In my last blog post I discussed one of Google’s new products, the News Timeline. Another new product recently unveiled is “similar images”. This uses some very clever technology and a marriage between their search algo’s and the ability to scan an image to bring you further images closely related to your original image search.

An example on the site shows a search for “Paris” and lists results that include the Eiffel tower, a picture of Notre Dame, a city map of Paris and a snapshot of Ms. Hilton. Clicking similar images brings up further images of the same, many times showing the whole series in a release of shots. The algorithm is very clever in determining what falls into the same genre.

It is easy to pass a lot of time with this application, casually browsing galleries and slowly being taken adrift. Sometimes the algorithm serves up the downright obscure and one wonders what exactly the connection is – but then it just opens new paths to more unexplored stuff.

I have always been an early adopter of Google labs technology. I was the first in our company to stumble across Google trends and much later discovered what they had done on the Insights front. Not many people want to join me on Latitude, but then again I have not seen much benefit apart from colleagues on holiday wanting to boast about where they are in the world. Google’s latest unveilings haven’t yet convinced me of their value, but the stuff is experimental and should / could lead to better things. The first is the “Google News timeline”. Now, when I first heard about this I thought Great I can finally sort the SERPs according to most recently published, which would go a long way toward helping me wade through the reams of mostly outdated information out there on the web. I most recently came across this problem when I was researching my recent presentation “to-ezine-or-not-to-ezine”, which is a debate on the SEO value of publishing articles in Ezines purely for the value of their links. Anyone searching the web for pertinent information on this topic is bound to find dozens of citations from rather well know SEO personalities sanctioning this practice and giving it the thumbs-up – yeah, back in 2006 when this practice did indeed work. Many of these posts live on today and have grown in strength and ranking and actually taint the picture quite substantially. Because the community has moved on, there is not much talk about the shortfalls of this practise, but rather an overwhelming majority about the next best thing.