3 ms·
The project was interesting because it involved the new process (Scrum), a new technology platform (Adobe Flex), and a fair amount of requirements (rework) and
by wallflower 18y ago
The project was interesting because it involved the new process (Scrum), a new technology platform (Adobe Flex), and a fair amount of requirements (rework) and technology (refactor) churn. And it was a "1.0" product. And there were high revenue expectations (which were exceeded+met, at least until the recession)
We did do fairly large stories (e.g. complete UI panels) but we go slammed when the requirements changed and we had to go back and rework what was 'done' in Sprint 4 during part of Sprint 6. I always joked that I was going to give a well done steak to my scrum master at project's end (e.g. a raw piece of meat).
I fear that my first gulp of Scrum has left a bad taste but it was an intense experience. We are much better now - very little requirements churn and we have a UXD group. We don't have Mighty Morphin' Wireframes now. Scrums are now managed according to the scrum master, which means if the SM is relaxed, hands-off (like ours is) we don't get pressured to update the burn down every day.
Burn downs. During the first project there was intense pressure to meet the burn down expectations so people would work overtime and that skewed their accuracy and added stress. What ended up happening in practice was that some tasks took hours from other tasks.
- DanielBMarkham 18y agoTo the degree you are relaxed, having fun, and working hard, it's working. Use that as a guide. Scrum/Agile isn't magic or new -- it's just stuff that smart teams have been doing for decades. Sorry to hear about the bad taste it left. Sometimes it isn't the way you work but just the initial mix of folks that make a big difference. Some people are okay with chaos and changing requirements (ie, they know how to nail down what they can and what they can't and do the work expecting change) and some folks aren't. Remember that agile teams aren't just a big list of stuff where you do a little each sprint. That's something you could train monkeys to do. You're supposed to be managing the big picture, working risky items first, designing for change, etc -- all of the good stuff that keeps you out of those "done" screw-ups. EDIT: Burn-down mania: make your sprints small (a week or two). If you end up working OT, it's simple: don't buy so many stories the next time. Some guys will say you should never work OT, but I find that impractical in the real world. Better to have smaller sprints. If something hurts, stop doing it. I also find PMs pressuring teams to buy more stories than they can do. The PM shouldn't be the CSM because of this conflict of interest. The team should be in charge of what it can do or not inside a sprint. Thanks for the story.
- wallflower 18y agoAgreed. Four-week sprints are more like a frigging 10-k. And some of them were run in perilous, icy conditions (1.0). However, that's the in-company standard now. We're ok with it - we've learned much more about team velocities and how to balance it with product owners' expectations.