Running Tested Features Yet Again

There’s a discussion going on on the Scrum list about whether the phrase “Agile Project Manager” might be a problem — since there really isn’t such a role in most “Agile” methods. In a comment on that thread, I found myself saying this:

Scrum works, in my opinion, because it requires two things:

  1. Produce Done-Done software on a regular basis;
  2. Remove every obstacle to doing item 1.

Agile teams often stall not because “Agile doesn’t work,” or because “Agile doesn’t fit our real world,” but because they don’t recognize that things like “We can’t get done in one Sprint because QA needs a whole Sprint to test a Sprint’s work” is not a description of the “real world” but a description of an obstacle to be removed.

This is, of course, no more than a restatement of my long-running Running Tested Features notion, as found, for example, here. My views often drift over time, but this one has not drifted for quite a while now. I might be onto something here …

I’ll take an early opportunity to ressurect the other blog articles on this subject, and reset my links to them.

Recent Articles

Scrum Information Base — Agile Skills

Chet Hendrickson and I have offered to help the Scrum Alliance build a broad and growing base of information relating to Scrum / Agile, and to the many skills and practices that can help teams be successful. Our offer has…

Why Did You Do That, Ron?

Piers Thompson sent me a good question about my recent database articles. I suspect others would like to hear the question and answers.

Events & Classes

  • No events.