A Study on PageRank Values of Paid-Result Search Engines
Posted 29 November 2002 - 11:39 AM
Posted 29 November 2002 - 12:19 PM
The PR differences between the engines with and without relationships with Google is slightly interesting, but it doesn't bother me if they favour a few sites. They don't do it across the board so it doesn't matter. The article might have been more credible if the writer had been able to state things like, Yahoo! only aquired PR10 after they entered into a business relationship with Google, but it didn't say one way or the other. It just isn't a piece of proper research. It's a long-winded piece of personal opinion and moans.
The writer will need to do a lot better if s/he wants to be taken seriously. I hope it isn't you
Posted 29 November 2002 - 12:21 PM
To pick just one of many obvious failing and faults, the document is entirely incorrect about Overture's PR value.
www.Goto.com is still PR8 if you can catch it before the redirect due to the high number and value of backlinks. Overture.com has PR of 8 for the root domain that most link to, but unfortunately redirects the browser to http://www.overture.com/d/home/ very swiftly, for which there are far, far fewer direct links. It seems that our friend's 'scientific study' looked at the value of the redirected page, but neglected to mention this in his 'study'.
In real scientific studies, just one single flaw can negate the entire study. This document contains scores of flaws that are obvious even at a glance.
I conclude that the value of this document is as a waste-paper bin liner.
Posted 29 November 2002 - 09:13 PM
I learnt that the author, Dr Garcia "used to work for a well known and reputable SEO company" so was "well awarre Overture was formerly known as GoTo. We understand page redirections and ip redirection as well".
Dr Garcia told me: "You miss the point that this is not about Overture, SearchKing or even Google itself but about an imperfect metrics an industry pretends to follow."
I'm glad he told me that, because from his 'study' I gathered that this was about PageRank. He even named PageRank within the title of the study. Still, far be it from me to suggest he's trying to change what it's about after being caught in inaccuracies.
Anyway, the reason I mention the email is that apparently mine is not the only one received, and a new footnote about the 'feedback' they have received now follows.
Even assuming Overture's index page has indeed a PageRank of 8 units, why give such a high value to a document with no content at all? Who in his right mind will cast a vote for a blank document or for "John Doe"?
It seems that Dr Garcia is unfamiliar with the concept of linking to root domains, or the fact that some suggest this is the best form of link following deep-link legal cases. Presumably Dr Garcia thinks that all links to the Amazon.com domain are just telling us about the homepage, and not intending us to dig deeper into the site. :roll:
Overture's redirectional page with no content is not that different from a "blind" doorway or "blind" splash page. If an average site has the same redirectional mechanism in its index page sure it will get in trouble with Google, AltaVista or other search engines. Why then not apply the same rule to other search engine index pages?
Dr Garcia apparently is unaware that Google not only support genuine redirection needs but indeed provide information on their site for how to do it properly and well. He also seems to have little or no understanding of PageRank, the fact that PageRank is entirely separate from any measure of on-page content or ranking criteria, or the basic concept of Link Popularity in general. Link counting counts links to a document instead of paying any attention to what is actually the content of the document.
If PageRank cannot discriminate against index pages with no content, but give "high citation importance" to the link, why then use that metric?
It seems that the poor doctor is entirely out of his depth when it comes to search technologies. It seems that Garcia believes that PR takes precedence over the algorithm, rather than being just one of many criteria taken into account to produce the final SERPs. Unfortunately, Garcia has obviously not encountered any of the thousands of instances where low-PR pages outrank higher PR-value pages for searches because the content of the document takes precedence.
There are certainly no shortage of poorly researched pseudo-scientific articles full of SEO misinformation around to keep Garcia's paper from being lonely.
Posted 30 November 2002 - 11:27 PM
Sadly he has declined (so far) and replied as follows:
Thank you sir,
We haven't visited the site you refers to or what visitors may have or are discussing with regard our research study. If something good come out of their discussions or if someone is willing to address the issues we have raised, then it was worth the effort. From our part, we have raised a serious question with regard to high PageRankings supposely assigned to Overture and to a blank index page with no content (a "blind" splash page) and that redirects. If PageRank count votes to a blank page with no content at all, then we have a serious flaw in the PageRank metric. On the other hand, if an average user has a blank index page which redirects, sure he/she will be penalized by Google, AltaVista and other search engines. If search engine use redirections in index pages with no content at all, why they are not penalized, too. Isn't this a double standard? The entire discussion on this particular matter is now in our FEEDBACK FROM READERS section.
Hope while discussing such important issues, people can see thru, without getting into personalistic attacks. As mentioned before, this is not about Google, Overture or any search engine in particular, but about an imperfect metric an industry is trying to follow.
With all due respect to you
Dr. E. Garcia
I have responded to Dr Garcia directly, and again have invited him here to rebut or debate my criticisms directly.
The ongoing points that I continue to press are as follows, extract taken verbatim from my last email to Dr Garcia:
As for your raising "a serious question with regard to high PageRankings supposedly assigned to Overture" I believe that it was readers such as myself who pointed out the higher PR score of Overture. Your initial focus was on how Google favoured smaller PPC companies above Overture, which is entirely the opposite message. If you are to gain anything from feedback, please stick to facts.
Your study's observations on high PR for splash pages, which you now equate to "blank pages" have likewise been added after the initial criticisms, and so cannot be used to sidestep the fact that those criticism were valid. If the criticisms were not valid, why have you found it necessary to update, and why are you using the arguments only made in that update to defend your article?
You seem to be taking criticism of the study, which surely deserved all criticisms made, as very personal ones. Your initial reply implied name-calling, which was certainly not a part of my initial feedback. I said that you didn't seem to know your field very well, and I stand by that. However that is not name-calling.
It is perfectly natural that you will be judged upon the work you produce. I feel that you are unscientific, inexperienced in SEO matters, and judging from your replies, are willing to twist facts to suit whatever agenda you have at the time. If this is not the case, then perhaps you need to spend a little more time on your work and communications.
Your study is unscientific, because your figures were wrong, and because you lack the ability to rectify the figures. If you want to see the PR of Overtures home page, simply change your browsers cookie handling so that it will prompt you to either accept or refuse the cookie Overture will set. When you arrive at the overture.com home page, you will clearly see the PR value before you choose to accept or refuse the cookie, because the redirection is partially dependant on information that may be in the cookie on your machine. Simple when you know how, isn't it?
I find you inexperienced in SEO matters because your study and your replies both imply that PageRank is the sole or primary criteria in ranking pages. It is not. No matter what search terms you study, you'll see thousands of occassions where lower PR pages are placed above higher PR pages.
PageRank is a metric about link-popularity - neither more nor less. Anyone with experience in SEO should know that, and should know that the keywords in the title of a document are more important than your PR score.
PR scores alone have little or no bearing in where a site will rank in search results. Keywords matter far more, otherwise Yahoo (PR10) would come up even for searches it had no relation to. Go ahead, search Google for the term 'keywords' and see if a Yahoo directory category (or a Google one for that matter) come up in the top ten results.
You truly come across as having no understanding of what PageRank is designed to do, what it is a metric of, or how it is used by Google. PageRank is designed to count links and provide a guide to the value of pages solely from a 'popularity vote' perspective. It counts all links, even to pages it cannot reach, because *that* is what PageRank does that no common webcrawler can do.
You argue about PageRank being an 'industry metric' despite the fact that PageRank is actually a trademark of Google, and therefore cannot legally be used by others at all. In fact, just to download and use the toolbar you had to accept a Term of Service that instructed you that you may NOT use PageRank data for your own purposes.
In short, your arguments are poorly made, full of holes, lack insight, and the best points in your document even from your own admission have all come about through reacting to criticisms and feedback.
You may notice from the tone of my reply to Garcia that I rather dislike the way he seems to be seeking to change the study after the fact and then try to use changes made as a result of criticism to refute the criticisms that inspired the changes. We all make honest mistakes, and it is generally better to be honest in admitting them.
If he'd just said "Oops, yes, I made a mistake" there'd no longer be a debate that could embarrass him.
Reply to this topic
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users