... questions about homework. How far did you get? (... into parsing formulas.) What would you need to go farther? (... more knowledge of regular expressions or parsing.) Knowledge does matter!
... Planning Game.
Points of discussion: Will an estimate vary if we are implementing a story from scratch versus against a base of existing code? How, and how much? Will some stories help us get started better than others? How?
Points of discussion: mechanics of accounting, both generally and in the customer's application; other needs of this customer: custom journals; accounts with names, numbers, and sub-accounts; periodic closing.
The team talks about the stories and the application. Make and refine estimates for each story on the board.
Begin to develop minimal team coding standards. The only requirement defined by the instructor at this point is a file-level comment that tracks mods to the file by date and pair. This will facilitate in-studio communication.
Begin to plan out-of-class time: 1-2 sessions of ~1.5 hours out of class every day.
Leave the results of the meeting on the whiteboard, with marker and post-it notes.
This section is a carryover from Session 5, left as context for the previous sections.
Over the next three weeks. we will grow the beginnings of a small accounting system. Here are some of the ideas that will be part of a complete system:
Next time, we will write up stories for this system and play the Planning Game to decide what you will work on for our one release (end of the course) and our three iterations (end of each week).
Most of the significant stories will be written at the beginning of the project and divided into "must have", "nice to have", and "won't have" piles. We will draw up a tentative list of stories for the first iteration, and a less certain list of stories for the remaining iterations. As your customer, I will try to ensure a certain coherence among the priorities I set for the stories. I'll make tradeoffs among the priorities based on feedback of various kinds, including your work each iteration.
As the project goes on, we may add new stories, but in general we expect the number of stories under consideration to trend toward zero.
Read an article by Jason Fried, Getting Real, Step 1: No Functional Spec.