Finding the Lowest Value
Max’s team moved into a new office, which brought with it the low-walled, “bee-hive” style cubicle partitions. Their project manager cheerfully explained that the new space “would optimize collaboration”, which in practice meant that every random conversation between any two developers turned into a work-stopping distraction for everyone else.
“Silos,” the project manager explained, “are against the ethos of collaboration. We’re all going to be full stack developers now.” Thus everyone’s job description and responsibilities changed overnight.
Add an overly ambitious release schedule and some unclear requirements, and the end result is a lot of underqualified developers rushing to hit targets with tools that they don’t fully understand, in an environment that isn’t conducive to concentration in the first place.
Max was doing his best to tune out the background noise, when Mariella stopped into Dalton’s cube. Dalton, sitting straight across from Max, was the resident “front-end expert”, or at least, he had been before everyone was now a full-stack developer. Mariella was a long-time backend JEE developer who hadn’t done much of the web portion of their application at all, and was doing her best to adapt to the new world.
Max did his best to ignore the conversation. He was right in the middle of a particularly tricky ORM-related bug, and was trying to figure out why one fetch operation was generating just awful SQL.
“Hrmmmm…” Dalton said, tapping at his desk and adding to the distraction while he thought. “That’s a tough one. Oh! You should use a filter!”
“A filter, what would I filter on?”
Max combed through the JPA annotations that controlled their data access, cursing the “magic” that generated SQL queries, but as he started to piece it together, Dalton and Mariella continued their “instructional” session.
“In the filter callback, you’d just check to see if each value is the lowest one, and if it is, return true, otherwise return false.” Dalton knocked out a little drum solo on his desk, to celebrate his cleverness.
“But… I wouldn’t know which value is the lowest one, yet,” Mariella said.
“Oh, yeah… I see what you mean. Yeah, this is a tricky one.”
Max traced through the code. Okay, so the
CUST_ID, so why is it generating a
LIKE comparison instead of an equals? Wait, I think I’ve-
“Ah ha!” Dalton said, chucking Max’s train of thought off the rails and through an HO-scale village. “You just sort the array and take the first value!” *Thumpa thumpa tadatada* went Dalton’s little desk drum solo.
“I guess that makes sense,” Mariella said.
At this point, Max couldn’t stay out of the conversation. “No! Don’t do that. Use
reduce. Sorting’s an n(lg n) operation.”
“Hunh?” Dalton said. His fingers nervously hovered over his desk, ready to play his next drum solo once he had a vague clue what Max was talking about. “In logs in? We’re not doing logging…”
Max tried again, in simple English. “Sorting is slow. The computer does a lot of extra work to sort all the elements.”
“No it won’t,” Dalton said. “It’ll just take the first element.”
“Ahem.” Max turned to discover the project manager looming over his cube. “We want to encourage collaboration,” the PM said, sternly, “but right now, Max, you’re being disruptive. Please be quiet and let the people around you work.”
And that was how Dalton’s Minimum Finding Algorithm got implemented, and released as part of their production code base.
[Advertisement] Manage IT infrastructure as code across all environments with Puppet. Puppet Enterprise now offers more control and insight, with role-based access control, activity logging and all-new Puppet Apps. Start your free trial today!