JavaPulse

a finger on the pulse of the freelance Java™ market in the Netherlands

Itay Talgam: Lead like the great conductors

Posted on | 24 October 2009 | No Comments

Great analogy for leading organizations.

Useful Links for Sun Certified Enterprise Architect (SCEA) Exam

Posted on | 23 June 2009 | No Comments
Tags: | | | | |

Note: Design Patterns are workarounds for limitations of the languages and platform. As the platform evolved, some design patterns have become obsolete. e.g. Service Locator is no longer needed with Dependency Injection
It is important to determine which design pattern is replaced by new constructs in the j2ee platform. More on this later.

Jeff Sutherland Talks Shock Therapy for SCRUM

Posted on | 22 March 2009 | 2 Comments
Tags: | | | |

On March 5, I attended a presentation by Jeff Sutherland, co-creator of Scrum, on Shock Therapy for Scrum.

Scrum Butt
Shock Therapy was introduced because most teams that think they are doing Scrum are actually only doing Scrum Butt – “We are doing Scrum, but we don’t do all of Scrum for some reason or another”. Perhaps there are good reasons for excluding some parts of Scrum, but without truly understanding the reason behind a Scrum practice or the implications of skipping it, agile teams struggle to become hyper-productive, or mistakenly, struggle with Scrum itself.

Nokia Test and Statistics
In 2005, Bas Vodde developed a test for Nokia to quickly assess the status of a team that claims to be doing agile. In 2008, Jeff Sutherland extended this test with multiple choice answers and a scoring system. This resulted in the Nokia Test. Although it does not cover all of Scrum or all best practices often used with Scrum (such as XP practices), it aims to save teams from the common dysfunctions of a Scrum team.

The Nokia Test is separated into 2 parts: the first to deal with iterative development and the second to deal more specifically with Scrum. The gist of the questions are as follows:

Part I Are you doing iterative development?
1. Are your sprints timeboxed to 4 weeks or less?
2. Are your software features tested and working at the end of each iteration?
3. Do you start with agile specifications (user stories, just-in-time specs)?

Part II Are you doing Scrum?
4. Do you know who your product owner is?
5. Do you have a product backlog prioritized by business value?
6. Is your product backlog estimated by the team (i.e.people who do the work)?
7. Does your team generate burndown charts and know its velocity?
8. Is the team’s work free from disruption by project managers (or anyone else)?

Of the above, only 50% of teams claiming to be agile pass Part I and only 10% pass Part II. These are world-wide statistics.

Take the Nokia Test yourself here.

Introducing Shock Therapy
Teams starting with Scrum often come across similar challenges or pitfalls that prevent them from benefiting fully from Scrum. In the past, it would take 2 years for a company to achieve 300% improvement. Some teams never figure it out. Shock Therapy has consistently led teams to 300% increase in productivity in 6-8 weeks.

Shock Therapy is:

“A good set of practices, but no choice.”

Jeff Sutherland compared Shock Therapy to martial arts training. You enter the dojo and do what the Sensei (teacher) says. You repeat exercises over and over until it is part of muscle memory. Only when you have mastered the basic practices are you allowed to improvise. And the last and most important – Before you have gained discipline, centering, and flexibility, you are a hazard to yourself and others.

Because Scrum, as a framework, is flexible and customizable to a company’s own practices, teams starting with Scrum often have too many choices and don’t know how to make decisions. We don’t expect customers to know what they really want until they have seen a prototype, so why do we expect Scrum teams to know exactly what they want before seeing a prototype? Shock Therapy is the prototype.

Case Study: Scott Downey as Scrum Sensei at MySpace

  • To take teams to hyper-productivity (>240%), on average took 2.9 days per team member where the team includes Scrum Master and Product Owner
  • Rules stay in place until 1) the team starts to go hyper-productive 2) the team completed 3 successful sprints 3) have shown a good business reason to change the rules
  • Rules are:
    • Everyone attends Scrum training before starting first sprint.
    • Sprints are 1 week long.
  • Scrum Sensei’s Definition of Done (DoD ensures quality as a balance to speed) is introduced after basic success:
    • Feature Complete
    • Code Complete
    • No Known Defects
    • Approved by Product Owner
    • Production Ready
  • Estimates are in Story Points
  • Scrum Sensei controls Scrum board until team is ready
  • Condensed Sprint Review, Retrospective, and Sprint Planning into one 4 hour meeting per week. Over time, the team needs less time and will move down to 90 minutes.
  • Incomplete user stories are rejected before the Sprint
  • Multitasking is forbidden: context-switching takes extra time and slow things down.
  • Scrum Sensei gets respect by removing the team’s worst impediment within a few days.
  • On average, after 6-12 weeks, teams are functioning at above 500%. (Scott Downey)
  • One team was at 1,650% after 16 one-week sprints
  • Teams usually start out resisting the Shock Therapy. After a few weeks, they are indifferent. “Then they scream bloody murder if I try to take Scott away from them.”

In Perspective: The Big Picture
In terms of organizational changes, bottom-up (participatory) changes tend to take longer to take root, but also tend to have more long lasting results. In contrast, top-down (directive) changes may have quicker results, but are harder to get ingrained into the organization (especially when measurement of change is not in place). Real organizational change need to be participatory, with strong leadership supporting the changes along the way.

I see Shock Therapy as a combination of top-down directive and participatory change. Starting with a top-down directive, backed by management who understands Scrum, teams are able to participate in the introduction of Scrum. Most skeptics reject Scrum practices because they cannot understand the benefits without experiencing them. These skeptics often are convinced after some smell of success. Experience show that Scrum teams introduced to Scrum in this way reach hyper-productive quicker and tend to continue the trend of improvement after the Scrum Sensei is gone.

  • Even if you don’t believe the numbers, can you really afford to ignore the possibility that they are real?
  • What would it mean for your business if you are able to reduce time-to-market of new funtionality to a matter of weeks rather than months or even years?
  • What would it mean for your business if your competitors achieve this time-to-market reduction?
keep looking »