Who knows how big the data is going to get
They say that the Internet of Things has a protocol problem. But the communications bottlenecks are nothing compared to when humans conspire to cock up
You can understand why the combination of supercomputers, artificial intelligence, robots and the Internet of Things make luddites nervous. If I’m being honest, my initial reactions to each new development in M2M aren’t much different. I tend to go through a combination of shock, denial and anger, after which I move to the phase of acceptance and then, finally, a period of mourning for my lost innocence.
Try as we might suppress it, there is a luddite tendency lurking within all of us. The crucial aspect is how we deal with it. Personally, I try to comfort myself with the knowledge that nobody has ever created any sort of IT system that comes close to sophistication of the human brain. The organic thought processing and communications systems of even the dimmest reality TV contestant would overshadow anything that Google will ever come up with. Hopefully.
Having said that, humans don’t always seem to combine their intellects to achieve the maximum collective good. To test this theory, I recently rode my bike head long into a barrier in a park at night, inflicting spinal injuries on myself. Though scientists are supposed to have no pre-conceived ideas about outcomes, I was a little disappointed at the way the experiment panned out. I discovered that Britain’s National Healthy Service is run by people with immense knowledge and intellect, such as ambulance paramedics, casualty nurses, orthopaedic consultants, pharmacists, neurologists, general practitioners, radiologists, GPs and physiotherapists (to quote a small sample of the people I’ve studied at lose quarters recently). But they all act in isolation.
While I yield to nobody in my admiration for their massive intellectual capacity and hard work, you can’t help noticing that they seem to operate as individual silos. Like the old fashioned telecoms systems we often criticise, these human silos aren’t great at communicating with each other, or even passing useful information to the patient.
You start to wonder if machine to machine communications could be designed to fine tune some of the processes of the NHS. At a recent Innovate show in London, for example, start up Acticheck was demonstrating a new wearable Smartband which can detect both lack of movement and falling body temperature in its wearer and make the decision to alert the emergency services. It also has the advantage of human in that its GPS capacity can pinpoint exactly where it is. That would have saved me 90 minutes of ambulance waiting time.
Obviously these are early days for Acticheck and M2M business culture. On the other hand, but what better time is there to stop the rot setting in and enforce the right disciplines for a culture of communication.
Many M2M systems are a closed shop, with devices from one manufacturer being concentrated in one house or building. But when communications widen, as in the Internet of Things, then we risk falling foul of the intransigence of different vendors with competing standards. If only that could be nipped in the bud. Currently a list of protocols for machine conversations sound less like a communications stack than a Tower of Babel, featuring IBM’s MQTT, OMG’s AMQP-based DDS, RESTful HTTP, XMPP, CoAP, NanoIP and SSI. I’ve seen hospital corridors with less confusing signposting.
Then there’s the actual data. In the world of M2M, traditional relational databases will give way to new, distributed types of silos. The arguments over whether NewSQL is better than NoSQL will rage for years. It will be a long time before the dust settles and clear winners emerge in the dispute between the likes of MongoDB, Cassandra and NuoDB.
According to 451 Research there are tens of thousands of developers listing NoSQL on their LinkedIn profiles, but the market needs hundreds of thousands. Old fashioned developers of relational databases, however, will be able to use their traditional SQL skills in developing for systems like NuoDb.
So at least there is some sort of evolutionary continuity going on. I’m glad someone is using their brain. Mine, thanks for asking, seems to be OK too, according to the CAT scans.