get's a PageRank of 8... even though this page does not exist. Does this seem a bit strange to anyone?

It seems the PageRank value is determined by the PageRank of the front page of the site minus 1 point for every subdirectory in the URL. It can't be right. It appears the toolbar is malfunctioning... either on the client's end or on Google's end.

Does anyone have any actual facts out there as to why this is happening?

I would really like to know why that is happening.

>>I would really like to know why that is happening.

Read the newsletter again, especially this sentence:

"In essence, Google interprets a link from page A to page B as a vote, by page A, for page B."

A page doesn't get link popularity by existing. It gets link popularity by having links to it. And you can vote for a dead page, just like in some states you can vote for a dead man.

I reread your post. And if there's some reason that you are sure this isn't what's happening, you didn't mention it.

And I've seen some pages the removal of which would have been a significant increase in the sum of human knowledge....so I suppose all of the world's 404-not-found pages should have ranked above them in truly relevant search engine results.

I don't want PageRank explained to me. I studied PageRank and other relevancy determining technologies in my thesis at university. I know what it's about.

I'm was just hoping someone out there had some 'facts' as to why the google toolbar appears to be malfunctioning. I've tried to email google and have recieved no response.

Here's the problem explained again. I have found that the toolbar is reporting PageRanks for pages that don't exist. Here's the condition: if google has part/all of a web site in their index, eg. http://www.yahoo.com, pages that don't exist within the site AT ALL, never have and never will, somehow get a PageRank.

The example i used was:
http://www.yahoo.com/asdf/asdf.html

to make my point, try:
http://www.yahoo.com/afdsjsddsafdsafsd/adsfdsafdsf.html

that page doesn't exist and there's no chance anything links to that page... however, it magically has a PageRank of 8 according to the toolbar.

Can anybody help me... I'm asking this of experts and people who actually have facts... not of people who just get on here and waffle.

Sorry, but I'm just a bit frustrated with this forum. Where are the experts?

Regards,
Adam W

For one thing, the original definition of PR by Larry Page, in his 1998 paper, had a value that was significant to four digits.

Then they give us this single-digit bogus toy on the toolbar. Originally the little yellow box that activates after a slight delay on mouseover showed the digit. Then they took this off. Clearly the 0 to 10 scale is logarithmic at some level, which makes the resolution even more dicey.

The purpose of this bogus toy on the toolbar is to justify the fact that Google wants to collect link data, using their 37-year cookie with the unique ID in it. They're building the world's best marketing database.

They'd have no reason to make the toolbar "phone home" with every page you surf without this bogus toy on the toolbar. Essentially, they're using the good name of PageRank to justify their spyware. They had to spell this out in their privacy policy because Alexa lost a class-action suit for failing to explain this.

You can't just "phone home" for the hell of it anymore; you need a reason because too many privacy advocates are watching. PageRank is Google's reason (or rather, their excuse).

So now all us SEO whizkids are scratching our heads over this little green bar, which tells us nearly nothing that's reliable and useful. You could get gray hairs watching that bar move with each new Google index.

I personally think they've introduced a plus-minus arbitrary factor into the green bar to keep you folks guessing. It's not in their interests to give you an accurate PageRank value.

It _IS_ in their interests to collect data on your surfing behavior.

There's too much speculation in this forum!!! I could rattle off all my ideas, but I want facts.

One more call for the experts... please.

Here's the conclusion of a piece by Monika Henzinger, Google's director of research, that appeared in IEEE Internet Computing, Jan/Feb 2001:

"Conclusion: Research into the hyperlink structure of the Web is just beginning, and more exciting applications (such as those outlined in the sidebar, 'Uses of Hyperlink Analysis in Web Information Retrieval') can be expected in the future.

"When Web search engines started to use hyperlink analysis, Web-positioning companies started to manipulate hyperlinks by creating hyperlinks that try to boost the ranking of their clients' pages, making it harder for Web search engines to return high-quality results. To offset this, Web search engines need to design more sophisticated techniques, and to keep them secret since any disclosure leads to more manipulation attempts."

_______________________

You throw a tantrum because we can't produce the scoop on why the PageRank feature on the toolbar behaves strangely. Read the last sentence in Henzinger's conclusion, over and over.

The bloody algorithm behind the bogus PageRank toolbar toy is a COMPANY TRADE SECRET, dude. Those who know can't say because they work for Google. Those who don't know are forced to speculate.

What would you have us do? Why don't you go ask Google for the information and see how far you get?

Don't give me these conspiracies and hidden agendas and paranoia and whatever. I'm not asking for the inside information on google's technologies and "COMPANY TRADE SECRET"'s ... I just want to know if the toolbar has a bug. And then I'll start my speculating from there.

"What would you have us do?"... post facts or don't post.

My 'tantrum' comes from reading too much bullsh1t in this forum.

None of them publish their algorythms, etc. which are subject to change anyway. Moreover, there are abberrations cropping up all the time. This is hardly an exact science. The main purpose of this forum is to share our experiences from which we seek to deduce sound approaches to SEO.

Reality is the way things are whether we like it or not. Some questions simply have no fixed or "right" answer. Some answers just turn out to be better than others. Live with it.

I was looking for an answer ("does dah toolbah have ah bug in it?") ... but I guess this is a place for the deep and meaningful, for delving into the depths of right and wrong, for discovering the truth and unlocking the SEO in every one of us!

My ARSE, SEO isn't an exact science!! I find it hard to imagine the technicians at google sitting around a cauldron, determining ranks by observing the positions of the stars or some other "COMPANY TRADE SECRET"... sorry.

Perhaps you haven't had much experience with... science...?

I just came looking for an answer to a technical, straight forward question ("does dah toolbah have ah bug in it?").

<"I studied PageRank and other relevancy determining technologies in my thesis at university. I know what it's about.">

Adam W,

When and if you find out, please let us know. You are the one claiming expertise.

Also, why not? It could be that this page rank is given for the consumer to use, and not for celebrities and scientists with a SE technology thesis.

Maybe is just intended for you to think that there is a bug.

Right away, I don't believe there is a bug, but keep on with your question, maybe, just maybe, you can make another thesis.

[This message has been edited by etreus (edited 06-25-2001).]

If it's a bug. Then it's a bug. End of story.

If it isn't... then what are we seeing? The speculation begins. (eg. are we catching a glimpse of what Google uses for it's initial conditions during the first iteration the PageRank algorithm? That being a page initially starts with a PageRank determined by a relationship between the value of the root page of the website and the number of subdirectories between the page and the root page? Makes sense. Initial conditions are required for recursing through bidirectional digraphs over several iterations where unknowns exist... perhaps Google found it more effective to create some sort of initial value of a page that's related to the number of subdirectories it resides rather than a general constant among all unknown pages... blah blah blah... )

I don't want to speculate yet. I just wanted to know if it is a bug.

Regards,
Adam

[toned down per moderator request]

[This message has been edited by defenestrius (edited 06-27-2001).]

I thought you were going to say, "But I did stay at a Holiday Inn last night!"~

Very unlikely it's a bug. I'm sure PageRank is reporting exactly what the top technical engineers at Google have programmed it to reflect. And yes, one could speculate that this may indeed be an initial weighting value inherited or extrapolated from other known sources which have an established page rank within the Google index.

Interesting yes, bug, probably not. Now if this were AltaVista, it might be another story...

[This message has been edited by hazmat (edited 06-25-2001).]

To someone who did three years of philosophy in grad school, a bug is relative to the programmer's intention. If it looks like a bug, it may not be a bug. It it does not look like a bug, it may still be a bug. That's what I learned in philosophy.

What I'm saying is that we don't have enough information about Google's intention with the toolbar to determine whether the word "bug" is even relevant in this discussion.

Reality includes a massive analog component, interwoven in various complicated ways that no one can fully grasp. Get used to it, or you'll burn out in no time.

To hazmat - "because it is within a site it still retains pagerank"... that's wrong. That's not PageRank. Do some reading. (I proved it below).

To all - do this exercise: go to your web browser and type in:
http://www.yourdomain.com/dsflsdklfj/adsfsadf.html

if Google has atleast some of your site in its index, that page should get a pagerank.

Now, create a new page called newpage.html on your website (if possible). This ensures that google doesn't have that exact page in it's index yet. Now type in:
http://www.yourdomain.com/newpage.html

in your browser. What happens? No pagerank.

Conclusion:

1: 404's get a pagerank.

2: Pages that really DO exist, but google doesn't know about, don't get a pagerank.

I guess that throws your 'category' idea out defenestrius. And I guess there goes your "because it is within a site it still retains pagerank" idea hazmat. Read the two conclusions again.

Too many people speak before they think. Why are people so eager to "grab the mic" in this forum???

Thank you to all those who read this topic, realised they didn't have any facts and decided to keep their 'speculation' to themselves.

Regards,
Adam

Still we have open the question of the origin of life, the universe and many others.

Is this a matter of speculation or is it a matter of solid science?

It is neither; it is in the fringe area inbetween.

You have speculated that there is a bug in Googles Page Rank. I have a feeling it is intended to be that 404s get a page rank in google, and so it is. It is probable that in the algorithm it will take into account themes and term vector and not only links in. That is the sophistication and advantage Google could have in comparison to other SEs. This is only speculation though.

404s and the Yahoo result page you showed us, DO exist.

Also believe that what you want is to have the Mic more often, and to make this thread as long as your thesis.

[This message has been edited by etreus (edited 06-26-2001).]