DevOps Fables: Jirassic Parts

Based on a true story

As the sound of the helicopter engine died away, it was replaced by the thunder of the waterfall right next to the landing pad. Why did I agree to this? thought Dr Evie Sattler as she disembarked, keeping her head down and running to the awaiting jeep. "Dr Sattler!" greeted the smiling John Hermand, "welcome to Isla Hopper, let's get to the complex."

"So the whole island is dedicated to software development?" asked Evie incredulously as the EV jeep zoomed along the boulevard between the helipad and the centre of the island. "Absolutely, Dr Sattler. This is the Parts Unlimited International Development Centre of Excellence. We have everything here any of our staff could ever want or need. We built the island in the centre of the bay and provided housing, amenities, transportation and, of course, the office. We spared no expense." Just then, the jeep pulled up in front of the gleaming office complex, a huge circular glass and steel building. "Come, Dr Sattler. Let's go get you set up. I need your report so I can get some nasty VCs off my back."

The huge two-storey wood-effect double doors were mostly for show and Hermand guided them to the employee entrance round the side. They proceeded to the elegant reception to get Evie signed-in. Once they had passed through the security gates they stepped into the central atrium. The PA system echoed an announcement, "Retro of Retros is due to start in East Dock in 10 minutes. All parties, please proceed to East Dock, now." "Impressive, isn't it?" said Hermand, "over in that section to our left is our Design, UX and Rapid Prototyping Centre. We call that 'Tryassic Parts'. The centre to the right used to be where developers were based, but now they mostly work from home - even though for most of them that's still on the island - we call it 'Createtious Parts'. The section in the centre here is our Agile PMO, where we plan and coordinate all of our work. Dear Dr Sattler, welcome, to Jirassic Parts..."

Night had fallen and the weather outside had turned as stormy as the atmosphere in the central command room of Jirassic Parts. "Explain it to me again, Hermand," said Evie wiping the perspiration from her head as she reviewed the printed reports strewn around her, "just how does your work, work?" "Like I said Dr Sattler, we wanted our Developers to be as productive as possible. So we paired them up and provided each pair with a dedicated backlog of work and a coach to help them achieve it with as few distractions as possible. The challenge we find is that we don't seem to be making as much progress as we would have expected given our expenditure. We thought that the bottleneck could have been the number of items our Product people were adding to the backlog so we invested in an AI to take all their ideas and convert them into user stories automatically. This seems to have helped - all backlogs have 1,000 items in them now but we still don't seem to be delivering faster."

"Wait, wait!" said Evie, "do all backlogs have exactly 1,000 items in them or is that an average?" "Er, an average, I think," said Hermand, "why does that matter?" "Because the version of Jira you are using limits out of the box queries to 1,000 results." "You mean - we could have more?" gulped Hermand. "You've built an automated requirements gathering machine, " replied Evie angrily, "you took approaches you didn't understand and mis-used them to create massive systems. You were so excited by technology that meant that you could, that you never stopped to ask if you really should do any of this. And now the systems are autonomous, unstable and potentially really dangerous." "Just how big a problem are we talking here?" asked Hermand as the lightning flashed outside. "Well, if I look at your average arrival rates of new tickets the AI is generating a new one every 5 minutes or so and your pair programmers are delivering one of these requirements approximately every 5 days. That means in the last day alone you generated enough new work to last you for the next six years."

Hermand's deafening scream of agony echoed round the now empty building as too late his mistake dawned on him. "NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO!!!" reverberated and amplified off the walls and made Evie cover her ears in pain. The self-styled "Velocity Raptor" hung his head in shame as Dr Sattler pointed out that his grand dream was little more than a illusion, a failed experiment. "Right," said Dr Sattler, "I think we know where some of the major issues arise. Tomorrow, we can start to improve and actually start delivering value. Are you ready to start?"

Lesson: Never run before your can walk.