Quote:
Originally Posted by clowntable
Estimating the hardness of tasks as a team activity ("planning poker") seems inferior to analytically coming up with a value on your own or consulting the required coworkers on a need basis for example.
When we do planning poker, each dev in the team analytically comes up with a value and shows it with his card. Then we compare and if there are big differences in estimates we discuss why we feel that way. This both keeps people honest and also helps people understand the task better when someone in the team might have a better understanding. I really think it works better than doing it solo.
Another reason is that in scrum the tasks aren't pre-assigned to each dev. Everyone has equal say and there is no lead dev. And in theory everyone should be able to jump into a task and complete it, although obviously in practice some people might have a preference or more skills for particular tasks.
Quote:
Originally Posted by suzzer99
I've worked at non-scrum places and it's really easy for programmers to slip through the cracks and just do nothing for months on end. That is the biggest value of daily status meetings imo.
this.
And the trick is to keep them short and quick. Anything more than 15 minutes is not good imo, which means teams need to be reasonably sized. All you're doing is saying what you did yesterday and what you're gonna do today. Sometimes issues are raised and if its outside the scope or timebox of standup then a followup meeting can happen.