I'm Keyvan Nayyeri, a 25 years old Ph.D. student at
the Computer Science department of
the University of Texas at San Antonio.
I'm also
a Software Architect and Developer and previously held a B.Sc.
degree in Applied Mathematics.
This is my blog where I publish content about various topics specifically Programming Languages and Compilers, Software
Engineering and Programming.
Google PageRank has been a an important factor for many web visitors to recognize popular websites on the web specifically in the search results. This rank is calculated based on a set of different complex formulas that have undergone several major and minor modifications in the past years. This has been an ongoing progress for PageRank formula in order to keep it from deviation, and improving its quality as a recognition factor.
Google has been trying to emphasize on the content, quality, and the real popularity in calculation of the PageRank, and has been avoiding some fake hacks to cheat the formula or at least these have been their primary goals when defining and applying the formula.
The most challenging point against this formula is the activities of some webmasters and spammers to improve their ranks without providing the valuable content to absorb popularity. Back in 2003-2005 era the most common method of spamming targeted this area to quickly improve the PageRank. From this point, Google applied many changes in the formula to stop these spammers and webmasters from wasting their time on doing anything other than providing the valuable content.
It’s Google’s tradition to recalculate ranks on a regular basis with new modifications, and with the most recent data. The last update happened in the latest days of December 2008 when they applied significant changes in the formula that affected almost all the websites.
In the past few days after the update I’ve been monitoring the new ranks for some sites as well as the type of the site and the way that it receives its traffic. My main reason to keep an eye on such changes is the direct effect that it may have on online spamming and the way that I should manipulate my spam rules on Waegis.
In essence, I should say that the recent update in the PageRank was the worst one that I’ve ever seen, and it has had such a bad effect on the web. Not only it cannot represent the quality for web pages anymore but also it has opened new doors for spammers to easily promote their sites.
One of the early things that I tried to evaluate was the capability of the factor to recognize the level of content on a site and/or webpage. We all know that Google has been officially moving from site-wide ranks to per-page ranks, so it could make sense to reduce the influence of homepage ranks in the new updates. But after all, the story was different because I still have not been able to realize that how this PageRank relates to the quality and/or popularity of the content.
Here is a simple story that makes you laugh: my blog rank has dropped to 4 while my Twitter page has a 5 PageRank. Interestingly, Scott Hanselman and Phil Haack lost their 6 PageRank to get a 5, and their Twitter pages have the same rank as their blogs. More interestingly, my Twitter page has the same rank as Scott Hanselman’s blog, and his blog has lower rank than his Twitter page! What the mess!
Obviously, these ranks cannot represent anything about quality or popularity because we all know what’s going on and we all know about the quality and popularity of the sites! Of course, I checked over 50-60 sites and blogs in different categories and with different levels and I could be sure that this new rank is not a good sign for anything. There was a blog with less than 2-3 posts per months almost without any external reference and without a popularity, and it had the exact same 4 rank as my blog or Simone’s. He may have used spamming or other techniques to get the rank and this is a possible claim, but isn’t Google supposed to use formula to stop this hell?
Generally, the recent update in the PageRank has reduced higher ranks (5 or higher) and increased lower ranks (4 or lower), so many of the lower sites without a good quality have the same rank as higher sites with a good content. Moreover, I could extract that this new formula strongly depends on the number of external links and doesn’t care about the type of link and other parameters that may be pretended. In my opinion the way that new formula works is very similar to how it was working in the fist days.
The other aspect is the influence of this change on the web and the way that it has damaged our web by encouraging spammers. Unfortunately, my Waegis statistics after the update show that the spammers are smoothly transiting from their older style to a newer style that emphasizes on the holes in this new formula, and somehow they use their spamming style that was common in 2004-2005.
As you may have noticed, in the latest days of 2008 some companies had started public programs to give out prizes to web users who go to sites and leave comments with some keywords. I may write a separate post about this dirty business, but it’s good to know that this last update in the PageRank formula helped these companies very well, and they could achieve their goals faster!
Honestly, I never been positive with the topic of SEO and page rank stuff and I’m a big advocate of quality and content. Unfortunately, Google PageRank has proven to be a bad tool rather than a good one because it’s been a source of dirty business. I mean that its disadvantages has outweighed its advantages after all. From a logical point of view, Google should put its emphasize on the quality and avoid opening holes for anyone to use any technique to cross the longer path of content providing! I don’t know what’s the point in the existence of a business around SEO but it looks like that this industry is just burgeoning!
As the final conclusion, if you like me think that Microsoft is the only software elephant that does stupid things, you’re absolutely wrong, because Google is worse! They should promote their Ph.Ds for this genius formula!
Craig
Jan 11, 2009 3:34 PM
#
I don't think PageRank has much influence on search results any more. Google are trying to deemphasize PageRank's value and produce search results based on better criteria.
Keyvan Nayyeri
Jan 11, 2009 3:37 PM
#
@Craig
I don't think so. If they do this, then PageRank will be worthless and they should drop that.
Craig
Jan 11, 2009 3:46 PM
#
I just googled 'PageRank Worthless' and this is the first thing that came up.
www.google.com.au/.../search
www.startupnation.com/.../1
Withing the SEO community PageRank has just been one small factor of many that influences how your site will rank within Google searches. It would probably not even considered the major factor.
Keyvan Nayyeri
Jan 11, 2009 3:54 PM
#
@Craig
Yes, there are many community reactions like that (as mine is somehow covering this) but there is no official annoucement to confirm the vaue of PageRank by Google.
Of course, I think that PageRank is still a major factor in the order of search results. If they put it out then PageRank is nothing and this is not something that a company likes for its trade mark.
arne
Jan 12, 2009 9:05 AM
#
I have read you article about CustomUsernameAndPassword validation and the article about getting the Client ip address. This is also how we do it. The problem is that you can not combine the two. You can not get the clients IP address when using custom validation cause the statement OperationContext context = OperationContext.Current; will always yield null as a result (at least to my knowledge).
Do you have a way to get the clients IP address in the overridden validate method?
Arne
Keyvan Nayyeri
Jan 12, 2009 9:08 AM
#
@arne
I don't think that I have disabled commeting for the original post. I had to delete your comment rather than leaving this reply.
arne
Jan 12, 2009 9:54 AM
#
Sorry, I have several of your comments and posted it on the wrong one. I will post the comment on Custom Username and Password Authentication in WCF 3.5 comment.
Arne
Dew Drop - January 12, 2009 | Alvin Ashcraft's Morning Dew
Jan 12, 2009 11:24 AM
#
Pingback from Dew Drop - January 12, 2009 | Alvin Ashcraft's Morning Dew
Dave Ward
Jan 12, 2009 1:04 PM
#
Every distinct URI has always had a distinct PageRank. Often, multiple URIs for the same document would even have different PageRanks themselves, depending on which were linked to more popularly (assuming one didn't 301 to the other).
PageRank is internally calculated continuously, and they tweak the formula daily. The updates you see in the toolbar are just display values that reflect a snapshot at a given time when the toolbar updates are pushed out. Internally, every URI has a floating point PageRank of pretty high precision.
You also have to take into account that the display values are manually penalized for some URIs. For example, I got bumped from 5 to 4 just for requesting links to a CodePlex project as part of a contest. That doesn't affect actual search results rankings though.
Leave a Comment