Multiple Links from ONE Page to ONE other

Do multiple links from the same page affect PR?

[sudden]

Hi, I just found a directory I might want to get a link from. However, while guestimating the PR value of a link from that site, a question came to my mind.

The page where my link would be on, has multiple links to the SAME page on it - and many of them. As I said, it´s a directory, with the following structure for each listing:

HEADLINE (linked externally)
DescriptionDescriptionDescriptionDescriptionDescriptionDescriptionDescriptionDescriptionDescriptionDescriptionDescription

www.website.com (linked externally) / category (linked internally)

So, for every listed website, there are two external links to the web site in question AND an additional internal link (to the category´s main page)

I hope you get the situation from my description, there are at about 30 links on the page which all link to the same internal page. Now: Will these links, looking at PR, count as one or as 30? The difference of the PR value for the other links on the page would be quite high.

Any thoughts?

[lolo7]

I think yes - they count as 30. But PR from these 30 links is the same if it would be 1 link becasue the summary PR divided by 30 and all 30 parts comes back to the page.

[John_Caius]

Precisely - thirty one-thirtieths equals the same as a single link. One link in a list of 200 counts for half as much as one link in a list of 100.

[cwebb]

Well, 30/31th is better than half the PR passed when there is one outbound and one internal link, BUT if you have 30 links from one page, Google will most likely decraese their value, so it counts significantly less than 30 links from 30 different pages!

[gmoney]

there are at about 30 links on the page which all link to the same internal page. Now: Will these links, looking at PR, count as one or as 30

I believe that Google would want to treat identical links off of one page as simply one link. Basically, I feel that Google would do something similar to the following: 1) make a list of links off the page, 2) reduce the list down to a unique list of links and 3) then do all their PageRank calculating business.

It seems like a natural (and simple) way to easily avoid PageRank manipulation so that the PageRank theory can work as it is intended. However, I don't have any real evidence of this.

Credits: Comments by PhilC (in a stickymail that is gone now) helped me to formulate my opinions on this one. If I recall, PhilC quoted some lines from the original PageRank papers to make his case.

[ciml]

> would want to treat identical links off of one page as simply one link

I am very sure that's the case. I have a page with three links, two go to the same URL. Each URL gets exactly the same PR from the link (⁺_- 1/30 of a Toolbar PR notch).

[John_Caius]

Just to clarify my slightly confusing wording above:

thirty one-thirtieths = thirty (one-thirtieths) = 30/30

:)

[sudden]

Thinking about it, counting all those links as a single one seems to be the only thing that makes sense. Thanks for sharing your thoughts!