots directives "noindex,nofollow" and a canonical tag with destination to "page A", will more PageRank pass to "page A" though the canonical link, because from "page B" the links cannot pass PR juice any further?

[Adam Humphreys]

That's one for John Mu in WMT forum for sure.