Search: Do The Google Dance

Kuro5hin has an interesting article about what the “Google Dance” is and how it affects your ranking on the worlds most popular search engine.

Long story short: Dance equals Data. Servers. Moving. New Results.

A more complete answer is that the “Google Dance” is the nickname that has been given to the time of the month (usually around the 28th) that the data that “GoogleBot” (Google‘s little spider/robot that goes round ‘reading’ the web) is introduced into the system. However, since Google has over 10,000 servers it does take some time for the data to propagate around (“propagate” has now become my favourite and most used word for some reason). It has been long known that the start of the “Dance” can be found be watching when the data on the www2 and www3 starts ‘reading differently’ than that on the main Google server (an illustration in the Kuro5hin article is to do a query for links to Yahoo!).

One thing that has always bothered me about this is – obviously the main www server at Google points at multiple machines and it is extremely likely it is the same situation for www2+3, so why has Google separated these blocks? Surely it’ll make more sense to allow zero public access to the 2+3 data centres and just let the data spread across the servers without the public knowledge. Why do “we” need to know when a dance starts?

Well, I asked the question, now I’ll answer it. A good reason for noticing when a dance starts is so that search engine optimisers/optimizers and search engine placement specialists can look at the new Google search results as soon as they become available and see if any changes in the Google PageRank algorithm have taken place. If site X drops in the rankings but site Y rises, find out why and fix it before the next visit of Googlebot. I personally know of a number of changes that could really reck havoc across the SEO field when Google puts them in place (yes, I did say “when” and not “if” – I’m aware that Google knows about these “tricks of the SEO trade” and are just finalising how to work around them- it’ll probably be a couple more months before the code is in place though). Google employees are quite rightly proud of the technology behind the ‘Big G’ and try not to mention what takes place or is planned, but sometimes you’ve just got to pay attention to what isn’t being said rather than what is.

Google is, at the time of writing, just coming to the end of one of its dances and I’ve already noticed a few minor changes. First of all, Google’s image search doesn’t seem as comprehensive as it was, they seem to have penalised a few more irrelevant sites (i.e. sites that included keywords some where on the page that had no relation to the content) and they also seem to have added a little extra ‘weighting’ to sites from certain sources. It doesn’t look like a major change, just a few minor tweaks.

Oh – the good news is is that the Google “FreshBot” seems to really like my blog. The “FreshBot” is the name for the Google crawler/spider/robot that visits sites more often than the usual “once a month”. FreshBot is updating Google with the contents of my blog at least once a week – sometimes as often as every other day! Go me!

There are also a couple of interesting blogs about Google that you may like to read: Google Village and Google Weblog. You may also find Ovidiu Predescu’s weblog worth keeping an eye on as he’s recently started working for ‘G’.