If you ever stop to think about what it is that has really been revolutionary about the internet, you realise it's the ease with which you can interrelate different sorts of information. In mere seconds you can access a wealth of information on whatever subject you choose, be listening to an musician you've never heard of before or simply find out where your nearest fast food joint is. And while search engines and social networking have clearly played their part in this, the true hero is the humble hyperlink. In a single click on a word, you can be anywhere, and without it the internet… well it would hardly work at all.
However one area that's been held back from taking full advantage of the hyperlink is the humble picture, and this is where ThingLink comes in. Rather than simply having one link associated with an image, it allows you to add multiple links all throughout the image.
Now, the service isn't brand new, and there are other competitors in this space but ThingLink offers its service for free and has just recently launched rich media links. These allow you to embed links to other services such as YouTube, Spotify, Wikipedia, Facebook, and Twitter allowing you to play a video or song, Like something on Facebook, or view a snippet of information all without leaving your page. All it takes is a bit of code to be embedded on your website and you're ready to start tagging away.
With the world still full of paper forms, ancient documents, and people that just prefer to scribble with a pen and paper, there's a market for a service that cheaply and reliably converts hand written documents into digital form. Unfortunately, for all the benefits that computers bring, for certain tasks humans are still just better, and reading human scribble is one of these tasks. Indeed its been suggested that we're still 50 years away from computers being able to effectively perform this task, though we're not foolish enough to completely dismiss the idea that it will happen sooner.
The trouble is, if you're dealing with a mass of potentially sensitive data, you can't just outsource the digitising of these documents to the lowest bidder. So what's the solution? Well, one possibility is to do the work internally, but this is a very expensive way of doing things. You could also have layer upon layer upon layer of security checks on the personal working for the outsourcing company you work with. Or you could take the route presented by MicroTask.
The company has pioneered a new service that allows huge volumes of hand written forms to be interpreted and verified completely independent of any high-level human interaction. The way it works is to break scanned documents up into tiny segments such as single words or dates that on their own lack the context to have any meaning, and thus have no security risk. Once broken down these segments are sent out to people to be interpreted and input as digital data. Data is then verified as correct by a system of comparison whereby multiple people solve the same puzzle and the system compares the answers. If there's no consensus the segment is sent out again to a larger group of people until a more definitive answer can be found.
The company can process individual documents or vast libraries of data, indeed it's currently in the process of digitising the entire Finnish library using a simple game interface called Digitalkoot that's open to the entire Finnish population to take part in. Indeed this combination of gamification and crowdsourcing is set to be big business in the near future, and MicroTask hopes to be riding that very wave. Good luck to them, we say.