Review:
Agile Project Management with Scrum, by Ken Schwaber
Publisher: |
Microsoft |
Copyright: |
2004 |
ISBN: |
0-7356-1993-X |
Format: |
Trade paperback |
Pages: |
153 |
Ken Schwaber is one of the two original developers of Scrum and one of the
signatories to the Agile Manifesto.
Agile Project Management with
Scrum is an introduction to Scrum by him that feels aimed primarily at
business people (project managers, executives, line managers) rather than
at technical people. There's very little about agile software development
in general; the emphasis is on business process, team organization, and
how the Scrum process looks to the rest of the organization.
After an introductory chapter laying out the basics of Scrum, the
remainder of the book is primarily case studies. Schwaber presents some
principle and then discuss an anonymized case study that illustrates the
principle. Each major Scrum role (ScrumMaster, Product Owner, and Team
Member) gets their own chapter, interspersed with chapters on planning,
reporting, and scaling. Finally, there are some concluding appendices
laying out the rules of Scrum in more detail, adding a brief (and very
disappointing) discussion of fixed-term, fixed-price contracts, and
providing some other supplemental information.
This is only the second regular book that I've read about
Scrum, but my
reading has been a bit out of order. We've been using Scrum for various
projects for about a year and a half, I've played both team member and
product owner roles, and we've been thinking hard about how to integrate
Scrum more into the way we do development. My impression of this book is
that it's intended as more of an introduction for people who aren't yet
comfortable with Scrum or who are still struggling to make the pieces fit
together. I'm therefore possibly not the target audience. Much of this
book is a rehash of things someone who has studied Scrum will already
know.
I'm of two minds about the case study presentation style of this book. On
one hand, it does help to make the points Schwaber discusses more
concrete, and it's difficult to talk about business process in any other
way. They're also more interesting and engaging than Schwaber's straight
rule descriptions. But I had difficulty digging into most of the ones
here. They're very simple and a little shallow, and I had trouble
applying many of them to practical situations I might be in. I would have
liked somewhat more detailed stories with more analysis and possible
alternatives within the situation, rather than simple illustrations of how
Scrum solves the problem that Schwaber had just said it would solve.
What I did like about this book is that it provides the background and
theory of Scrum as empirical process control much more clearly than any
material I'd previously read. Schwaber is good at explaining what
empirical process control is and why it's a superior approach, and he gave
me new vocabulary and basic concepts to use to talk about Scrum. I also
liked the high-level look from a management perspective without getting
into the details of story points and velocity estimation (Schwaber doesn't
use them at all in this book, relying instead on traditional "days of
work" estimates). Much of my previous experience with Scrum was tightly
focused on its advantages for estimation and responsiveness to changes,
and it's nice to see some of the other advantages of Scrum (developer job
satisfaction, improved focus, incremental deployment) highlighted more.
Another part of this book I appreciated was the discussion of sprint
length. Schwaber doesn't go into the possible alternatives, largely
assuming that a month is the best interval, but the discussion of the
impact and implications of a month sprint length helped me see the
advantages of longer sprints. (To date, we've been doing either one week
or two week sprints.) I came away with the feeling that a hybrid model
where longer sprints are used for business purposes and then subdivided
into shorter sprints for developer coordination might be particularly
effective, although that's not an idea that Schwaber presents.
This is not a particularly compelling read, particularly if, like me,
you're not a huge fan of genericized case studies. Schwaber's writing is
clear, but wooden and occasionally awkward, and the book is often dry.
But it talks about Scrum from a management angle instead of a developer
angle, which is an additional perspective that I found valuable. Possibly
worth taking a look if you want more detail about Scrum from a high-level
management perspective, although I don't think you're missing a lot by not
reading it.
Rating: 6 out of 10