I recently performed a bit of feature testing with a new feature Opera have incorporated into the development builds of their browser. The feature is called ‘Opera Turbo’ and as the name suggests, it aims to speed up the web browsing experience.

So, how does it perform this small miracle? Opera Turbo essentially sits between the customer ISP and the target web server. Instead of retrieving content directly from the web server in question it is retrieved via the Opera Turbo servers which perform optimisation on the data-set. Optimisation reduces the site footprint via compression and thus the ISP overhead. Optionally, Opera Turbo can also pre-render pages at the server side, reducing the processing needs of the client device.

In my testing ‘Turbo’ mode excelled at compressing inefficiently compressed web imagery. This is more prevalent than you would think, what with the sheer amount of blog and user contributed content on the web today; most of it from users blessed with blistering Internet connections. This does come at a cost though and images quality is degreded significantly. Obviously this is hard to convey to you whilst not falling foul of the same mistakes LifeHacker did, but the following screenshots (note: not original imagery) give you a fair impression of the effect albiet not 100% faithfully.

Before - 170kb

After - 6.6kb (96.14% smaller!)

The caveat of this potentially interesting technology is of course that you’ll never be completely in control of it. It’s closed source and operates on Opera owned server parks. Of course we’d never be as libelous to suggest Opera would in any way tamper or eavesdrop on your web traffic, but this does put you somewhat at the mercy of yet another layer of machinery standing between you and your content.

I note in the full paper that calculating the additional latency of taking a hop via Opera Turbo servers would make an interesting study, and this is something especially pertinent depending on where you want to use the software. As an English resident I was routed via a Swedish server for example. This does make you wonder just how many resources Opera are dedicating to the service and just where their priorities lie as a profit making organisation. If I was online in an economically infertile Africa would I too have to traverse Europe to make my connection? It’s an awful shame that the framework is not yet Open-Sourced as there’s surely some great potential in trialling it at an ISP level in areas such as sub-saharan Africa where backhaul connections are playing catch-up.

Recently I contacted Senior Editor Adam Pash over at Lifehacker bringing this technical issue to his attention but unfortunately did not receive a reply. It might be unfair to do so, but you could draw out of this that such technical issues really aren’t prioritised in most organisations as long as the target audience feels the commodity is functionally operational. It’s just a shame that lessons are not learned, across the Gawker Media Network.

If this has interested you and you want to get the full details of exactly what testing I performed, you can download the full paper here