homepage Welcome to WebmasterWorld Guest from 54.197.94.241
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

    
Hmmm, Wikipedia links may not be passing PR
despite the fact that they voted against using nofollow
oddsod




msg:3186446
 5:26 pm on Dec 12, 2006 (gmt 0)

I've just spent a fair amount of time researching old versions of wiki articles (pre cut-off for last PR update). External links and reference section links that have been there for almost a year don't seem to be gaining any (toolbar) PR from en.wikipedia.

I appreciate that PR juice is shared among all the outgoing links in any article (and the internal ones). But even small PR7 articles with just one or two OGLs, and hardly any wikified words, don't seem to be passing PR. The link destinations are PR3s and even PR2s in some cases. For all my experience with PR this just doesn't add up.

OK, toolbar PR doesn't matter etc., etc. I know the argument and I agree that it plays very little part in SERPs so let's avoid making this a discussion about whether PR helps in SERPs ;). This post is about Wiki links: Am I alone in thinking that Google is not passing PR for them, strange though it may seem? Has anyone else noticed this behaviour?

 

jomaxx




msg:3186463
 5:31 pm on Dec 12, 2006 (gmt 0)

I think I've read that some Wikipedia links use nofollow and some don't. Are you looking at the source code to see which is the case?

oddsod




msg:3186488
 5:48 pm on Dec 12, 2006 (gmt 0)

Yes, checked that first. I know that for a while Wiki used/considered using nofollow. So I checked the source for the articles as they were in the past (just in case for some strange reason there was a condom in place at that time). Searching for site:en.wikipedia.org mypagename.html does bring up the relevant wiki page/s. So Google is seeing the links but, according to my suspicions, not passing the juice.

Robert Charlton




msg:3186549
 6:15 pm on Dec 12, 2006 (gmt 0)

Searching for site:en.wikipedia.org mypagename.html does bring up the relevant wiki page/s. So Google is seeing the links but, according to my suspicions, not passing the juice.

You can't search source code in Google. You can only search what Google sees on the page. The search you describe is a search for visible anchor text only, not for how Google might regard the coding of the link.

oddsod




msg:3186625
 6:51 pm on Dec 12, 2006 (gmt 0)

OK, I missed a paragraph break, sorry. :)

<br>
Also, searching for site:en.wikipedia.org mypagename.html does bring up the relevant wiki page/s. So Google is seeing the links but, according to my suspicions, not passing the juice.

I'm not using the existence of the anchor text as any advanced proof. It's simply that wiki is probably not sticking my link in a js wrapper or suchlike when Googlebot visits them. The issue here is the PR.

PS: Sometimes my anchor is the URL

bwnbwn




msg:3186670
 7:26 pm on Dec 12, 2006 (gmt 0)

wiki was a nice place to visit and contribute info so I spent allot of time creating aticles adding content and linking to source pages some mine some not, problem is u get a 10 year old can come in do a domain search and delete all the links as spam replaced them he comes back wacko so just quit trying to contribute to this as it is way way out of hand now.

The Info is not correct unproven etc.

oddsod




msg:3186678
 7:34 pm on Dec 12, 2006 (gmt 0)

The Info is not correct unproven etc.

Are you referring to my suspicion that wiki links aren't passing PR? Have you done any testing on this?

AndyA




msg:3186696
 7:55 pm on Dec 12, 2006 (gmt 0)

I don't believe that Wiki links pass much PR, if any at all. I haven't really studied this theory or tested it, but there are several rather long articles, which I had nothing to do with in any way, that link to several pages on my site, and those pages don't appear to be getting any bounce because of the links.

In fact, this same site has quite a few links going to it from different - but related - topics (it is considered an authority site in its field, but not by Google).

oddsod




msg:3186704
 8:02 pm on Dec 12, 2006 (gmt 0)

I don't believe that Wiki links pass much PR

Ah, so it's not just me? :)

But if it looks like a (non-condomed) link and smells like a link....

Would Google manually block the passing of PR from Wikipedia? I mean, these are links that anybody can create. Just like on link farms. Or is this tin-foil territory?

leadegroot




msg:3186728
 8:38 pm on Dec 12, 2006 (gmt 0)

It doesn't surprise me at all - Matt Cutts has hinted that this would happen.

Interestingly (and its only a sample of one), I had a link removed from a wikipedia page - and the traffic to that site from Google increased afterwards. Almost as if having a link on Wikipedia is actually a negative filter point...
Probably a coincidence, but I hadn't been working on that site anywhere around that time...
I don't think an algorithm of 'only in external links, re-added a few times by same user, etc' would be too difficult given the public history of each page on the site

rytis




msg:3186745
 8:48 pm on Dec 12, 2006 (gmt 0)

I am not saying yes or no, but theoretically SEs may (and should) decide to pass or suppress the link benefits (PageRank - authority - anchor text value - etc..), to pass a percentage of any component, to apply this policy to any site manually or algorithmycally (e.g. too many links to "bad" sites in Google's eyes), to make this passing/non-passing obvious or masqueraded (toolbar PR) and so on.

On en.wikipedia.org quality - last time ended up there few days ago, the links for article were like current live.com SERPs - yes, on subject, but not even 2nd tier best choices, 3rd tier.

Wizard




msg:3186765
 9:18 pm on Dec 12, 2006 (gmt 0)

Since I have gained a link from PR7 en.wikipedia page, my site climbed from #50 to #8 in its main keyphrase (included in wikipedia outbound link anchor text). After last tPR update my site is still PR4 but it was notced the last tPR export was a snapshot from June and I gained the wiki link in July.

So I think wiki links still have a big value, but there might be exceptions.

jetteroheller




msg:3186780
 9:40 pm on Dec 12, 2006 (gmt 0)

I have a link from a PR7 article in Wikipedia.
But on this page are 452 links in total.

PR3 from the linked folder on my site
PR2 from other pages of the linked folder on my site

I compared with an other folder on this site, about same creation date, but no PR7 links from Wikipedia. Exactly the same, folder index PR3, rest of folder PR2

Scarecrow




msg:3186983
 2:26 am on Dec 13, 2006 (gmt 0)

Wikipedia started a biography on me without my permission 14 months ago. Since then I've been fighting them to take it down because I don't want a troll magnet in Wikipedia that has to be watched for the rest of my life.

I started an anti-Wikipedia site because they wouldn't take the bio down. Then the bio got longer and longer, because the anonymous teenage administrators on Wikipedia decided that I should be taught a lesson.

They have five of my sites listed as "external links" at the bottom of my bio. For months they were live links. Then they decided that all five should be put on the Wikipedia "spam blacklist." That was because I started redirecting them to another anti-Wikipedia site (not mine) when I detected the Wikipedia referrer. I stopped doing that last June, but they're still mad over this, so the five sites are de-linked, and anyone who comes along and tries to edit them into hot links gets intercepted by the spam blacklist and the edit is aborted.

The bottom line to this silliness is that for certain periods during this long war, the URLs were hot, while during other periods they were de-linked and you had to paste it into your browser address bar. But either way, it makes absolutely no difference to my traffic. In fact, when the links are hot, I get only about a dozen extra visitors to each of my sites from Wikipedia. No big deal.

The problem I have with Google and Wikipedia is that Google ranks the Wikipedia pages insanely high in the search results. Then the scraper sites grab Wikipedia's content, thinking that it's good content for AdSense. And guess what -- even the scraper sites rank well! There are a couple scraper sites that have grabbed my bio, and split it up into 10 different pages to avoid duplicate penalties, and amazingly enough, each page of the scraper site does well for the terms I'm searching.

Google has problems with Wikipedia. If they deprive Wikipedia of its ability to pass PR, that would only be the start of a long list of things they should to do fix their Wikipedia problem.

For example, the article Talk pages, the User pages, and the User_talk pages are indexed by bots. Wikipedia won't disallow them in robots.txt because Wikipedia's own internal search function is so crummy, that Wikipedia editors wouldn't be able to find anything. But these non-article pages are filled with libel and flame wars. If Wikipedia isn't smart enough to keep those out of the bots, then the bots should do it for them.

In a search for my name, the bio comes up as number one, and under that is my User_talk page. All it says when you click on it is, "This user is banned from editing Wikipedia."

What a joke. Google, take a look at how you index Wikipedia. Half of it should be treated as spam.

oddsod




msg:3187275
 11:06 am on Dec 13, 2006 (gmt 0)

I agree with you that there may be rubbish on Wikipedia and that lots more could be done to sort it and that Google could rank my pages above Wikipedia in SERPs ;) but .... why not pass PR? That's the bit that doesn't make sense.

I compared with an other folder on this site, about same creation date, but no PR7 links from Wikipedia. Exactly the same, folder index PR3, rest of folder PR2

Thanks for sharing that. It's not conclusive proof, of course, but it's what I'm seeing time and time again, on several different sites.

theoretically SEs may (and should) decide to pass or suppress the link benefits (PageRank - authority - anchor text value - etc..), to pass a percentage of any component

Don't get me wrong, I agree that SEs can do whatever they want with what is essentially their algo. I'm trying to establish whether others have seen this same lack of tPR and if it is the case that PR is blocked I'm hoping to discuss the logic. If Google thinks a particular (Wiki) page should be #1 in the SERPs why would it not (give PR) value to long term links on that page? Particularly those quoted as "References"?

It doesn't surprise me at all - Matt Cutts has hinted that this would happen.

I must have missed that somewhere. You wouldn't have a link, would you, leadergroot? I'd be interested in reading what exactly was said on the Wikipedia link issue.

photopassjapan




msg:3187278
 11:10 am on Dec 13, 2006 (gmt 0)

Have you seen searchmash lately?

:D

Go and check.
I haven't seen it for months, and since then two interesting things showed up on the sidebar to the right... things i clearly remember posting here, and posted by others. I wonder if this will even be taken to G in full time. Wouldn't hurt them at all. Neither us. Nor users.

If they are implemented it's gonna be an interesting day for webmasters throughout the world. I mean a LOT of people would say, hey G actually KNEW all the way! But also it would chill a lot of nerves in my opinion. This IS the way it should have been right from the start of picking up wiki and blogs.

Am i right or am i right :P

leadegroot




msg:3187298
 11:39 am on Dec 13, 2006 (gmt 0)

He didn't mention wikipedia explicitly (IIRC) (and I don't have the energy to go hunting the ref atm) it was more the mention that if they don't believe the links are quality and you don't put a link condom on it - they will do it for you.

oddsod




msg:3187452
 2:15 pm on Dec 13, 2006 (gmt 0)

if they don't believe the links are quality and you don't put a link condom on it - they will do it for you

Ah, OK. I thought it was specific to Wikipedia. Maybe Google is blocking all links because of the low quality of some Externals... and Reference links not getting PR is the incidental damage.

selomelo




msg:3187964
 9:05 pm on Dec 13, 2006 (gmt 0)

In fact, I noticed a couple of months ago that the Wiki in my native language (Turkish) explicitly uses "nofollow." When I first noticed it, I checked also the English version to see if it uses the "nofollow" attribute too. No, English Wiki does not use it. I do not know if this is isolated to the Wiki in Turkish. It is worth to check for it in other languages.For an example, you can have a look at source of [tr.wikipedia.org...]

Eltiti




msg:3190468
 1:34 pm on Dec 16, 2006 (gmt 0)

Selomelo: I checked the Dutch, French, German and Spanish versions of the article you refer to, and they all use "nofollow".

[edited by: Eltiti at 1:34 pm (utc) on Dec. 16, 2006]

mattg3




msg:3190516
 3:06 pm on Dec 16, 2006 (gmt 0)

I have many longlasting links from English and German WP. It at least brings interested very targeted users ..

I don't think the english WP links count as I can't find any listed as backward links. Might be the intentionally broken link command though.

Just found 2 backlinks from the vietnamese Wikipedia and also from the danish one. I speak neither of these languages..

jetteroheller




msg:3190526
 3:38 pm on Dec 16, 2006 (gmt 0)

Maybe it's we have here a discussion about

only the first 100 links on a page are counted.

This external link section on a Wikipedia page is always far beyond the first 100 links.

On the other side, I have created a new english Wikipedia article, because there was a PR7 article, where my site had been not related enough.

In the new english Wikipedia article, I placed 3 links to my site and linked from the PR7 article to my new one. This Wikipedia article reached PR5 in only 2 month.

oddsod




msg:3190650
 7:08 pm on Dec 16, 2006 (gmt 0)

>> This Wikipedia article reached PR5 in only 2 month.

I take it your opinion is that your page got PR from the Wikipedia article. I'd be interested in hearing from anybody else who feels they got PR from the English wikipedia. I'd dearly love for my theory to be debunked.

The point about the first 100 links wasn't one I considered but I went back and checked and at least two of my long standing "Reference" links are within the first 100 hrefs.

mattg3, I believe that the backlinks you're not seeing is solely because of the "broken" Google. Yahoo's backlinks work and, as they too subscribe to "no follow", Yahoo's backlink check should tell you if someone is using the link condom.

mattg3




msg:3190683
 7:49 pm on Dec 16, 2006 (gmt 0)

Yahoo: on domain alone

Pages (1,108,755) Inlinks (63,395)

on wikipedia
We did not find results for: http://www.example.com domain:wikipedia.org. Try the suggestions below or type a new query above.

Hmm

jetteroheller




msg:3190731
 9:50 pm on Dec 16, 2006 (gmt 0)

>> This Wikipedia article reached PR5 in only 2 month.
I take it your opinion is that your page got PR from the Wikipedia article

Not my page, my Wikipedia article got PR5 in 2 month.
My own site was not influenced by the link

wikipedia/important city PR 7 link to wikipedia/event in important city

wikipedia/event in important city PR5 link to example.com/my-big-reportage-about-event-in-important-city

example.com/my-big-reportage-about-event-in-important-city PR4

steveb




msg:3190738
 10:14 pm on Dec 16, 2006 (gmt 0)

It's easy to find wikipedia links listed as backlink in Google and Yahoo, but that doesn't address an passage of toolbar PR. (And even if they don't apsss toolbar PR that doesn't mean they don't pass real PR there at the plex.)

tedster




msg:3190797
 12:12 am on Dec 17, 2006 (gmt 0)

Relevance signals do seem to pass - at least I've seen a couple cases that look like a wikipedia link quickly helped with search terms that were only ambiguously present on-page and in previous backlinks. But passing PR does seem questionable to me, too. As noted above, it is hard to measure with the plethora of outbound links on many wikipedia articles.

At the same time, I'd expect a link's sticking power to be a factor on any wiki. The whole idea of a wiki involves the "community" in policing the dross. So link age would probably play into the picture, and possiby more intensely than on most sites. So I'm theorizing here that PR might pass more effectively over time, as a link gains history without being edited/deleted/replaced etc.

nippi




msg:3190903
 6:50 am on Dec 17, 2006 (gmt 0)

makes sense to pass no value from wikipedia.

Only way to stop abuse.

oddsod




msg:3191004
 11:44 am on Dec 17, 2006 (gmt 0)

nippi, that may apply to "External Links". A blanket ban would be Google's loss as Wiki article "Reference" sections benefit from carefully hand selected academic works of relevance to the page (kinda like the attraction DMOZ originally had).

I too believe that there is some SERP benefit from being linked from a Wiki article. The extent of the benefit probably varies from article to article. And, as you say Tedster, it makes sense to value long-standing links more.

... it is hard to measure with the plethora of outbound links

I agree. Though one or two of "my" articles are quite small PR7s with a less than a handful of outgoing links and not many wikified terms. They are what originally led me to believe that tPR wasn't passing.

selomelo




msg:3191161
 4:55 pm on Dec 17, 2006 (gmt 0)

I can confirm that Wikipedia may not be passing PR, at least in the usual way.

I have a link from a Wikipedia PR6 page to a relatively new page on my site. The Wikipedia page has only a few outbound links (less than 10) The page on my site has some really good inbound links (with varying PRs, some PR4, and some PR5). Currently its PR is 4 suggesting that Wikipedia may not be passing any PR. Nevertheless, both the link as well as the page are relatively new, and age may be a factor here.

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