The world according to algorithms

I wrote this post over three years ago! Gosh – kinda of more scary now. Yikes.

_____________________________________________________________________________________________

My personal, trusted search agent, my husband, cut out an article for me about DemandMedia, an innovator in offering a service for web owners to pull algorithm driven, highly moentizable content – fast and cheap.

Then a few minutes later I read about Cheaptweet.com and how it uses an algorithm to mine Twitter feeds for deals on clothes, electronics and services.

I began to notice a pattern.

The next day I read about new search methods that were smarter because of, you guessed it, algorithmic technology.

Now with a thud, I realized, a bit to my horror, that algorithmic logic drives a big part of our lives. It drives our searches and, as a result, what we learn about. It drives which ads we see and crunches through a formula to present us with the most relevant, contextual based ad possible. It filters what offers we see or don’t see online.  And the ever iterative algorithmic engines can even choose our future mates.

I even think some algorithm predicted the end of the world to happen sometime in 2012 *sigh*.

It then blindingly dawned on me (better late than never) that my perception of the world was being shaped by algorithms – aggregation of data points. I was taken aback by the fact that my world perception was not formed as I thought by my experiences with real people – but by mechanical machines spitting out numerical answers to questions I had not yet asked.

I realize I see the world through number colored lens. I am not sure I like the effect.

This shouldn’t be bothering me – but it does.

Judy Shapiro

http://twitter.com/judyshapiro

Advertisements

A digital fairytale

Once upon a time, there was a king that owned lots and lots of content. Now this king, let’s just call him King M, controlled how the people got his content. They had to pay him a content tax or sit through commercials. Either way, the people all had to pay him and he was very very happy. And very very rich.

But then one day, the Internet threatened his kingdom because people could get content lots of ways and did not have to pay King M his due.

He was very very unhappy.

So to fight the evil Internet and it’s main warrier, Google, King M tried to make all his pages seem invisible to Google. King M hoped that the content would be safe behind his gated sites.

Alas, King M underestimated how determined people would be to get his content. And while this tactic may work for a bit – this strategy can not win the war. Content, like water, will ooze through the cracks on the Internet, and find its way out.

King M would do better to understand that content simply attracts – yes better content attracts better – but that’s all content can do in the digital world – attract audiences. No longer would people will be willing to pay for content – no matter how much King M raged.  The people will pay for interesting, new ways to engage with the content – like in rich community places or new ways to access the content. People will pay for new ways to consume the content.

But King M did not see the wisdom of this tactic. Unfortunately for the people living in King M’s kingdom, they will have to bear the brunt of King M foolish battle strategy. Hopefully, King M will learn his lesson before too much digital blood is shed.

One can only hope.

Judy Shapiro

http://twitter.com/judyshapiro

The centralized search model is over. Welcome to your personal search.

I was having dinner with a dear colleague one night after a long day. He was trying to explain how uTube will become the video search engine in the future and that fact was key to understanding a new way to optimize search in video. “Nothing will come close” he said.

Rather impolitely, I kept saying, actually insisting, “No – they won’t be”. He looked at me rather incredulously and not without good reason. He was well ensconced within bosom of the techno-glitteri and he moved in the high-glam world of high tech. He knew stuff about uTube that most people didn’t. I was not just challenging him; I was challenging most of the accepted wisdom of the techno Silicon Valley world; that in the internet game there will be a few search winners, Google and uTube. Game over. That belief was required for the rest of our conversation to continue.

Poor man. He could not get me to agree to that simple, well understood principle.  I could not buy into a monopolistic search belief system. I sense the centralized search model is ready to fall apart. I was not just being argumentative or combative, but I sensed a “new” trend that has been operative for 20 years was starting to asset itself and only few people seem to see it.

I call the trend, “the techno-edge effect” and the main principle is that new technologies migrate from the enterprise level to the consumer level, to the “edge”, at some point. Some simple examples to illustrate my point:

  • Corporate Audix systems (messaging systems) evolved to become consumer answering services and ultimately devices (1980s)
  • Desktop PCs became personal PCs (1990s)
  • VoIP for enterprise migrated to consumers via services like Vonage (2000s)
  • Centralized software development to crowd sourcing (2007)
  • Podcasting was a corporate activity, now anyone with a webcam can be a broadcaster (2008)
  • Centralized news service to citizen journalism (2008)
  • “Cloud” computing would have only been contemplated for business a mere few years ago – now the model of the new Netbooks is that your data is “in the clouds” (2010…)

(Enough yet? If not email me, I can send you about 2 dozen more J)

The point is that the march to the technology edge is unrelenting and undeniable. The only question is how fast a particular technology will move to the edge. One could argue that Twitter became so successful so fast because it encouraged a high level of customization and personalization – it moved out to the edge really really fast.

Very interesting, but what pray tell, does this have to do with the entrenched and well accepted belief that search will be dominated by a few centralized companies?

Everything! Because I believe technology has reached the point where we will be able to create a totally personal web not through some centralized company, but through the transformative ingredient of trust. Now that people are creating trust for themselves (via communities, blogs etc) the power is shifting relentlessly from centralized search providers, (like a Google or uTube) to distributed power of the Trusted Web.

The model of the Trusted Web and decentralized search

Instead of semantic search or intelligent search agents from big companies driving the web’s evolution, I contend that each person’s ability to drive trust into every web action will be the animating force that moves us from centralized search paradigms to a new, decentralized one. In the new model, we will be able to search better because our trusted communities are doing search for us. We can better trust sites, because people we know had good experiences and we will learn about new things on the internet with services like Twine or HopSurf that gives us ideas based on people who are similar to me.

In the future, we will rely on the power of our networks to inject trust into our search – we won’t go to Google first. If some specific event requires that I get search help from a diverse set of backgrounds, I can create a virtual, new trusted group from all my networks. We will still search internet, but we will start with our trusted network first moving out only as the need requires.

This new proactive model of creating trust is not some future, far off concept. It is happening here and now. We now use trust based content rating systems to determine what content is more trustworthy. Reputation systems allow us to better trust verified SMEs (subject matter experts) versus just any reviewer. The explosive growth of communities demonstrates how people are proactively creating trust through shared interests. Twitter, Comodo HopSurf and Twine are interesting in this decentralized model because they provide an individualized community-based “trusted information filter” to help sort through the deluge of relevant data. Forums are yet another mechanism for people to create individualized trust by letting users share experiences.

All this adds up to the inescapable trend that the techno-edge effect for the Internet is that trust will be dismantling the centralized search monolithic model we have today. We are now moving to this new trust decentralized model. A model I call the Trusted Web.

Watch this space.

Judy Shapiro

http://twitter.com/judyshapiro

%d bloggers like this: