I'm against scrum in general, but even more so for scrum applied to Data Science or research in general. If you had to do proper scrum with these things you'd have tasks like: thinking hard about this - 5 points; staring at the screen looking at charts - 8 points; redoing everything with slightly different parameters - 8 points.
Does it make sense?
@alan Scrum is generally better for production than for research. But research also needs to be managed, and doing that usually means defining the task goal as having answered a question or three.
And if time runs out, you still have partial answers that allow you to define the next question better, etc.
It works, if you want it to.
@jens you need to be managed 😜
Jokes aside, the point is that it's "needed" only for non-working people, aka managers/product managers/etc., if I'm spending 3 months on a topic I know all the ins and outs, and I know how to record the things I care about along the process. I don't need to be handheld and I'm perfectly capable to help the people who need some organization without standups, kanbans and whatnots.
This Mastodon instance is for people interested in technology. Discussions aren't limited to technology, because tech folks shouldn't be limited to technology either!