TURNING METADATA INTO INTELLIGENT SEMANTIC DATA

One of the things I want to experiment more with next year is semantic data or as Tim Berners-Lee increasingly calls it linked data. Once data is tagged and coded for semantic meaning, or once applications emerge that 'extract' semantics we are on the verge of something very exciting.

Google has been experimenting with semantic search for a while, i.e search for the query "~what is the capital of France?" and you get a semantic search. Or try "~who is the president of nigeria?"

I recently came across, a new application from a subsidiary of Thomson Reuters called Open Calais. It is free, has an api, and is a semantic application that actually works. It describes itself as follows:

The OpenCalais Web Service automatically creates rich semantic metadata for the content you submit – in well under a second. Using natural language processing (NLP), machine learning and other methods, Calais analyzes your document and finds the entities within it. But, Calais goes well beyond classic entity identification and returns the facts and events hidden within your text as well.

The tags are delivered to you; you can then incorporate them into other applications - for search, news aggregation, blogs, catalogs, you name it.

It looks at unstructured data and pulls out content that is separated into the three classes of content it recognises: named entities, facts, and events, as shown in the diagram below.

Watch the video below (apologies for the cheesiness), and play with their simulator. More information at Open Calais

SOCIAL MEDIA COUNTER

A fun app showing the growth of social media by Gary Hayes.

THE BATTLE FOR SOCIAL SEARCH

Google has launched into social search by concluding deals with both Twitter and Facebook. I'm interested in where this might go.

The division between social media and search will disappear, and we should expect to see more spamming of social media to increase search traffic. Google miscalculated and missed a key evolution of search, after both Facebook and Twitter have won the real time search battle.

Google will remain number 1 in search but the search business is going to become increasingly fragmented. Mobile search is still an open field, and we are only at the dawn of semantic search. Even within social search, there is in my head, a further division, which includes real-time search (which Google is focused on), and geo search. There are many billions of dollars at stake, and the game just got a bit more unpredictable and interesting.

This is all still incredibly experimental but we all now need to master something new (again).

MANAGING YOUR DIGITAL ECOSYSTEM

Here is a piece about digital ecosystems I recently had published in Marketing Week:

Every brand needs to manage its digital ecosystem. Like a jungle ecosystem, it is a wild, complex and sometimes dangerous place. The web is just one part of that digital ecosystem; it is as wide as it is deep and continues to grow. Nobody knows how big the web is and there is no real map.

Your brand is being talked about throughout its nooks and crannies, sometimes positively and sometimes negatively. Whether you are hosting the conversation, observing the conversation, or completely oblivious to its existence, it is affecting brand equity, sales, and loyalty

The greater digital ecosystem is multi-layered. However, unlike in archaeology where today’s living layer sits on yesterday’s fossilised layers, the digital ecosystem’s generations co-exist.

The first layer of the ecosystem is its foundation, and is the network of computers that links us all; we call this the internet. Selecting what hardware you buy and rollout, and deciding how that infrastructure is managed, structured, linked, and maintained are the key questions for this layer.

The second layer links documents together in a web navigated by hyperlinks; we call this the World Wide Web. In web 1.0, we focused on online banners, virals, websites, traditional search engine optimisation, and search engine marketing, as the main tools for driving success. Brands and media owners controlled the conversations here.

The third layer is a graph of people connected by social media platforms, in what Mark Zuckerberg, founder of Facebook calls a social graph. This is web 2.0 territory and the key to success here is to understand how to create contagious ideas that change the conversation and manage communities of interest towards advocacy, whether they be in the blogosphere, on Facebook, Flickr, Youtube, Linkedin or MySpace.

A fourth layer is in its infancy. This is the internet of things, where things like cans, books, shoes or parts of cars are equipped with minuscule identifying devices. This is where we bridge the gulf between the virtual and real world.

The fifth layer is the semantic web. Largely still in the realm of science fiction, it’s a place where all things will have a Uniform Resource Identifier, meaning that they will be identifiable and open to manipulation by machine based intelligent agents.

This rapid evolution points to the idea that this is marketing’s Cambrian Period. In archaeology, the Cambrian Explosion was the period on Earth which saw a rapid increase in the number of complex animals. Marketing’s Cambrian Explosion is the period when we will see a rapid increase in the complexity of the media landscape.

To successfully manage your ecosystem, the smart marketer needs to build an engine with a strong brand idea at its heart. However, this idea is less about an image, and more about an experience centred on branded content and branded utilities that sell by helping or entertaining the audience.

This content and these utilities give the brand permission to engage right across the ecosystem - reaching from areas where people are unaware of the brand, all the way up the purchase funnel to places where your most loyal customers inhabit.

Perhaps most importantly, with the measurability of digital technology, each engagement must be optimised, so the brand works out which parts of the ecosystem are the most profitable places to hunt and which tools contribute most to these profits. Brands now have a choice: create an engine that allows you to glide through the ecosystem; or end up as a fossilised dinosaur. The choice is that stark.

ONLINE VOTING CONTESTS AND MECHANISM DESIGN IN SOCIAL MEDIA

In the UK talent contest - X-Factor - the twins pictured above did spectacularly well despite their lack of talent. Jedward as they became known were voted through week after week by the public even though the panel of experts headed by Simon Cowell implored the public to vote them out. Punk legend Johny Rotten even got involved saying "They're just lazily untalented. It's not fair, it's not right and a perfectly good singer gets voted off while that dopiness gets promoted. How is that singing?”

The Jedward Problem is an example of mechanism design gone wrong. Lots of what we do in social media involves promotions, competitions, games, and other forms of mechanism design. Purists dislike using these techniques. They see them as downmarket, and a form of attention cheat. They say brands should invest in long-term brand building, and application development. They are of course right when you view this issue from within the social hive, but in the broader world of business there is no more complex set of challenges than those in mechanism design. The 2007 Nobel Prize for Economics was awarded to Leonid Hurwicz, Eric Maskin, and Roger Myerson "for having laid the foundations of mechanism design theory". It's complex stuff. Let me explain with the help of Wikipedia.

The distinguishing features of these games are:

• that a game "designer" chooses the game structure rather than inheriting one
• that the designer is interested in the game's outcome

The designer of a mechanism generally hopes either

• to design a mechanism that "implements" a social choice function (e.g. high bid, low bid, no bid, or cheat)
  
• to find the mechanism that maximizes some value criterion (e.g. profit)

Hurwicz, and many others have used complex maths to analyse everything from elections, to financial markets, and this thinking is at the heart of the Google AdWords auction system. Bloomberg quoted Hal Varian, chief economist at Google as saying "We are one of the most active users of mechanistic design,''. In the auctions, advertisers place bids to have their messages displayed on search result pages, but their final positions in the auction are not calculated purely on price, but also on what Google calls a quality score. Improving quality scores is a type of game. Deciding an advertisers position in search results is a mechanism design problem.

These games, incentives, and mechanisms are all over the web. In a previous post I wrote about Keynes' idea of the beauty contest and I have also previously written about what I call message, medium, and modus. For me, modus - the mechanism design you choose for your idea - is frequently more important than the creative proposition in the diffusion of your idea. If you can engineer the mechanism to be social and contagious then you can achieve your goals. But beware of online voting cheats, fake email addresses, and many other devices.

This approach basically means you can use other people's social ties to propagate your ideas. The strength of their social ties, their trust relationships, and their influence becomes the fuel that drives your campaign. Jedward have been voted off X-factor, but now they are free I'm sure they will be willing to share their knowledge of mechanism design.