Software Estimation Skills in the Wild
Description
A two-panel meme. The top panel contains black text on a white background that reads, 'This guy is a software engineer, you can tell by his awesome estimation skills'. The bottom panel is a photograph taken in a parking lot, likely outside a large hardware or furniture store. In the photo, a man is struggling to fit a very long, thick wooden beam or plank into a small, silver hatchback car. The car's trunk and a side door are open, but the beam is comically oversized for the vehicle, sticking out many feet. This meme serves as a visual metaphor for the notorious difficulty software engineers have with accurately estimating the time and effort required for their tasks. The man's gross miscalculation of physical space is humorously equated to a developer underestimating the complexity of a project, a common experience that leads to missed deadlines and 'scope creep'
Comments
15Comment deleted
He clearly estimated the task in story points but measured his trunk capacity in ideal man-days
Somewhere a PM just converted that eight-meter plank into a "quick half-day task" - and the architect signed off because it fit in the Gantt chart
This is what happens when you estimate a task as 'small' without checking if it fits in your current sprint capacity - suddenly you're blocking three other teams and the PM can't back out of the parking garage
This is what happens when you estimate a 2-story point task as a 1-pointer because 'it's just moving some data between two containers' - technically correct, catastrophically underestimated, and somehow still in production
Classic dev estimation: 'It'll fit in one sprint' - until velocity parks it in the next quarter
Classic planning-poker: it's a 3-point task, right up until it spans two cars, blocks the release train, and turns 'split across sprints' into tight coupling
Proof that story points don’t convert to cubic feet - P50 in planning, P95 in parking, parallelized across two hatchbacks
He should sprint instead Comment deleted
Unironically I have the best estimates of math and stats such topics in my work. And I still underestimate time needed for a project Comment deleted
always multiply by a fudge factor Comment deleted
2 x 0 = 0 Comment deleted
1. Make a realistic estimation. 2. Multiply by a factor of 3 to 5. 3. Now you have an optimistic estimate. Continue improving you realistic estimation skills at step 1 on subsequent occassions. Comment deleted
Bro is almost Pythagoras Comment deleted
When css doesn't load Comment deleted
css added 'if' command Comment deleted