IoT is a new area and we all have a lot to learn. Some of us approach IoT like a systems development project, which it kinda is. Others approach IoT like a classic Agile project so they can be competitive, which is kinda what you need to do. But IoT is really its own animal. An IoT platform is a single unique entity that is theoretically immortal, unlike systems such as planes and cars. An IoT platform also has special and often costly deployment considerations, which is just one challenge to using an Agile continuous engineering and delivery approach. At the same time, IoT is rapidly growing in the consumer and business markets which demand modern approaches like Minimum Viable Products and Incremental Deployment.
The talk is IoT Continuous Engineering: Pitfalls and Solutions for Agile and Systems Developers. I’ll be discussing approaches that Agile developers embrace but systems developers often don’t (and vice-versa) to point out some techniques that work (or not) with IoT. And I’ll illustrate IoT continuous engineering issues and potential solutions that are unique to IoT. The talk will extend the work I did on the IBM Garage Method for IoT.
Here’s the abstract:
New IoT developers have a high risk of failure when starting to work on IoT systems. There are unique issues with developing in the IoT that must be addressed or the project will fail technically or commercially. IoT systems are more complex, have higher risk and cost of failure, require physical deployment considerations, and critical architectural concerns must be addressed early. Similarly, Systems Engineers often don’t have the perspective of iterative development, continuous engineering, and requirements flexibility that makes an IoT project successful in an Internet enabled world.
This session is an introduction to successful practices that are unfamiliar to engineers entering IoT development. We’ll illustrate how to successfully deliver IoT systems by:
- Continuously and iteratively deploying products that include multi-platform hardware and software components
- Understanding the high cost of IoT development failures as compared to other types of software development
- Making post-deployment cognitive analytics part of the development lifecycle
- Addressing risks early to increase the likelihood of successful delivery
- Illustrating the dangers of ignoring IoT architectural issues
- Addressing critical IoT architectural issues without committing “analysis paralysis”
- Avoiding unhelpful Agile and systems development practices and perspectives
- Embracing Agile and systems development practices and perspectives that are helpful for IoT product development
If you plan to be at ATO in October, please come by and join the discussion!