Defragging the Team - A good Analogy!

[Source: Article in Gantthead by Ian Whittingham, PMP ]

If nothing else, it makes me feel vaguely virtuous, in the way that having my blood pressure taken or cholesterol checked does. No matter, I always have to click around the Programs menu to find it, hidden away in the Accessories folder under System Tools, sandwiched between Disk Cleanup and Files and Settings Transfer Wizard. But when I click Analyze, I realize I have not been as diligently attentive to the needs of my laptop as I should have been.

A braided ribbon diagnostic displays the state of my C drive: wide expanses of white space delimited by irregularly spaced red threads, interspersed with blue lines of varying thickness. Just as in nature, file management appears to abhor a vacuum. A pop-up window chides me: “You should defragment this volume.”

What the diagnostic graphically depicts is the extent to which the files on my PC have become fragmented across my C drive. But if I switch to the peripheral memory store that occupies my E drive, the analysis is much more reassuring: large blocks of blue predominate the ribbon display. This depicts the ideal state of contiguity in which all files should reside. The bluer, the better.

The variations in these color-banded patterns--red for fragmented files, blue for contiguous files, green for unmovable files and white for free space--are generated by algorithmic behavior designed to overcome the limitations imposed on the storage of information by both the physical properties of my laptop’s storage media and the algorithms that write files to the drive. Ideally, to optimize read/writes and utilize efficiently the media’s storage capacity, files must be moved around with the aim of allocating them contiguously in the memory store, so that any available free space is maximized.

However, the continuous process of adding new, amending existing and deleting unwanted files creates churn within the memory store. As churn impedes contiguous file allocation, file extents are created to workaround the distribution of the free space available. But as more files are added, amended or deleted, more extents are created, causing more fragmentation across the volume and a consequent degradation in performance. All of this occurs in the background, unnoticed by me; which makes defragging my disk drive a necessary and periodic chore.

And so I click Defragment and wait for the ribbon display to imperceptibly re-arrange itself, bit by bit, as files are compacted and relocated. But as the diagnostic ticks closer to 100 percent complete, watching the restoration of my disk drive’s integrity I am suddenly struck by the recognition of a best practice that might equally apply to a project team as to my PC: the periodic defragging of the project team.

Project teams are just as susceptible to performance degradation, caused by churn, as the storage media on my laptop is. But what are the symptoms of churn in a project team? And how might they be remediated so that team performance is restored?

A defining characteristic of highly performant project teams is integrity. In project teams, integrity is most commonly manifest in how the team accomplishes its work. Integrity derives from the ability of each team member to operate in a coordinated and proactive manner such that the effectiveness of individual actions is reciprocated and compounded by those of other team members. Thus, the project tasks that each member performs are integral to the overall performance of the team.

There are many ways to achieve team integrity, but the most obvious root to this is through effective role definition. Each team member must have a clearly defined and acknowledged role within the team, with minimum overlap with other team roles (except where required in the case of back-up, or to maximize resource allocation to a critical path activity or function). However, this involves much more than just compiling a RACI (Responsible, Accountable, Consultative, and Informational) matrix of team member roles, and then posting it to the project website. The key here is acknowledged role.

Unless the responsibilities and accountabilities of an individual team member are explicitly acknowledged by the other team members, boundaries and ownership can rapidly become blurred or porous. If project tasks are being left unresolved and unclaimed, or if multiple team members are working on the same task in apparently disjointed or uncoordinated ways, these may be incipient symptoms of churn.

In highly matrixed organizations, contention for team resources by competing projects may also be contributory to churn. If team members are dividing their time (and attention) among multiple project threads, there can be loss of focus on meeting project objectives. As pressure increases to meet deadlines and work packages or tasks are completed in isolation without regard to integration with other project tasks, further dislocation can occur. This coping behavior is similar to the file extent workaround: tasks are allocated to the time available without regard to quality or scope of work. Thus, tasks become fragmented from their objectives. On software development projects, for example, code builds created in such environments often exhibit a higher rate of bug fixes and post-release patches.

By their intrinsic nature, distributed teams are more susceptible to fragmentation than co-located teams. While large scale meetings might appear to be an effective forum for project leaders to set direction and communicate status, it may not be true for each project team member. For example, conducting a project meeting via telecom with 20 participants across four times zones for 75 minutes each week does not necessarily create a strong sense of team cohesion, nor is it an effective use of each participant’s time. Consider that such a meeting will consume 25 man hours per week of time (almost two man months per year) and will tie up 190 point-to-point communication channels among the participants.

Just as partitioning the volume on a disk drive can reduce the number of read/writes and reduce churn, you can apply a similar principle to team member interactions such that only the specific needs of the project that are relevant to a team member’s domain expertise or functional role are engaged. But to achieve this, a trade-off has to be made in the amount of facilitation effort required of the project manager. The communication needs of each team member--as it affects the performance of their project contribution--must be well understood and acted upon by the project manager to be effective.

As in the process of disk defragmentation, the project manager also needs to create contiguity within the project team. Viewed simplistically, contiguity can be taken to mean nothing more than project team member co-location. And certainly agile project teams fit this analogy very well, especially in the results they achieve. However, contiguity can also mean much more than just physical continuity and interaction. Contiguity should be a source of team cohesion and integrity.

One way in which it can be achieved in distributed and virtual teams is by continually relating the functional responsibilities and tasks of team members to the achievement of business objectives. When teams lose sight of the objectives they also lose the point of the tasks they are executing. And although this may not always ameliorate the effects of churn, it is one way in which project managers can help their teams to defrag themselves and restore team performance.

Building Trust and Agile Practices

[Source: InfoQ]

During Agile 2008, Dr. Linda Rising held a presentation centered on experiments conducted many years ago, presenting how deep, powerfully affecting, and difficult to avoid are human “prejudices” and “stereotypes” as seen from the perspective of psychology and cognitive science. In the second half of the session, she explained how it is possible to minimize and overcome the impact of prejudice. This article is a summary of that presentation.

The following experiment was mentioned throughout the session, an experiment conducted in 1954 and known as the Robbers Cave Experiment:

Two groups of 12-year-old boys were taken to a Boy Scouts camp, but neither group knew of the other’s existence and believed that their group was the only group in the camp. (Of course, they traveled to the camp by different buses.) During the first week, the two groups carried out their activities separately. They swam in the lake, built hide-outs, and pitched tents. The accompanying adults (who were actually the researchers conducting the experiment) ensured that there was no contact between the two groups. The members of each group bonded as a team.

After one week, the two groups became aware of each other’s presence (as contrived by the adults). At this point, despite their not actually seeing each other, the two groups began distinguishing between “us” and “them” and saying “they (the other group) intruded on our territory”. The researchers were surprised by the extremes in boys’ reactions: how quickly the members of each group had bonded initially with each other and how quickly both groups had seen the other as “the enemy.”

The two groups were then brought together. In a plan set-up by the “adults”, the two groups participated in competitive games such as baseball and tug-of-war, and the total scores for various games were recorded for each group, with the adults acting as judges. The group that won with the highest total score for the day was awarded prizes. When that happened, the losing group burned the flag of the winning group and raided their camp; the two groups were on the verge of declaring all-out war on each other.

Dr. Rising broke off her story about the experiment here and turned to the topic of people’s prejudices. Humans are born with an instinctive ability to instantly determine whether what is in front of them is safe or dangerous, edible or inedible, enemy or friend. Since this ability is instinctive, it is an ability that humans acquired through competition for survival and the process of evolution. For humans in primitive times, slowness in distinguishing whether the person in front of them is a member of their own tribe or a member of an enemy tribe was a life-or-death issue. However, because of this instinctive ability, humans also came to categorize other humans — in other words, stereotyping them — the moment they laid eyes on them.

People categorize other people: enemy or friend; family or stranger; insider or outsider. This decision is made in an extremely short period of time, and categorizing then leads to stereotyping and simplification, with people making presumptions about others based on first appearances. Prejudice has two characteristics:
• Everyone has prejudices
• Nobody realizes that they are prejudiced

In other words, everybody mistakenly believes that they are not prejudiced. Consequently, while people believe themselves to be making rational decisions, they in fact constantly justify their behavior and that of their friends, family, and colleagues, and constantly presume the actions of strangers and outsiders to be “bad.”

Stereotypes simplify the way people view others. Despite the rich and complex individuality inherent in each person, people label others focusing on general outward appearance and ignoring details and difficult-to-observe features. Dr. Rising gave this example during her explanation:

“People who are married tend to use presumptive expressions when talking about their spouses, don’t they? They ‘never’ tidy up the kitchen, or they’re ‘always’ complaining. This is another kind of labeling due to prejudices.”

Moreover, when supervisors and managers evaluate the abilities of the staff working under them, the same prejudices and stereotyping occurs. In many cases, a supervisor “determines” the ability of a worker in about three weeks, labeling them as either “can do” or “can’t do” workers. Once a prejudice has been formed, the supervisor views all the actions of that worker through this filter. If two workers make the same mistake, in the case of the “Can’t do” worker, the supervisor will think, “There he/she goes again, making the same mistakes,” while in the case of the “Can do” worker the supervisor will think, “Maybe he/she wasn’t feeling well.” Eventually, the supervisor can only recognize actions that affirm their prejudice.

Not only this, prejudices can also cause people to lose the capacity to demonstrate their own abilities. Here Dr. Rising introduced another experiment. In general, females are believed to be less proficient in mathematics than males (in the case of the United States). In this experiment, the subjects were given a mathematics test. If no mention of gender was made before the test was administered, there was no difference in the results between males and females. However, if comments were made about the relationship between mathematical ability and gender before the test was administered, the females scored lower than the males. Furthermore, the same results were obtained by simply having a column on the test paper for subjects to fill in their gender, without any comments about gender being made. This is an example of consciousness of their own gender activating prejudices and inhibiting a person’s natural abilities.

Prejudices and stereotyping are rooted in human instincts, and so influence everyone. However, these can also be obstacles when many people are attempting to work together to complete a complex job or reach a common objective, missing chances for people to display their individual abilities, inviting mistakes, and in some cases even lowering the capabilities of the people who have prejudices. How is it possible to escape from these prejudices and/or lessen their impact?

Continuing the story of the first experiment, Dr. Rising said:
The researchers tried to stop the conflict between the two groups, but simply making the two groups carry out activities together produced no noticeable results. The researchers then staged an “incident” in which the camp’s water supply was cut off and all the boys had to check the water pipe (more than 1 km long) to see where it was blocked. When the blockage was found and removed, the members of both groups joined together in celebrating. By involving all the boys in the resolution of an “incident” that affected the entire camp, the conflict between the two groups evaporated.

Humans can cooperate. Given a common objective, they can work together to tackle the task at hand. In a study using monkeys, the experiment was set up so that two monkeys obtained food by cooperating with each other. Initially both of the monkeys obtained food, but the experiment was changed so that only one monkey obtained food even when the two monkeys cooperated. However, even when one of them realized that it would not receive any food, it cooperated with the other, and the monkey who received the food shared it with the other monkey.

Humans can cooperate. Dr. Rising says that this human ability is also instinctive. In order to cooperate, there is no need for the two parties to like each other. All that is necessary is that one party acknowledges the efforts of the other, and the other party recognizes the efforts of the first. A relationship of mutual respect for the ability and contribution of the other party is born from that. People feel happy to be respected and trusted; such instincts are inherent to human nature.

Taking advantage of these natural human instincts, Dr. Rising points out that Agile practices are outstanding from the standpoint of enabling people to display their abilities, especially the fact that face-to-face communication promotes cooperation…

Dr. Rising’s presentation at Agile 2008 discussed how tied up in prejudices humans are, how the capacity of teams is diminished by stereotyping, and how Agile teams overcome these negative impacts. In the opinion of the author, such knowledge cuts through the workings of the human spirit and brain and seems somehow excitingly close to the human limit.