In the previous video, Robin Calhoun, Senior Product Manager, brought up a point about traceability of the future that will offer a heat map to trigger you when changes important to you occur in the product lifecycle. It will let you know what pieces of the process you should be paying attention to and it will flag you if something is not quite following the right pattern.
This time, Derwyn Harris, Director of Product Marketing, wonders whether traceability will allow some sort of artificial intelligence to manage building products for us, if not instead of us, in the future.
Robin thinks computers can take over the repetitive tasks in a developmental cycle, whereas humans will always need to make the final decisions. If an AI highlights an issue with testing or a lack of, a human can provide a reason (for example, it was a national holiday that day) and then qualify what happened (missing testing on that particular day is expected). If there is something a human needs to act on, the system likely will not be able to give a reasoning why they should or shouldn’t proceed, or why the deadline was missed and what needs to happen next. In other words, AI will be about saving time with the repetitive tasks, and the human element will remain and focus on adding reasoning and meaning to the developmental cycle.
Derwyn wonders what this means for building products today. It’s clear a lot of people today spend the majority of their work doing the repetitive tasks machines could be doing for them. According to Robin, when you are taking the time to interrelate data, there is still value in looking at it and you shouldn’t assume the mechanism will tell you what to do, similarly to remaining engaged when driving an autonomous car. Even if there are still a lot of manual tasks people spend their day on when they could save that time for less tactical work, they are ultimately responsible for the outcome of what they’ve built. There’s a sense of ownership and liability. “I wouldn’t want to say: Jama built my product for me. I don’t think this will ever be true. I think it will be: Jama helped me focus on building the right thing and in the past it would have taken me a year, while now it takes me six months.” This is because the time was spent on validating the work vs. building and validating the work, where Jama was the first level of QA and a human provided the final check.
Traceability, the way we’re building it today at Jama, opens up a lot of possibilities in the future. What does the transition look like for a single person managing excel traceability spreadsheets for a very specific purpose? What’s the education, or perhaps resistance to change that future traceability will bring? Robin hopes that what we’ve built at Jama solves the issues that an excel sheet solves and additionally makes it easier and more intuitive, or even enjoyable to work with. “The thing that an excel doc can’t do is give you feedback. It just sort of is. It presents what you told it. It’s kind of a dead thing. To manage that, you really have to build steps, checks and procedures to do it right.” The new view we’re building will include excel’s capabilities but it will give feedback, too, for example, it will tell you you’ve done this pattern 50 times and you’ve always had a certain type of relationship between epics and stories or a system of requirements and individual components, and it looks like you should do it again. Having the feedback loop within your own data will be valuable because it will keep you in the right pattern. “I hope that this transition is just doing the same but faster and better and it can set smart, educated humans free to use their intelligence and knowledge faster.”
Robin and Derwyn will continue their conversation on modern traceability and discuss how it helps or whether it’s against Agile methodologies and concepts in the systems space. Stay tuned!