Is Google juice finite?

Posted by multippt

There is an observed trend that the overall Google PageRank decreases over time. Looking at some older models of Google PageRank, this trend could be explained by the growing number of sites, and the spreading of PageRank.

Take the case of most sites for example. As PageRank is spread through links (excluding nofollow links), one can guess that the amount of value each link subsequently decreases with more and more links. So, more isn’t always merrier, if you plan to keep (or increase) PageRank.

More »

Google: Lifting your pagerank penalty

Posted by multippt

If you are one of the few whose pagerank got hit by the recent update, it may have been because of a penalty.

Google (actually, Matt Cutts) suggested several steps to removing the penalty:
1) Read the Webmaster Guidelines
2) If you have any sort of paid links on your site, remove them (or if you cannot, you can reason it out)
3) Submit re-inclusion/re-consideration request
*Of course, there are some who do not have such links, but still receive a penalty for it. Might be because of some elements on the page that look something like that (eventhough it isn’t).

Of course, it isn’t just all that: You need to ensure the problem does not come again.

This way suggested by Google is a great start for tackling the problem of people trying to profit from PageRank, which Google does not want.

The Google alogarithm cycle

Posted by multippt

For about a few days, most pagerank checking tools were unable to display PageRank results (although the Google toolbar does so without problems).

Apparently, it appears some folks does not seem affected by the problem (notably DigPagerank, which now works again through some miracle, as well as the tool Andy Beard has). This came in light when it was discovered Google changed the algorithm for its PageRank checker (by testing the new and old versions of the toolbars). Of course, there are rumors that it might have already been found.

It is not sure whether the change is Google’s plans, or probably just another rumor. However, the change in alogarithm would mean that we would get a nice PageRank update a little earlier than expected.

Deconstructing DigPagerank

Posted by multippt

DigPagerank, a “new” pagerank checker to the scene, is actually unlike other pagerank checkers. Upon peeking its innards (of course, doing a little bit more than just “View Source”), I noticed a few things that actually make it stand out from other (older) pagerank checkers.
*Newer checkers seem to be of the same level as digpagerank, using both server and client side.

How DigPagerank actually works
Digpagerank first compiles a list of datacenters (which are hardwired into Digpagerank), and then uses a little AJAX [i.e. some sweet pieces of JavaScript] to call out to Google datacenters as if it were the Firefox Google toolbar. Of course, there is more behind-the-scenes work, like making a nice code that is specific for certain links. That’s why clicking on the “Stop” button won’t stop Digpagerank from doing its work. Of course, due to AJAX’s built in security, one cannot access external servers like Google Datacenters directly. How does it solve this problem? Well, it actually uses a special kind of proxy, known as a “transparent” proxy. This does not hide your IP address (i.e. the thing that identifies a computer), so it actually queries Google datacenters using your IP address (or an IP address that is close to it).

What about it?
Because of this, Digpagerank cannot be banned by Google via IP (because it means banning legitimate users which it does not want to do; unless it changes the alogarithm), and it complies with the Google TOS because it uses the exact same interface Google gives to its users (and sends the exact same information the Google toolbar sends). Then again, you would not feel very secure knowing that Digpagerank is using your IP address to query Google Datacenters 30 times at once…

Compared to other PageRank checkers, they usually use a proxy to access Google Datacenters. Of course, it is a matter of debate whether it actually is against the Google TOS (it is in the grey area). Of course, no one cares about PageRank anyway, or do they?

What’s with Pagerank checkers?
Ever since the Google alogarithm was found (sorry, you’ll have to find it on your own ;)), many Pagerank checkers were made with the intent to allow users who do not have access to the Google bar to view PageRank.

What does Digpagerank say about this?
Lie low. ;)

What does Google say about this?
Google may or may not bother about this. As long as it complies with the TOS, it is fine.

Hey where’s the real details?
I’m not releasing them because:
1) They are not under the GNU or Creative Commons license

Google should make PR an API

Posted by multippt

Ever wondered why Google decided to make OpenSocial open, but not its famous green bar? PageRank has often been regarded as one of the most useful tools in determining the quality of a site, though not neccessarily the most accurate. As with any ranking system, there are bound to be problems like selling of links, which Google countered during the last PageRank update. PageRank may be regarded as “dead” and “inaccurate” by some, but it is one of a few ranking systems that are available.

PageRank should be made an API. After all, Google is stabbing its own feet limiting PageRank to only those who visit the Google directory and have the Google toolbar (which is not supported by some web browsers). PageRank checkers are in the grey area in the TOS, and cannot be regarded as following or breaking the Terms of Service.

Then again, maybe Yahoo should make something like YahooRank and replace Google’s PageRank. That should give more options.

Google OpenSocial

Posted by multippt

Great! Yet another API added to Google’s arsenal. Google OpenSocial allows you to integrate social features on your site.

Of course, as with any Google API, using it is not that simple. Be prepared to read the lengthy documentation. Nonetheless, it is an opportunity to get your site to interact with sites like Orkut, and Friendster (and many more).

This API allows you to access some nice information, like profile information, friends’ information, as well as their activities on those social sites.

The difference between Facebook? This Google API requires only HTML and JavaScript (or if you are creative enough, Flash), so most sites can use it without changing much stuff.

Say goodbye to PageRank… forever…

Posted by multippt

A source comfirms that this is not true. However, it does give insight to how Google may actually improve on PageRank.

In a bid to prevent people from selling links because of high PageRank, Google decides to abolish the PageRank ranking system altogether. Here are some new measures planned by Google:

1) PageRank reset

All sites will have their PageRank resetted to 0. After that, PageRank will be replaced with a blue bar called VisitorRank (VR). VisitorRank is similar to PageRank, except it does not rely on links alone. VR is incredibly difficult (if at all) to cheat provided if there are some additional measures. VR can measure traffic activity (another worry for privacy? Ask Alexa), ensuring you have an accurate vote. Visitors get to vote for the site, similar to StumbleUpon.

2) No backlink information

Google has plans to abolish the “link:” feature in its search engine. This may prevent people from showing that their links have plenty of backlinks (but then again, Yahoo does have the Site Explorer feature).

3) “Do-follow”

Google implemented the “no-follow” feature. However, few were using it. Thus Google decided to do the reverse, ensuring you get “genuine” links. All links without the “do-follow” relevance attribute will not be counted.

These changes are estimated to be out by mid-November.

Via GoogleMill