Can inbound links really hurt you?

[kgun]

Originally Posted by crankydave

What? I asked a few questions and then stated that consequences (both good and bad) because of things beyond your control happen all the time. That is fact.

Is the weather beyond your control? Is "Mr Market" beyond your control? Is the economy beyond your control?

So this is your proof?

Don't require more of other members than of yourself.


--------------------------------------------------------------

Originally Posted by kgun

Private copy to me since you have edited too many of my posts where you disagree.

Can inbound links really hurt you?

<this is what happens when I edit a post. See my name at the bottom? You owe me an apology AFAIC CD>

[kgun]

Originally Posted by kgun

<this is what happens when I edit a post. See my name at the bottom? You owe me an apology AFAIC CD>

That may be correct. Does that make any difference?

This

Hello from Ukraine!

post was edited a few days ago with no name at the bottom.

And I urge you to start a new thread if you wan't to continue this discussion. This thread has gone enough off topic in my view.

[Fancy schmancy]

Originally Posted by kgun

<this is what happens when I edit a post. See my name at the bottom? You owe me an apology AFAIC CD>

Who should apologize? Crankydave you should face that the evidence against you is conclusive. And this discussion really doesn’t belong in this thread.

[crankydave]

Originally Posted by kgun

That may be correct. Does that make any difference?

This

Hello from Ukraine!

post was edited a few days ago with no name at the bottom.

And I urge you to start a new thread if you wan't to continue this discussion. This thread has gone enough off topic in my view.

I don't know nor care what your problem today is.

YOU spun this thread into another direction with your baseless, unfounded, ridiculous, insulting, accusations.

And right now, I am quite content in exposing your insulting nonsense for what it is.

I'm tired of it kgun. It's not the first time, not the 2nd, not the 3rd, nor the 4th, etc. You have made a habit of leveling totally baseless and unfounded accusations and insults at mods and WPW.

I'll tolerate it no longer.

Dave

[cz]

Quote:
Originally Posted by deepsand


Specifically how can IBLs HARM - as opposed to doing nothing for - you?

What I mean't by this, is that if you go on a campaign of purchasing IBLs and then stop, for whatever reason, the removal of those IBLs will certainly damage your SERPs.

If your pockets are not deep enough don't go on campaigns of purchasing IBLs even from Googly loved sites, in fact especially not from these sites as the removal slap will be harder..

This one seems to make sense to me.

I have a new PRO site I launched and I exchanged links with a related site. He has a few non-relevant links on his site but they're going to nice sites that sell general merchandise and only a few of them. I hope that doesn't harm me. It's the first link exchange for the new website.

KGUN: I lurk here and for years I would add that I don't understand most of what you post as it doesn't make sense. Plus you act like you know everything and have a chip on your shoulder or the need to be right about every darn thing. CrankyDave contributes solid info that is understandable and is not playing "the big hard-a*s" role all the time, so I'd tend to listen to him more than I would to you - if I could understand what you're talking about 1/2 the time. Sorry but I couldn't contain myself fellow members.

[kgun]

Originally Posted by cz

This one seems to make sense to me.
I have a new PRO site I launched and I exchanged links with a related site. He has a few non-relevant links on his site but they're going to nice sites that sell general merchandise and only a few of them. I hope that doesn't harm me. It's the first link exchange for the new website.

Your quote from deepsand about link selling is another example of why this thread is growing over all boundaries.

Originally Posted by kgun

1. This thread is not about link buying / selling. That has been discussed extensively before / elsewhere (see the may 9th 2009 top links on my social bookmarks.) on WPW.

I can not require that you read everything I cite and quote in this thread. This is my personal opinion.

1. If you can take one minute, click the above link, and note what the first 20 social bookmarks are about you will understand what I mean.
2. I don't see any substantial new information in this marathon thread beyond the first page.
3. That is why I call it a circle squaring thread. Sorry if that sounds arrogant.

These

Link Selling - Unethical or Just Unwise?

The Ongoing PageRank Massacre Hits Big Players

two are on the last page of my social bookmarks (Page 5 if you view 100 bookmarks / page).

[kgun]

Here is my hopefully final answer: (Wait to the link redirects to post #16).

Links::: Life or death of your eBusiness.

[cz]

I hope so kgun,

I can read! I know what the topic of the thread is and the deepsand quote was right on topic because an inbound, is an inbound - bought, swapped, whatever.

I agreed with him because I paid for a link campaign for yrs on a relevant forum linking out to one of my sites.

When that campaign was stopped - within 60 days listings, customers, sales plummeted. That is my experience validating his opinion kgun.

My other sentence was an example of how my new link swap illustrates what others have posted - it's not the inbound link that can get you so much as the link going out.

So I guess you're saying my opinion and real life experience of his quote is dog-meat ay? I've been subscribed to this thread since the beginning and didn't post to spam or anything. I'll butt out sou you can have the last word, as usual.

[kgun]

Originally Posted by cz

I can read! I know what the topic of the thread is and the deepsand quote was right on topic because an inbound, is an inbound - bought, swapped, whatever.

I think deepsand is the best to answer that question. I shall not exclude that deepsand brought new arguments to the link buying / selling equation that has not been discussed elsewhere.

[deepsand]

Originally Posted by cz

I know what the topic of the thread is and the deepsand quote was right on topic because an inbound, is an inbound - bought, swapped, whatever.

I agreed with him because I paid for a link campaign for yrs on a relevant forum linking out to one of my sites.

When that campaign was stopped - within 60 days listings, customers, sales plummeted. That is my experience validating his opinion kgun.

For sake of clarification, please understand that kgun is not disagreeing with you re. the effect(s) of the loss of any IBLs; rather, it is the case that both he & I maintain, not as an opinion, but as a matter of fact, as is demonstrable from the PR algorithm itself, that such cannot constitute a penalty.

What kgun & I have consistently objected to is the use of anecdotal "evidence" as "proof" that such a "penalty" exists. As oft times previously noted, while such a penalty may is theoretically possible, it must be part of the SERP algorithm; and, to date, there is no empirical proof of such.

[Feydakin]

Except, of course, for the tests that were run and showed that it could be done under certain circumstances..

[deepsand]

Citation(s) required.

[Feydakin]

Originally Posted by deepsand

Citation(s) required.

You used the wrong word.. It should read

Citation(s) wanted.

[williamc]

I have a feeling Feydakin is speaking of a test I did a while back which CrankyDave was also aware of at the time.

It got a site basically deindexed within 72 hours of placing links from a certain site, which I had a feeling was going to be the result.
PageRank of the linked-to site also zero'd out at the same time.

Was it one of the algo's in place, or was it a hand job? Who can say but Google themselves, but the fact is, it worked.

Now that I think on it, I believe Mel and Janeth may have been privy to that test as well.

But I do so love hearing from all these other experts who have never actually tested a damn thing themselves.

[cz]

deepsand,

I never said anything about a penalty just that I experienced what you mentioned here:

What I mean't by this, is that if you go on a campaign of purchasing IBLs and then stop, for whatever reason, the removal of those IBLs will certainly damage your SERPs.

I can't give any empirical data or tests, but it did happen to me as you say. I think because I lost the juice, not because of a penalty. Now I keep up my links and don't stop my campaigns. I think long and hard about cost and such, before comitting to placing an ad to my site, making reasonably sure I can maintain the costs - especially if it includes ads that slowly pile up to a lot of IBL's.

I'll butt out but I wanted to reply to you deepsand.

[deepsand]

Originally Posted by cz

deepsand,

I never said anything about a penalty ...

Neither said or intended to imply that you did.

I was simply trying to put kgun's post in context for you, particularly in light of his reference to your mention of my post.

You will find that, when matters re. mathematics arise, kgun is likely the more knowledgeable here. And, the PR & SERP algorithms are, simply put, mathematical functions.

[Terry Van Horne]

deepsands whole argument seems to be about negative PR. I don't know that negative is possible but the question was "Can inbound links really hurt you?" IMO, the Google guidelines says that buying links not using "NoFollow" can... so the question really being debated is... the How can they hurt? If I were Google the last place I'd do anything to "hurt"/punish a site is with PageRank. A dampening filter is the easiest way to "hide" the punishment. It also means you can apply it to any value with no trace of the cause.

[kgun]

First of all what are we talking about?

1. PageRank?
2. An existing link?
3. The addition of a link?
4. The removal of a link?
5. A bought link?

Originally Posted by williamc

It got a site basically deindexed within 72 hours of placing links from a certain site, which I had a feeling was going to be the result.

PageRank of the linked-to site also zero'd out at the same time.

If you know the concept of linear correlation, you should know that if the data is scattered evenly in a circular disk except for one statistical outlier, the removal of this outlier brings the correlation coefficient to zero. So the significance of the fitted line depends solely on this isolated outlier.

Similarily with the removal of one significant link. The pagerank can depending on how it is programmed decrease to zero. Nothing magical in that.

Originally Posted by williamc

Was it one of the algo's in place, or was it a hand job? Who can say but Google themselves, but the fact is, it worked.

I can see how a manual job can do that. Report the site to Google as black or grey and being belived. But that is another story.

Originally Posted by williamc

Now that I think on it, I believe Mel and Janeth may have been privy to that test as well.

You are no more correct even if the whole world agree with you.

Originally Posted by williamc

But I do so love hearing from all these other experts who have never actually tested a damn thing themselves.

Can you mention any? Do you think of SEO experts?

Originally Posted by Feydakin

Citation(s) wanted.

Uncientific answer.

Links, the life and death of your eBusiness.

To repeat, in order of my priority. Are we talking about?

1. PageRank?
2. An existing link?
3. The addition of a link?
4. The removal of a link?
5. A bought link?

I am talking about 1, be it toolbar, directory or internal.

As an economist and a quasi mathematician I assume that it can not be negative.

If you are talking about SERP position the effect of IBL's have to be judged among a lot of other factors. That opens for a lot of speculation and informal guessing without proof.

Does this Creating a Science of the Web | Web Science Research Initiative URL need links to be used by surfers? Does this Yahoo! site need a search engine to be used?

[crankydave]

Originally Posted by kgun

]I am talking about 1, be it toolbar, directory or internal.

Then you are way off topic. This thread is not about pagerank.

Dave

[kgun]

About SERP position?

[williamc]

Originally Posted by kgun

Similarily with the removal of one significant link. The pagerank can depending on how it is programmed decrease to zero. Nothing magical in that.

Maybe you should reread my post that you quoted above Kgun. I did not say a link was removed, I said when links from a certain site were ADDED.

I assume you are posting too fast my friend, as you should never have missed that normally.

[kgun]

Originally Posted by williamc

Maybe you should reread my post that you quoted above Kgun. I did not say a link was removed, I said when links from a certain site were ADDED.

1. Do you know how many IBL's that site had before the addition of that link?
2. Do you know the Toobar PageRank before the link was added?
3. I assume that you do not know the internal pagerank

[kgun]

It is late night in Norway. I have to go to bed since you use too long time to reply.

Good news:

I leave now with a screendump of this page that may be pasted into a PDF document later..

Bad news:

I may return tomorrow or in 2012 when Microsoft get their new operating system

[williamc]

Originally Posted by kgun

1. Do you know how many IBL's that site had before the addition of that link?
2. Do you know the Toobar PageRank before the link was added?
3. I assume that you do not know the internal pagerank

Seeing as how it was built as a test site, and I controlled all links to it, yes, I am certain no other links were lost. The PageRank of the site was static for several months until 72 hours after that test at which time it lost both static rankings and PR.

Kgun: you are trying to beat what is known with hypothesis. It can not be done, unless there is sufficient new supportive evidence to the contrary.

As I said, it may very well have been a handjob, however, that is neither here nor there, as it incurred a penalty regardless of which method of doling out said penalty was used.

The thread was at least partially about "can links hurt you". In some cases they can yes. This is not science, I leave the guesswork to you, this is merely what was witnessed by numerous people who are not idiots when it comes to optimization.

[crankydave]

Originally Posted by kgun

1. Do you know how many IBL's that site had before the addition of that link?
2. Do you know the Toobar PageRank before the link was added?
3. I assume that you do not know the internal pagerank

Don't mean to step on Will's toes but I can answer those questions...

1. Yes
2. Yes
3. Of course not

Not to mention it was not the only site tested. I brought that up, in this thread, pages ago. You either did not read it, forgot you read it, or ignored it.

Dave

[crankydave]

Originally Posted by kgun

I may return tomorrow or in 2012 when Microsoft get their new operating system

I vote for which ever happens AFTER you post an apology for your baseless accusations against me.

nighty-night

Dave

[kgun]

Originally Posted by crankydave

I vote for which ever happens AFTER you post an apology for your baseless accusations against me.

nighty-night

Dave

If you shall have a chance of getting that, it must be in anothe thread started by a site Admin. I don't like to take threads unnecessarily off topic.

So if we are not talking about ToolBar pagerank are we

1. talking about SERP position?
2. One of Googles many penalties
3. or something else?

Good night.

[crankydave]

Originally Posted by kgun

I don't like to take threads unnecessarily off topic.

Since when?

Originally Posted by kgun

So if we are not talking about ToolBar pagerank are we?

...

You do realize what thread you are posting in do you not?

Dave

[Tubby]

I just noticed his thread "can inbound links really hurt you"Presumably I had ignored it because the topic did not concern me. Sometimes I take a look at a thread because its a day old or more and only has a single post.

Whatever the reason, this morning I noticed this thread. I saw the last post was by crankydave, and there were 777 replies. So I took a peek.

777 replies tells me there must be some sort of squabble, I see names like Kgun, crankydave, cz, kgun, deepsand, Feydakin, kgun, williamc, Terry Van Horne - and to be truthful I did read back a few posts and never quite concluded what is sustaining this thread.

Can someone do a quick summary - I thought inbound links could not hurt me, Is this not a yes or no question? I do not really want to jump in on what looks like a fun thread when I do not know whats been going on under the covers.

[Feydakin]

In a nutshell -

There is a small group of us that tested and confirmed that under certain circumstances you can alter a website's link profile with IBLs to enough of an extent that the site will suffer in it's rankings..

Basically we attacked a few sites with a specific link profile and caused those sites to drop in the rankings..

Those sites all had a few elements in common, a very weak link profile for example, and we were able to control certain types of links aimed at those sites.. But throwing more links at the sites they went down in the rankings..

The contention is that people don't believe us and want data that we are unwilling to give.. Data like the types of links used, the link profiles of the sites attacked, and the URLs of all sites involved.. That information is simply not going to be given out.. And since we refuse to release that data they refuse to believe us..

Again, this can not be done to just "any" old website.. Conditions need to be just right both on the site being attacked and the links being used..

No, it is not a yes or no question as Janeth pointed out many pages ago, Google said "almost" impossible.. So we tested it.. We found that it can be done..

And soon we will see a round of show us the proof and nuh uhs..

[williamc]

Just to reiterate:

Originally Posted by janeth

I was just pointed to a private thread where this was tested and I even participated in the thread (guess I am getting old) and they did make sites loose their rankings by throwing different types of links at them.

But they were not from forums or blogs.

Now, so far, Aaron wall has stated he has seen it done, Eric Ward states that he would put his 13+ year rep on the line that it can be done, Janeth participated in a test thread and witnessed it being done and the results, feydakin and CrankyDave were actual participants and witnessed it being done. And I am the one who brought the test up in the first place after seeing it done to a few sites inadvertantly. I also supplied the "bad" site from which to get links from. I asked Dave and Feydakin in on the test so that it was not merely my own word and to have some others around to show why my explanation might not hold water. Janeth, Mel, and a few others also participated in the thread, going over everything with a fine tooth comb trying to find some/any other explanation for what transpired.

All I see otherwise is a bunch of people upset that their opinions don't seem to hold water against people who have SEEN it used effectively..

[Tubby]

Thanks Feydakin

[Tubby]

Thanks Feydakin. you said.
" and caused those sites to drop in the rankings.."

Over the years I have done plenty of things myself that made my own site suffer in 'rankings.' What 'rankings' are we discussing. and would these changed rankings effect the traffic arriving at the sites?

note
I am on the IBL's cannot harm you, side of the fence -
(but willing to absorb observations made by WPW members)

[deepsand]

Originally Posted by Terry Van Horne

deepsands whole argument seems to be about negative PR.

A careful reading of my posts shows that I did in fact allow for the possibility of a metric re. link "quality" external to PR.

Originally Posted by Terry Van Horne

I don't know that negative is possible ...

By definition, PR is a probability value; and, therefore, cannot lie outside of the range 0 to 1.

Originally Posted by Terry Van Horne

If I were Google the last place I'd do anything to "hurt"/punish a site is with PageRank. A dampening filter is the easiest way to "hide" the punishment. It also means you can apply it to any value with no trace of the cause.

There has been speculation that the so-called "damping factor" in the PR algorithm might be so employed. However, as said factor is itself a probability value, which cannot be negative, it can do no more than wholly discount any PR carried by IBLs. Furthermore, reducing such factor's value to 0 will simultaneously raise the non-link related factor of PR by the ratio 1/(1-d), where d = damping factor value that has been discounted.

Therefore, any metric re. link "quality" must be employed elsewhere; i.e., in the SERP algorithm.

It has also been speculated by some that a site might be de-indexed owing the "quality" of IBLs. However, a moment's reflection will serve to dismiss that notion.

Were a site to be de-indexed, there would be no systemic memory of its existence, with the result that it would eventually be re-indexed, possibly de-indexed again, re-indexed once more, de-indexed again, ad infinitum. As a matter of practical application, the site would remain in the index, with one or more metric values and/or flags serving to alert the SERP engine of certain factors regarding its suitability for being public listed and under what circumstances.

[kgun]

Deepsand.

I can not give you another green rep point now.

Instead you were included here below the Russian Academy of Sciences, but above Harvard. So so long you have moved Harvard down one place.

Tell me if you want the link removed, since somebody has "proved" in this thread that an IBL may hurt your site.

[deepsand]

I am most genuinely flattered; thank you.

And, although I am not the Penn State Ticket Man, as the "man behind the curtain," so to speak, I do have a vested interest in the financial success of that enterprise. Absent any reason to expect harm from your link, I leave the matter in your hands.

[kgun]

To make the anchor text meaningful and the link semantic, I replaced it with a link to your WPW profile.

I can exchange that link with a more personal link, e.g. to a blog or other site that you own.

[kgun]

Originally Posted by deepsand

Therefore, any metric re. link "quality" must be employed elsewhere; i.e., in the SERP algorithm.

Yes, and if a page fluctuates on the SERP's that is quite normal. It is impossible to know

1. Is it a consequence of n additional pages entering into or disappearing from the same competitive part of the m dimensional space where m equals the number of factors in the SERP algorithm.
2. Is it a consequence of a variable effect of these additional factors. You can use this thread as a simple step 1 in your speculation.
3. Or is it the consequence of a added or a deleted link?

Originally Posted by deepsand

It has also been speculated by some that a site might be de-indexed owing the "quality" of IBLs. However, a moment's reflection will serve to dismiss that notion.

In other words is there an add external link penalty that we don't know about? Personally I strongly doubt that. I not only doubt it, but regard it as meaningless viewed against a credibility metric of the algorithms.

[williamc]

Once again, kjell and deepsand are speculating, whereas the testers had control of all aspects of the test and actually know what is what. If you guys wish to concern yourselves with guesswork thats fine with me, but do not mislead others who are actually trying to use facts to get ahead in the engines and stay there. That requires real world experience, something neither of you seem to have.

Kgun: you have a very astute mind and I have enjoyed several debates with you in the past, and we have both come out ahead of the game in those instances, however, this time around, there has been controlled testing done and the results speak volumes. You admit to not knowing googles algo correct? So stop guessing where there have been tests that already show the outcome in some cases. As Feydakin pointed out, it is not a blanket type deal, conditions have to be right both on the linked and linking sites. The original question was answered factually, trying to cloud that is simply decieving and misleading at this point. I realize that you do not argue due to an insecurity complex that you have to always be right like deepsand, but the outcome is clouding the issue regardless.

You said it best on page 3.

Originally Posted by kgun

Don't pollute my theory with facts.

[Feydakin]

Originally Posted by Tubby

Thanks Feydakin. you said.
" and caused those sites to drop in the rankings.."

Over the years I have done plenty of things myself that made my own site suffer in 'rankings.' What 'rankings' are we discussing. and would these changed rankings effect the traffic arriving at the sites?

It was a blanket drop in the SERPs.. Including phrases where they ranked #1.. Things like their own name and local search phrases.. To the extent that all Google traffic vanished..

[Feydakin]

Now for my 'guess' as to what is happening..

I don't think that this is part of the normal algorithm at all.. I think that Google has tagged and flagged many websites as "dirty" on some level.. When those tagged sites link out to new sites, or old sites, with weak link profiles, the amount of 'dirty' that those sites have attributed to them is passed on the to site being linked to..

Since it is new, or has a weak link profile, it is effectively penalized out of the system until the link profile can be countered and made good.. This can be by removing the 'tagged links' or by improving the link profile of the targeted site..

Believe, don't believe.. Not really caring.. I know the premise of the testing, the methods used, and the sites targeted.. And while there is 'nothing' 100% clean in this type of research, we were able to reproduce it effectively enough to know that it can be done..

[crankydave]

Originally Posted by Feydakin

It was a blanket drop in the SERPs.. Including phrases where they ranked #1.. Things like their own name and local search phrases.. To the extent that all Google traffic vanished..

Just to add (although pointed out once pages ago) the drop was precipitous, not just a few spots, and remained until the links were removed. Additionally, the change(s) happened very quickly. Either the site(s) were being watched (or flagged) if a hand job was the case or it was automated or a combination of both. My guess (yes a guess) is a combination.

Dave

[Tubby]

One of many reasons I call into WPW is because its members do observe. . . I observe. Occasionally I make my observations known. I like very much to read and understand these observations. I also like to read speculations. ( I even derive amusement from Kguns occasional assertions of how it aught to be done)

Like most experienced WPW members, I weight the information, observations, speculations, mathematical probabilities in accordance with the source, add to it my own experience and sometimes It assists me to redirect the pathway I am moving along.

Sometimes if feels a bit like driving home on a dark night trying to pick out the particular Kangaroos that are likely to jump out in front of me. . . but a thread like this does enable all of us a little more periferal vision. (I like periferal vision.)

Google to me has always been a hazy unknowable hazard, and I can offer nothing in my own observations to add to this topic. . . Thanks Feydakin for personalising the thread towards my personal query.

[ctabuk]

So what you guys are all saying (in one form or another) is that we now have 2 definitions of 'anchor'

1 Anchor text to lift a site

2 Anchor links that if become attached will drag the site down.

So, if a reputable site suddenly is attacked by this method, what will the SE's (especially Google) do to deal with it?

Are you all saying run xenu at least once a month to see if you have been sabataged. List the links and report them to google. Or continue to add fresh content and add good quality links to and from the site?

[Feydakin]

Ok, last time, little words..

An established site can not be affected in this way.. We tried.. It had 'no effect'.. It requires that the target site be one with a weak, or new, link profile.. I'm not sure why people keep missing this part, it has been said often enough..

This would imply that you can solve the problem by removing the links or by improving your link profile..

[ctabuk]

Without trying to be even slightly obtuse, your statement, albeit it concise is missing one relevant factor.

An established site can and will evolve by adding new pages with new longtailed url's. It's called diversification. So if a new page was created for a fresh search term, then surely it could be 'attacked'?

But having said that, then your methods would I assume work. My thanks. Actually having reread your post - that in effect is what you are saying. Hmmm Interesting. I know that CD asked if I would add a link to his site, which I gladly did - I guess it was part of the experiment.

[kgun]

Originally Posted by Feydakin

An established site can not be affected in this way.. We tried.. It had 'no effect'.. It requires that the target site be one with a weak, or new, link profile.. I'm not sure why people keep missing this part, it has been said often enough..

If logical, the reason should be that the effect is negligible, not observable or identifiable.

williamc

Yes your cite is correct, but I have also written in this thread, that algorithms are not perfect. Algorithms may have side effects.

I doubt that interpreted within a probability context of PageRank (Toolbar or internal) the algorithms have negative side effects.

Interpreted within the more complex SERP algorithm I doubt that the addition of a link from page A to page B can hurt page B. But I think it can have a zero effect.

Citing conclusions without evidence, results and data that can be retested is unscientific to say the least.

Finally CD, my apology for being rude (and frustrated). This has been (is?) a good thread after all. The original poster seem to have left long time ago.

[williamc]

Originally Posted by kgun

Interpreted within the more complex SERP algorithm I doubt that the addition of a link from page A to page B can hurt page B. But I think it can have a zero effect.

You are still hung up on "probabilities" and guesswork. It has already been stated that we do not know if the links being added was simply a flag for review, and a handjob was applied. That "seems" unlikely as the result was reversed right after removing the links, without any reinclusion requests being made which would indeed point to a more algo related nature, but that is an unknown.

[Feydakin]

Originally Posted by kgun

If logical, the reason should be that the effect is negligible, not observable or identifiable.

Except, the effect was observed and identified.. And to the people attacked in this manner it is not negligible..

[kgun]

Originally Posted by williamc

It has already been stated that we do not know if the links being added was simply a flag for review, and a handjob was applied. That "seems" unlikely as the result was reversed right after removing the links, without any reinclusion requests being made which would indeed point to a more algo related nature, but that is an unknown.

That would be a typical example where the facts did not fit the theory. In Einstein's words, the SE company should change the facts (the algorithm)