Google "distinguishes natural links from unnatural links"?

http://sitemaps.blogspot.com/2006/05/live-in-our-hometown.html

 

It is impossible for google to tell what is natural. New york times did an editorial one time about something or other (I forget). Inside the editorial, they used a phrase (and linked it to a page on my site that used the same phrase) that had nothing to do with the article itself, but it showed up in google backlinks forever. I am assuming that google thinks the link is "natural"

I'd bet google's primary methods of finding "un-natural links" primarily include:
sitewide links
footer links
links to many different sites in rapid succession

Beyond that, how can it tell what is natural and what isn't? Can they read the webmaster's mind as he codes up a page?

 

I think they might have some type of algorithm, but that doesn't mean that it's going to work as well as they imagine.

At the very worst I'd expect they could simply ignore incoming links instead of giving you credit for them, otherwise it is too easy to throw up thousands of pages with "bad" links to someone you don't like... or to spam someone into thousands of free for all links pages.

 

I find the claim rather silly and rather insulting.

 

Minstrel, the way Google knows which links are natural and which ones might not be is through their own trusted sites program. Now how they come up with which sites they trust is something I still don’t totally understand and probably never will. My guess is that they use some type of trust rank system backed on old-line high PR trusted sites. A link from them will carry great weight with Google while other links will have little to no effect. Any site with a PR of more than 7 for several years would most likely fit this bill of being trusted. Any page on those sites, which link to the outside world, will have at least some value and add to the value of the linked to sites long term prospects of moving up in Google trust rank as well.

Somewhere in this equation is an archive factor, which has become more important over the past few years. Static pages, which have not changed in years, will rank very well even without a bunch of links pointing at them. My guess is that Google is looking for old pages that never change and trust those pages more because they are less likely to be modified to reflect SEO from updates to the Google system. There have been some examples of pages with no incoming links coming up number one with very little SEO to them. Most of these pages had one thing in common and that is they have not been modified for over one year.

Even on dynamic pages there are ways that Google can see if the page has changed within a period of time. More than likely Google keeps a record of every time a page has changed since they first found it. The more that page changes, the worse the page will rank. Now this is speculation on my part, but it is clear that Google loves old, unchanged content and somehow rewards those pages. I’m not that familiar with dynamic generated pages, but I’m sure there are certain parameters which contain content and changes to this content could be track even without a last mod date.

If I’m right, folks with dynamic pages have an advantage over static pages because they are able to changes page layout site wide without changing the actual content on the page. I have noticed on several different occasions that Google placement moves down if I modify static pages and over time it moves back up if those page don’t change again. I think part of this “trusted link” deal more has to do with the length of time a page goes unchanged than how many natural or unnatural links are pointing to it.

 

Trusted sites like our friends over at O'Reilly? Oh, wait a second - they sell text links for big bucks for the sole purpose of SE manipulation. So, Google is saying we should not trust them?

Don't trust O'Reilly? Come on!

Now I am confused....

 

I'm confused too or I would be sitting on the top floor of the Luxor drinking a cold one and having the time of my life. I'm just speculating like everyone about somethings I have thought about, recently.

 

Not to be abrupt about it, but Google has no idea which links are natural vs. which are "unnatural". That is my point.

 

I agree with ya 100%. However, to them natural might mean "trusted" not natural the way the rest of us think about that term. If they trust a site, then to them any link on those sites is natural, while if they don't trust the site, any link from there is not natural. That's my point. Sorry, if I didn't make that clear in my previous post.

 

Yes. Why would anyone want to buy a link from a site that has been PR7 for several years?

What worries me about statements like these emanating from Google lately is that i think it means one of two things:

1. Google THINKS they can distinguish natural from unnatural links - but I don't have great confidence they can do that accurately - see Lorien's post above

2. Google knows damn well they can't but have taken to releasing BS statements like this in some cockamamie notion that this will convince people they really still know what they're doing and we should all ignore the fact that at the moment they have royally mucked up their index

 

They can't always tell which ones are natural and which ones are unnatural, BUT there are certain conditions in which they CAN tell. Ex: Reciprocal links are obviously one category of links that are mostly unnatural and Google might start discounting or ignoring these links at some stage (unless they already do so). Also, links within the same IP group can also indicate that the sites have the same owner and the owner is just interlinking these sites to increase link popularity. Pages with many links on them (directories) or pages that contain the following text "sponsored links", "our sponsors" etc. just above the links might also indicate that these links have been purchase (with cash, reciprocal link, sexual favours .... whatever) .... So, it's not a perfect methodology but they should be able to tell in a lot of occasions which links are "unnatural".

 

Take a look at http://radar.oreilly.com/archives/2005/08/search_engine_s_2.html - I would submit that O'Reilly is a trusted website, but do not play by Googles' rules for text link advertising AND get away with it. Anyone else "caught" using "bought links" or other "unnatural" linking are penalized.

But, I think we are in agreement that Google can't do what they claim they can.

 

That is simply not true. Relevant reciprocal links are the most natural of links in a purely organic linking universe. What's unnatural are non-relevant reciprocal links.

With todays virtual servers, that is not necessarily true either.

Furthermore, I don't see any evidence that Google is doing either of these things. Do you?

 

Indeed. If that were so, 90% of the blogs would dissapear from the SERPs - relevant reciprocal links drive blogs to the top of SERPs.

Ugh - and don't start with the "Google can tell which links are relevant." Because they can't.

 

Please don't think I'm on the side of Google in this issue, I'm just trying to get inside what they might be thinking at Google. Remember this whole PR idea was theirs to begin with and while some have found ways to beat it, my guess is that they will stick with some kind of link value system until they are the largest or smallest company in the world. They believe in that system and promote it at every turn, even while some folks were tearing it apart the past few years.

 

But also remember that PR is REALLY simple - it can't determine "intent" or measure "quality" - it is very simply an indicator of potential based on counts of links pointing to web pages. Nothing more and nothing less.

 

Wouldn't google be able to use the same methology they use to serve relevant adsense ads on sites, to also determine relevant links on my site? Or no?

For instance, they serve adsense ads to my site related to classifieds and photos which are always accurate to my content. Would they not be able to then determine by keywords and content which links are not related?

Any thoughts on this?

Also, if google can determine what is natural and what is not natural, how come out of the sites displaying in my link: query, the majority of sites listed are free directories? Does this mean they view directories as natural links?

 

I've been wondering about this for long time.

If you have a directory site, a succession of links is part of the structure, as it occurs at DMOZ.

Is that un-natural to be indexed or ranked?

 

Haha I almost spat my drink out from the laughing. Thats a pretty wreckless claim on behalf of google since they can't even identify guestbook spam links I wouldn't expect them to be able to identify purchased links. I think google is starting to give googlebot a little bit too much credit.

 

I would think that Google analysis for each backlink involves repetition. You could not say "natural links" happen when the same text for backlinks are used 33 gazillion times. I know some are trying to use several different anchor texts as a fooling measure to slip one past Google.

Let's get real guys. Google have got Gazillions of money to pay the best researching guys in the world to find "unnatural links". Let's not be too gullible about this.