joel spolsky from fogbugz calls it 'Evidence Based Scheduling' ...
http://www.joelonsoftware.com/items/2007/10/26.html it's really just graphing estimation accuracy history for tasks using Monte Carlo Simulation / Analysis. bottom line is that it would be great to get a handle on the accuracy of our estimates and be able to see visually how well we're doing with it so that we can improve. in the same way that fogbugz does it... i want to see the probability that we're going to meet a certain deadline based on previous estimate history and accuracy from our developers. one thing that would probably need to change within openGoo to accomodate this... is that we'd need to identify 'who' actually created the estimate for a task. generally we allow our own developers to create the estimates for tasks that are assigned to them.... but currently in openGoo there is no way to identify who was actually responsible for the estimate on the task (unless you're just looking strictly at who the task was assigned to and consider this, but what if multiple people are assigned to a task, or the task assignment is changed?). if we were doing true estimation probability simulations... it would be good to base those projections on the numbers generated by previous estimates / completed task time from all project assignees for the current project.
