Is 100x1 =1x100, with links Options

[Aji]

Hi all,

How far is this true,

"Both Google and Yahoo don't care if links come from one site or a thousand sites. They care about the page that is providing the link. Whether you have a thousand links from a thousand sites or a thousand links from one site makes no difference---think about it this way: if they discriminated against links coming from single domains they would be discrimintating against one of the best ways to measure a sites content, which is the volume of pages a site possesses. Think about the true titans of the internet---they all have thousands of pages."

I read it in a newsletter, how far is this true.

Aji

[awall19]

Many different quality sources is generally better than many links from one quality resource.

Things are calculated out on a page by page basis, but it would be foolish to think that search engines do not have some sort of protection in place against using a single site to promote a ton of others.

Don't get me wrong, more links is better, but having links from many sites is better than having links from just one (assuming similar quality sites)

[Brian Turner]

Hi all,

How far is this true,

"Both Google and Yahoo don't care if links come from one site or a thousand sites. They care about the page that is providing the link. Whether you have a thousand links from a thousand sites or a thousand links from one site makes no difference---think about it this way: if they discriminated against links coming from single domains they would be discrimintating against one of the best ways to measure a sites content, which is the volume of pages a site possesses. Think about the true titans of the internet---they all have thousands of pages."

I read it in a newsletter, how far is this true.

Aji

As with general back to basics SEO, a wide range of IP blocks linking to your site/pages is best - but there still remain advantages of thousands of links from the same domain. Not in isolation, though.

[Scottie]

Think about it logically...

Link popularity is a measure of what the rest of the "community" thinks of your site. If lots of sites link to yours, it means lots of individuals think enough of your site to link to it.

If one big site links to you from all pages, that's an important vote too, but it doesn't say much about what the rest of the community thinks of your site.

It makes sense for a search engine to look at how many unique domains are linking to your site as well as how many pages... they get a clearer picture of the "importance" of your site that way.

[Jill]

Aji, please don't quote from another source without crediting that source as it's copyright infringement.

Please provide us with the source and/or delete the quote and paraphrase it instead.

Jill

[Aji]

Hi Jill,

I mentioned newsletter, just to hide the identity of the person who mailed me. This is a part of a personal mail from a link manager, who asked for a link exchange. He was very sure of this technique(100 links from a site is equivalent to 100 links from 100 sites), so just wanted to confirm it. I think it works with some of the current SE algos.

There is no copyright issue.

Aji

[Jill]

Oh so in other words, he just made that up. Which actually makes more sense. I didn't think any reputable newsletter publisher would say such a thing.

Of course links from domains that are not from the same domain are going to matter more than ones from the same domain. That's simple common sense.

Jill

[Aji]

We get around 1000 links from a site, which consistutes of a forum as well. I think this has helped us to get a better rank in SERPs, but I am not sure how much was the contribution from those 1000 links.

Aji

[Ron Carnell]

I'm going to disagree with our esteemed panel of experts for two reasons. (IMG:http://www.highrankings.com/forum/style_emoticons/default/smile.gif)

Reason #1: While I'm sure the original formulas published at Stanford have changed substantially since first being implemented, I also think those original formulas probably reflect Google's *philosophy* of links, which likely has changed much less. In those formulas, a link is a link is a link.

Reason #2: Links from within a site are STILL a valid indication of importance. I have 8,000 links pointing to my home page and three links pointing to an article page buried four levels deep. I don't consciously do that to increase the PR of the home page. I do it because the home page will *always* be more important than the article.

[Jill]

I agree that 1000 links from one site certainly do help, but I disagree that they would be counted exactly the same as 1000 links from completely different sites.

It just defies common sense.

[Brian Turner]

I agree that 1000 links from one site certainly do help, but I disagree that they would be counted exactly the same as 1000 links from completely different sites.

It just defies common sense.

I have pretty much seen this happen on my own sites with regards to Google - I've added links to domain1 from thousands of pages on domain2 (IP on a different c block) and seen NO effect on the keyword I was targeted for domain1. A bit disturbing really. I actually started to notice this particularly around September last year. I'm under the impression that IP spread is less of an issue with Yahoo!, but I've yet to test this fully.

[Jill]

Brian, can you clarify?

Are you saying that you've seen with Google that the 1000 links from one site (or one IP block) don't carry much weight? (I think that's what you're saying, but just want to be clear.)

Thanks!

Jill

[Ron Carnell]

I agree that 1000 links from one site certainly do help, but I disagree that they would be counted exactly the same as 1000 links from completely different sites.

It just defies common sense.

LOL. So do relativity and quantum mechanics, Jill. These are mathematicians we're talking about, after all. (IMG:http://www.highrankings.com/forum/style_emoticons/default/smile.gif)

Remember when everyone was convinced that links from ODP and Yahoo counted more heavily than other links? Google has repeatedly denied that, and I believe them (as, I suspect, you probably do, too). The PR formula, as originally implemented, is a statistical measure that very carefully refuses to differentiate one link from any other link. Could it be modified to contained weighted results? Sure, but that would be like conducting a statistical poll where smart people's opinion counted more heavily than average people's opinions did. Such a poll would still be heuristic, but it would no longer be statistical, and more importantly in my opinion, it would no longer be elegant. Math geeks really, really, really like elegant. (IMG:http://www.highrankings.com/forum/style_emoticons/default/smile.gif)

Nonetheless, Jill, I think your common sense approach still applies. Just in a different way.

A thousand links from completely different sites WILL count more, I think, simply because those thousand links are more likely to be counted.

If I put up a new site tomorrow with 1,000 pages, what are the chances that all the pages will be crawled and indexed? General consensus is that the depth of a crawl is going to depend on the PR of the linking pages, usually meaning the home page. Those 1,000 inter-site links to my home page are going to potentially raise the PR to my home page. But if my home page starts out with a low PR (under 4, if my experience counts for anything), most of those 1,000 pages will never be crawled. It's a chicken-and-egg thing, and the only way to break it is to get links from outside the web site.

Okay, so I get some external links, raise the home page PR, and my 1,000 other pages now get crawled. Do they count as much? I believe they do. But they are all fairly low PR pages (assuming they get all their weight from the home page), and because of the steep logarithmic nature of PR calculations, they no longer have the dramatic effect I might have once expected. Put another way, those 1,000 pages would have surely taken my home page from PR0 to PR4 in heartbeat if I could have gotten them indexed. But now that I'm a PR4 (mostly because of external links) and can actually get them indexed, they may not be enough to take me up to PR5. The hill is simply a whole lot steeper. If I now add 10,000 pages, I'm right back to the chicken-and-egg problem of getting them indexed.

It's actually a very elegant system. (IMG:http://www.highrankings.com/forum/style_emoticons/default/smile.gif)

In summary, I believe a link is a link is a link, and the algorithm looks ONLY at Page Rank when calculating Page Rank, caring not a diddle whether the link is on the same site, the same server, or the same IP block. I know that was the way the math was originally developed, and while I'm aware it might have subsequently changed, I just don't think it has. Mostly because it doesn't have to change. It works better without complicating it with weighting factors.

(IMG:http://www.highrankings.com/forum/style_emoticons/default/smile.gif)

[Jill]

Ron, I agree that strictly based on PageRank that what you're saying is most likely correct.

BUT...

There's more to Google's algorithm than PageRank (which you of course know!) and therefore I believe it would be easy enough for them to add a little something to the ranking algorithm that takes into account whether or not the PageRank is all generated from one site or a whole bunch of sites.

Granted, in the past, I believe that there was less emphasis on this, but it appears (anecdotally) that they may be making more concerted efforts to not count every link equally in one way or another.

LOL. So do relativity and quantum mechanics, Jill. These are mathematicians we're talking about, after all.

Ackk! Now you sound like my husband! (IMG:http://www.highrankings.com/forum/style_emoticons/default/wink.gif)

[cline]

This really looks like one of those arguments that is designed to deceive by saying true or nearly true things.

Both Google and Yahoo don't care if links come from one site or a thousand sites. They care about the page that is providing the link.

It is true that they care about the page that is providing the link, and not the site; however, this does not warrant jumping to the conclusion that multiple links on the same site are equal to links from multiple sites, all other things being equal.

Whether you have a thousand links from a thousand sites or a thousand links from one site makes no difference---think about it this way: if they discriminated against links coming from single domains they would be discrimintating against one of the best ways to measure a sites content, which is the volume of pages a site possesses. Think about the true titans of the internet---they all have thousands of pages.

Again, more true statements used to lead to false conclusions. Yes, in general, more content is better. But what does this have to do with discriminating against multiple links coming from single domains? And why that loaded word "discriminate"?