Prognosticating about the shape, size and flavor of The Internet of Things has practically become an industry unto itself. In addition to predictions from the usual suspects like Gartner and IDC, a long list of consultancies, technology firms and independent thought leaders stand ready to predict how big the IoT opportunity is and what companies stand to benefit from its success.
What’s less often talked about is how these larger platform- and ecosystem changes will affect the way that technology gets created. That’s why I was drawn to this post over on Bocsh’s Internet of Things blog. In it, Jim Morrish of Machina Research that talks about how changes that are already underway are folding the silo’d market for products with “machine to machine” (or M2M) capabilities into an even more diverse “Internet of Things” ecosystem.
As Morrish sees it, the market for Internet of Things technology is at a “tipping point” (yes, that phrase again!), as new tools and technologies emerge that allow formerly stove-piped, industry specific applications and standards to begin connecting with each other and sharing information. (Bosch is one of the companies that’s playing in this ‘industrial Internet’ space with its device management and business process management offerings, thus the blog post.)
Rather than talking just to each other, machine-to-machine applications will now be written to what Morrish calls the M2M/IoT Application Platform, a broader ecosystem that abstracts information from a wide range of data sources – including traditional corporate and IT systems, as well as legacy M2M platforms.
The effect of that will be to put power into the hands of application developers, who will have free(er) reign to shape the applications that will define the Internet of Things. Freed from the onerous task of mastering proprietary application logic or stove piped platforms, in other words, application developers will be able to rely on off the shelf development tools, protocols, and features that connect them to a much wider pool of data. That might be corporate and other IT systems internally, or field-deployed “smart” devices. (Think: diagnostic information from the organization’s fleet of electric vehicles, or a smart grid deployment.)
“In the world of the M2M/IoT Application Platform,” he writes, “the application developer is king.”
The broad outlines of what Morrish predicts can already be seen. Vendors such as Thingworx and Xively already offer cloud-based PAAS (platforms as a service) that fast-track “Internet of Things” products. Rather than having to build an “IoT” product from the ground to the cloud (literally), a company like Turbid, which makes storm water management products for industries like mining and construction, can plug into a an existing platform like Xively’s, providing customers with the ability to remotely monitor and manage their hardware, and removing the need for costly on-site visits.
With standard APIs that provide access to features like messaging, data archiving, provisioning and directory services, much of the heavy lifting needed to enable these solutions is already done, leaving what Morrish calls “a seamless and intuitive environment in which IoT application developers can focus on developing application functionality, without worrying about underlying and supporting mechanics.”
That all sounds great. But if you’ve been reading this blog, you know well enough that, when it comes to application development, “all that glitters is not gold.” Specifically: as more and more of the hard work of application development and product creation is abstracted in vendor supplied APIs or reusable code snippets, application development (and product creation) accelerate. That’s good, but often comes with a cost: less attention to important issues such as code quality and data integrity. We need to be careful, less the enlightened developer king turns into a callous and reckless tyrant.