is outgoing links weakening me ?

[janeth]

Originally Posted by inertia

Linking out can hurt you, if you link to crap. Well known fact. So it stands to reason that linking to good/relevant sites can help you. If google can figure out that you've linked to crap then it can figure out that you've linked to quality and relevance also. Surely!?

lol

Do you understand that linking out causes you to loose PR?

And that last year someone wrote a white paper about Trust Rank and no one ever proved it was being used.

[MuNKyonline]

I'm thinking that due to the fact that the TrustRank name is a few years old now and we've heard nothing new from Google on the subject that this is more of a label set by the internet communities to define a filter that search engines use to determine spam sites.

There is no patent for such a thing in the name of Google, there are only a couple of patents related to this subject that are owned by Yahoo.

So in terms of proof that it exists, its pretty likely, it may not be called trustrank but I believe that there has to be something, especially in Googles algorithm that will account for out bound linking and its relevancy to spam technques.

[janeth]

Originally Posted by MuNKyonline

I'm thinking that due to the fact that the TrustRank name is a few years old now and we've heard nothing new from Google on the subject that this is more of a label set by the internet communities to define a filter that search engines use to determine spam sites.

There is no patent for such a thing in the name of Google, there are only a couple of patents related to this subject that are owned by Yahoo.

So in terms of proof that it exists, its pretty likely, it may not be called trustrank but I believe that there has to be something, especially in Googles algorythm that will account for out bound linking and its relevancy to spam technques.

Because you feel it is so, it should be so?

[kgun]

Some important post from this thread as I see it.

Originally Posted by janeth

Maybe you should start by reading this

Pagerank Explained. Google's PageRank and how to make the most of it.

Originally Posted by kgun

Now, I wonder, have you read it yourself?



More info here:

1. Google PageRank - Introduction
2. How Google crawls my site

Can you explain exactly what rel="nofollow" does?


Since so few know the attribute, can it end up as a sort of segregation?

Originally Posted by Webnauts

Google says:

"A Google feature that helps determine the rank of a site in our search results. PageRank relies on the uniquely democratic nature of the web by using its vast link structure as an indicator of an individual page's value. Important, high-quality sites receive a higher PageRank, which Google remembers each time it conducts a search. Google combines PageRank with sophisticated text-matching techniques to find pages that are both important and relevant to your searches."

Source: PageRank


So if OBLs harm, where do you see the democracy? If my site is linked from other sites, but I am not linking to anybody, is that what Google means with democracy?

1. If you read the article linked to in Janeth's post, you distribute away pagerank if you link out (at least without rel="nofollow") so you loose (unadjusted - if there is such a metric) pagerank.
2. But there may be what I will call an adaptive (adjusted) pagerank that dependes on
3. Note Johns post cited above, more precisely:Important, high-quality sites receive a higher PageRank, which Google remembers each time it conducts a search. Google combines PageRank with sophisticated text-matching techniques to find pages that are both important and relevant to your searches. (Is this only related to SERP's?)

[crankydave]

Originally Posted by ctabuk

NO NO NO - Dave - you are actually leading this thread away - what I am saying and webnauts, munky,inertia etc are saying is that if you link out to an 'authority site' in your trade or profession.

Now supposing a site was created called Master Jewelers of America - and it featured the most incredible articles and pages and pages of content relating to your profession. Then it formed a trade recognised board of Directors and held training courses in your universities and had articles written about it in Forbes - now if you had the chance to link your site there - would you turn down the opportunity?

Not at all David. I am remaining true to what I've said all along... link out only when and where it's appropriate and only for the benefit of your visitors AND that I'd personally not recommend linking out strictly for the "perceived" value a SE might place on the OBL.

The links I posted illustrate that it's not the extension of a domain that determines "authority" but the information that it contains.

Here is what you posted...

CD - Authority sites - in the UK you have to obtain permission to use a .gov and as virtually all of my OBL's are .govs is it the same in the US?

You directly imply that "qualifying" to use a .gov extension "automatically" makes you an authority. I disagree and provided a couple of links that illustrate why I disagree.

Then you posted this...

Janeth - you are either being exceeding ill informed somewhere along the line but here is an 'authority site' I link to it because it is 'topic related' and that is the whole point of this discussion. Think SEARCHER link to sites that are in your trade or profession.

Here it is Home : Housing Corporation

Again, the direct implication is that a site with a .gov extension is "automatically" an "authority site". Which again, I disagree for the above reasons.

David... would you link to or want a link from either of the 2 pages I used to illustrate my position because they are "authorities" because of a .gov extension?

Dave

[kgun]

Originally Posted by MuNKyonline

I'm thinking that due to the fact that the TrustRank name is a few years old now and we've heard nothing new from Google on the subject that this is more of a label set by the internet communities to define a filter that search engines use to determine spam sites.

There is no patent for such a thing in the name of Google, there are only a couple of patents related to this subject that are owned by Yahoo.

So in terms of proof that it exists, its pretty likely, it may not be called trustrank but I believe that there has to be something, especially in Googles algorithm that will account for out bound linking and its relevancy to spam technques.

Can I patent

Semantic Link Rank?

1. This Advanced semantic linking and transclusion. is a semantic link (related to the topic discussed in this thread).
2. Is this an authority or trust or both link?

[inertia]

Originally Posted by janeth

lol

Do you understand that linking out causes you to loose PR?

And that last year someone wrote a white paper about Trust Rank and no one ever proved it was being used.

I am leaving to bash my head against a brick wall. It will be more productive.

[crankydave]

Originally Posted by kgun

[*]Note Johns post cited above, more precisely:Important, high-quality sites receive a higher PageRank, which Google remembers each time it conducts a search. Google combines PageRank with sophisticated text-matching techniques to find pages that are both important and relevant to your searches. (Is this only related to SERP's?)[/LIST]

Yes, this is Google's premise for their PageRank however I don't see it as being automatically indicative of a "high-quality" site for this reason...

A site that recieves a single link from a PR6 page can indeed receive a PR5 as displayed by the toolbar. Is that site a high-quality site? Is it of higher quality than a site that receives hundreds of links but is only a PR3 as measured by the toolbar? Why?

Dave

[kgun]

Originally Posted by crankydave

Yes, this is Google's premise for their PageRank however I don't see it as being automatically indicative of a "high-quality" site for this reason...

A site that recieves a single link from a PR6 page can indeed receive a PR5 as displayed by the toolbar. Is that site a high-quality site? Is it of higher quality than a site that receives hundreds of links but is only a PR3 as measured by the toolbar? Why?

Dave

sophisticated text-matching techniques

1. may be (are according to the cited statement?) used to adapt / adjust the (final toolbar or internal) PageRank.
2. So the overall net effect of linking out may vary from context to context and be very difficult to measure. You know my opinion on univariate informal testing.

[janeth]

Originally Posted by inertia

I am leaving to bash my head against a brick wall. It will be more productive.

I think for both of us, because for some reason you just don't get it. And thats really sad when you are charging people money for SEO work.

[inertia]

BASH BASH BASH!!!

Dont expand the issue by saying it's "really sad when you are charging people money for SEO work" based on one disagreement about outbound links! You are trying to damage my reputation by implying i shouldnt be selling SEO because we disagree about one thing.

I get what youre saying, outbound links loose your site PR - i agree! But i would rather google know I link out to valuable resources for my visitors than worry about some bloody PR!

Are you on drugs woman?!? Have you run out of glue to sniff?!

I think i should reiterate this point:

You are trying to damage my reputation by implying i shouldnt be selling SEO because we disagree about one thing.

[kevsta]

to be fair i don't think threads like this do anyone's reputation all that much good.

sure liven the place up a bit though...

[janeth]

Originally Posted by inertia

BASH BASH BASH!!!

Dont expand the issue by saying it's "really sad when you are charging people money for SEO work" based on one disagreement about outbound links! You are trying to damage my reputation by implying i shouldnt be selling SEO because we disagree about one thing.

We are not disagreeing, I've tried to show you how the search engine work and what they are doing. You've argued with what you and others think might be true.

I would not want someone working off what they think or what others had told them. If they were working on my site. I would want them to work off what they know to be true.

You stated information as fact that was totally untrue and you do things on other people's sites because you think it might help.

But not understanding how thing work and letting your pride get in the way of your learning you stand a chance of doing more damage than good. The sad part is that it want be on your site but to a clients site.

Originally Posted by inertia

I get what youre saying, outbound links loose your site PR - i agree! But i would rather google know I link out to valuable resources for my visitors than worry about some bloody PR!

When we started this thread you were linking out because you thought it helped your ranking.

You've made personal attacks and have accused me of things because you are unable to accept the fact that you had gotten some bad information.

Originally Posted by inertia

Are you on drugs woman?!? Have you run out of glue to sniff?!

lol, I'd choose my words more carefully.

Originally Posted by inertia

I think i should reiterate this point:

You are trying to damage my reputation by implying i shouldnt be selling SEO because we disagree about one thing.

We are not disagreeing, I've tried to show you how the search engine work and what they are doing. You've argued with what you and others think might be true.

I would not want someone working off what they think or what others had told them. If they were working on my site. I would want them to work off what they know to be true.

You stated information as fact that was totally untrue and you do things on other people's sites because you think it might help.

But not understanding how thing work and letting your pride get in the way of your learning you stand a chance of doing more damage than good. The sad part is that it want be on your site but to a clients site.

[crankydave]

Originally Posted by kgun

sophisticated text-matching techniques

1. may be (are according to the cited statement?) used to adapt / adjust the (final toolbar or internal) PageRank.
2. So the overall net effect of linking out may vary from context to context and be very difficult to measure. You know my opinion on univariate informal testing.

This would neccessitate a change in the way PR is calculated. Since this statement by Google has been quoted verbatim since at least as early as 2003, using it as a citation for adapting /adjusting PR, internally or externally, would not be accurate IMO. That's not to say it's impossible for it to being done now just that the citation shouldn't be used for that basis since it's at least 5 years old.

Yes, I agree that it's possible the overall net effect of linking out can vary contexually and yes it is certainly very difficult to measure based on the information that is made available to us.

Dave

[crankydave]

I'm going to echo Mike's sentiments posted in another thread...

Play nice. I'll even say please... and thank you.

Dave

[kgun]

Originally Posted by crankydave

Yes, I agree that it's possible the overall net effect of linking out can vary contexually and yes it is certainly very difficult to measure based on the information that is made available to us.

I sign that quote with my WPW digital signature.

[janeth]

Originally Posted by kevsta

to be fair i don't think threads like this do anyone's reputation all that much good.

sure liven the place up a bit though...

I agree but since I'm not an SEO I've got nothing to loose. And I do enjoy a good debate. Although this one has about died out.

[janeth]

Originally Posted by crankydave

This would neccessitate a change in the way PR is calculated. Since this statement by Google has been quoted verbatim since at least as early as 2003, using it as a citation for adapting /adjusting PR, internally or externally, would not be accurate IMO. That's not to say it's impossible for it to being done now just that the citation shouldn't be used for that basis since it's at least 5 years old.

Yes, I agree that it's possible the overall net effect of linking out can vary contexually and yes it is certainly very difficult to measure based on the information that is made available to us.

Dave

I agree. Without all the information that Google is using there would be no way to test. And no way of knowing rather Google is using it or not.

As Dave has said before. I'd link out for my visitors and pass PR or not pass PR according to the way I feel about the sites I'm linking to.

As well as the number.

But I would not do it in hopes of getting an unknown boost on the rankings. Nor would I tell my clients it would help them to get ranked.

[Webnauts]

Am I also an authority site? Section 508: The Road to Accessibility

Or check here some more: linkdomain:webnauts.net site:.edu - Yahoo! Search Results

Hey, just kidding...

[Webnauts]

Hm, looks like Google are telling BS themselves:

"...Several people at the BlogHer session asked about linking out to other sites. Won't this cause your readers to abandon your site? Won't this cause you to "leak out" your PageRank? No, and no..."

Source: Official Google Webmaster Central Blog: BlogHer 2007: Building your audience

I didn't say anything. Google did that. So please don't beat me up!

[inertia]

Originally Posted by janeth

You've made personal attacks and have accused me of things

Cant find any, apart from my last post.

Originally Posted by janeths sly personal attacks

And you're here to protect Google? I'm sure Google will give you a cookie.

Are you making this up as you go along.

You should run for president you change like the wind.

Because for some reason you just don't get it. And thats really sad when you are charging people money for SEO work.

I apologise to all forum users for my last post, it was enticed by the generalised, off topic comments above. Apart from my last post I dont feel I made any personal attacks. Ill bow out now.

[janeth]

Originally Posted by Webnauts

Hm, looks like Google are telling BS themselves:



Source: Official Google Webmaster Central Blog: BlogHer 2007: Building your audience

I didn't say anything. Google did that. So please don't beat me up!

It looks to me like an interview with Elise Bauer. I don't think Elise Bauer works for Google does she?

am I missing something?

[Webnauts]

Originally Posted by janeth

It looks to me like an interview with Elise Bauer. I don't think Elise Bauer works for Google does she?

am I missing something?

Sure!!! Google is posting on their own webmaster central inaccurate information! Google, google ...

Sorry, but this is getting too ridiculous for me. In other words far below my level.

Anyway, I wish you all a nice discussion.

[crankydave]

Originally Posted by Webnauts

Hm, looks like Google are telling BS themselves:

Source: Official Google Webmaster Central Blog: BlogHer 2007: Building your audience

I didn't say anything. Google did that. So please don't beat me up!

LOL John...

But you left out this part of the quote that was in reference to "leak out"...

And PageRank isn't a set of scales, where incoming links are weighted against outgoing ones and cancel each other out.

Now, either the author doesn't understand what exactly PR "leak" is or they are addressing all those who misunderstand it and what a simple and very narrow concept it refers to. It's very easy to say it "doesn't exist" to all the folks that misunderstand it rather than to try and explain it. I've tried to do the latter but it's always turned into a full blown mess. Perhaps I'll try one more time for you John and see if it makes sense.

1. A page has a finite amount of PageRank it can pass to other pages.
   
2. That PageRank is passed via the links on the page.
   
3. Those links on the page can be external links or internal links.
   
4. Any amount of PR that is passed via external links is not passed within the site.
   
5. The amount of PR that is passed, or "voted" if you like, by the external link is commonly referred to as being "leaked" because it is not being passed internally.

Example 1... If all the links on a page are internal links then 100% of the available PR is being passed (voted) within the site therefore there is no leak.

Example 2... If all the links on a page are external links then all the available PR is being passed (voted) outside of the site therefore there is a 100% leak.

All a PR leak refers to is the amount of PR that is passed (voted) to external sites that is not being passed (voted) within the site. Anytime less than 100% of the available PR is being passed (voted) within the site, the amount that is not being passed (voted) internally is commonly referred to as a "leak".

Dave

[kevsta]

right, can we make this post a sticky for noobs please? this is the simplest most concise definition Ive seen.

i will be linking to this post and possibly plagiarising it for confused customers from now on

Originally Posted by crankydave

1. A page has a finite amount of PageRank it can pass to other pages.
2. That PageRank is passed via the links on the page.
3. Those links on the page can be external links or internal links.
4. Any amount of PR that is passed via external links is not passed within the site.
5. The amount of PR that is passed, or "voted" if you like, by the external link is commonly referred to as being "leaked" because it is not being passed internally.

Example 1... If all the links on a page are internal links then 100% available PR is being passed (voted) within the site therefore there is no leak.

Example 2... If all the links on a page are external links then all the available PR is being passed (voted) outside of the site therefore there is a 100% leak.

All a PR leak refers to is the amount of PR that is passed (voted) to external sites that is not being passed (voted) within the site. Anytime less than 100% of the available PR is being passed (voted) within the site, the amount that is not being passed (voted) internally is commonly referred to as a "leak".

Dave

[janeth]

Originally Posted by Webnauts

Sure!!! Google is posting on their own webmaster central inaccurate information! Google, google ...

Sorry, but this is getting too ridiculous for me. In other words far below my level.

Anyway, I wish you all a nice discussion.

You do not believe that links leak PR?

[Webnauts]

Originally Posted by crankydave

LOL John...

But you left out this part of the quote that was in reference to "leak out"...



Now, either the author doesn't understand what exactly PR "leak" is or they are addressing all those who misunderstand it and what a simple and very narrow concept it refers to. It's very easy to say it "doesn't exist" to all the folks that misunderstand it rather than to try and explain it. I've tried to do the latter but it's always turned into a full blown mess. Perhaps I'll try one more time for you John and see if it makes sense.

1. A page has a finite amount of PageRank it can pass to other pages.
2. That PageRank is passed via the links on the page.
3. Those links on the page can be external links or internal links.
4. Any amount of PR that is passed via external links is not passed within the site.
5. The amount of PR that is passed, or "voted" if you like, by the external link is commonly referred to as being "leaked" because it is not being passed internally.

Example 1... If all the links on a page are internal links then 100% of the available PR is being passed (voted) within the site therefore there is no leak.

Example 2... If all the links on a page are external links then all the available PR is being passed (voted) outside of the site therefore there is a 100% leak.

All a PR leak refers to is the amount of PR that is passed (voted) to external sites that is not being passed (voted) within the site. Anytime less than 100% of the available PR is being passed (voted) within the site, the amount that is not being passed (voted) internally is commonly referred to as a "leak".

Dave

I fully agree with your David. Very well explained. I hope you don't mind, but I had to give you a big Rep+ vote!

[Webnauts]

Originally Posted by janeth

You do not believe that links leak PR?

I hope Kgun would have been here now. Maybe he will come later? Wait and see.
He will sure ask the question:

How came first? The chicken or the egg?

I think many here never understood what he always met saying that. But David answered that in his post above at point 1.

[janeth]

Originally Posted by Webnauts

I hope Kgun would have been here now. Maybe he will come later? Wait and see.
He will sure ask the question:

How came first? The chicken or the egg?

I think many here never understood what he always met saying that. But David answered that in his post above at point 1.

I'm very sorry to be beneath you in my posting. I hope one day to be able to rise to your level.

[crankydave]

Originally Posted by Webnauts

I fully agree with your David. Very well explained. I hope you don't mind, but I had to give you a big Rep+ vote!

Thank you John! I think this is the first time I'm been able to explain it where it made sense.

Originally Posted by kevsta

right, can we make this post a sticky for noobs please? this is the simplest most concise definition Ive seen.

i will be linking to this post and possibly plagiarising it for confused customers from now on

And thank you very much kevsta!

Dave

[kgun]

Originally Posted by kevsta

right, can we make this post a sticky for noobs please? this is the simplest most concise definition Ive seen.

i will be linking to this post and possibly plagiarising it for confused customers from now on

This is the traditional model and can be found in the important classic links about PageRank cited above. Pagerank is also nonlinear (logarithmic).

[TryUsOut]

This has turned out to be a very interesting thread. I respect each one's opinion and been wallowing on some good points that were raised throughout the discussions. However, this are all speculations for me (as we all don't exactly have all the facts there is to know on how search engine algo works) as I am not a firm believer of everything that I read. I do consider good points and some logical discussions - but I tend to weigh everything based on actual results.

Does anyone here has a site willing to make a little experiment on the following:
1. Linkout to so called "authority sites" or those that have a high trust factor (those having high PR) - so it maybe proven that linking out to such sites does or doesn't help in SERP
2. Do the same thing in number one and see after the next PR update if such links would affect the site's PR

This I think would be very interesting and should make a closure on some of the things that has been debated over this thread.

[kevsta]

Originally Posted by TryUsOut

This has turned out to be a very interesting thread. I respect each one's opinion and been wallowing on some good points that were raised throughout the discussions. However, this are all speculations for me (as we all don't exactly have all the facts there is to know on how search engine algo works) as I am not a firm believer of everything that I read. I do consider good points and some logical discussions - but I tend to weigh everything based on actual results.

Does anyone here has a site willing to make a little experiment on the following:
1. Linkout to so called "authority sites" or those that have a high trust factor (those having high PR) - so it maybe proven that linking out to such sites does or doesn't help in SERP
2. Do the same thing in number one and see after the next PR update if such links would affect the site's PR

This I think would be very interesting and should make a closure on some of the things that has been debated over this thread.

we're talking about the real PR flow which we cant see or measure, not toolbar which we can, but cant trust or believe, so it would prove nothing.

i have various experimental pages on my (mostly experimental) site where i've done this, and others where i haven't. but because they have different content and therefore compete in different searches and we cant measure real PR anyway its nigh on impossible to draw any firm conclusions.

i would say that linking out once or twice to wiki and the like doesn't hurt noticably. it may even help, but i doubt it.

i could remove the links, but one or two links would likely be too small a fraction of the available page PR to make enough difference to prove anything, especially with Google's permanent state of flux these days meaning the positions swing fairly widely day to day anyway.

i have PR4 pages with hardly any content performing in similar fashion to greyed out (toolbar) pages that have good content. (both launched same day strangely enough too)

all i can really state with confidence is that widespread use of nofollow in your own site doesnt seem to hurt rankings and occasional linking out to wiki and the like doesn't either.

you'd need a small site with a lot of external links off the index page to test this i think.