In the past couple of weeks I've attended a couple of IBM events and heard some great, inspiring speakers. In particular I enjoyed the sessions where we were thinking about the future of technology and the Internet of Things. It was the first time I'd ever heard a cow being described as a device, but it's really got me thinking.
How is a cow a device?
If I was a farmer I think it would be great if each of my herd collected data about themselves - their health, milk yields and location. I've already thought for a long time that it would be great if that the chip I had put in my cat over ten years ago was be GPS enabled - if you don't have pets or your pets aren't chipped, the idea of this is that if the pet is lost, a vet can scan the chip, check the details against a database and return them to the correct human.
They have extended some of the capabilities of the chip - I have a cat flap that is programmed to only open to my cat's chip thereby eliminating the risk of feline intruders that might beat her up, eat her food or, god forbid, sleep in her bed. It's great but again I've had it for at least six years - surely there should be more that technology's doing for me and my cat?
Very occasionally she manages to get lost and it's a mission to find her again - she's good at hiding so we've never got as far as her being delivered to the vet by some caring third party, scanned and returned to me. What happens is I print off a couple of hundred posters and put them through doors, in shop windows, walk round and round for hours until someone alerts me to her presence. Wouldn't it be great if my phone could alert me to her presence - the 'Find My Cat' app linked to the GPS capability in her chip? That would save me a lot of hours and worry.
I think that technically we are almost certainly capable of building this technology today, but when we consider this particular commercial application I think is where we run into trouble. How much would I be prepared to pay for this? £50? Definitely? £500? Maybe not. For £50 today I can buy a small GPS unit to go on my cat's collar. But my cat doesn't have a collar. I worry that she'll catch it on something and hurt herself.
How Secure is the Internet of Things?
No doubt it will get there though if Moore's Law has anything to do with. And then some interesting security questions come up. What if someone can hack my cat? If I were someone famous like, say, Paris Hilton, and famous for my love of pets, could paparazzi use a dog to geo-locate a person? Could they kidnap animals for ransom?
More common thinking around the Internet of Things centres on concepts such as the connected home. Research from Fortinet suggests that:
"A majority (61 percent) of all respondents believe that the connected home (a home in which household appliances and home electronics are seamlessly connected to the Internet) is “extremely likely” to become a reality in the next five years."
So what are the barriers to a fridge ordering milk, the heating being switched on from the office or the homeowner receiving alerts when a door or window is opened or a room is flooded?
Potential consumers of these technologies' concerns are around data privacy and protection. If your fridge knows how to order from Waitrose using your bank details, how easy is it for a hacker to also acquire this information? If your house is full of expensive devices - televisions, sound equipment, cameras and so on that are all identifiable over the internet, does this make you a sitting duck for a burglar with some tech skills?
So how do we allay fears and take technology to market? Just as when we considered Mobile (one of the four key innovation drivers today - Cloud, Analytics, Mobile and Social as identified by IBM - is the IoT next?) we identified that DevOps is key to organisations being able to delivery quality software, fast and take information security considerations in hand by properly involving all in the software development process. If companies want to compete and offer compelling, trustworthy solutions in the Internet of Things they will have to do DevOps.
Which should be fine - as according the research, there's a 5 year window until the IoT, or the connected home at least, is matured. What will be the state of DevOps then? In our #DevOpsFriday5 series we've been asking the question: "What’s your prediction for what DevOps will look like in 2020?"
DevOps Dead or Done?
Answers have been varied:
DevOps matures will fold in to the larger "manufacturing process management" family and adopt an even larger set of management tools
Tools and Platforms for DevOps will provide more full system orchestration and self-healing capabilities
It will have evolved in to something else. Probably some sort of loosely-coupled approach that extend well in to other business areas and connect them cross functionally
Further widespread rollout of DevOps culture, behaviours and tooling and the resulting need for greater software delivery efficiency will result in even more organisational dysfunctions being exposed
In 2020 the concept of Continuous Delivery will be the norm for most organizations and by then we will all be on to something different
What do you want to see in the Internet of Things? What do you think DevOps will look like in 2020?