Saturday, June 24, 2006

CTC 2006: Ken Thompson on Bioteaming

On the boat from pier to pier
Ken talking to Annette Kramer (who also blogs about this session.)

Ken Thompson, BioTeaming: Natural Models for Virtual Teams
Thursday morning, Collaborative Technologies Conference, Boston, June 22

Note, live blogging caveats apply - I capture as much as I can and I don't worry about typos, etc. I also want to note it was terrific to meet Ken F2F. He is smart and FUN!

Ken's Slides (useful to make sense of the notes below)

I’m the bumblebee. Most people think of ants, not very friendly. Bumblebees more friendly. Some people have commented that my figure is more like a bumblebee.

This is going to be participative. A little survey to start. Be honest. Put your hand up if you feel the work teams you are involved in regularly fulfill their potential. Put your hand up if the technology you use improves the team work. Your involvement in your work teams is as satisfying as your sports and personal groups.

I’ve worked with teams for 25 years, coached them for 10 and my feelings have been constant disappointment. IT could be me, like Jonah and the whale and I bring those teams downs, or it could be there is something fundamentally wrong with teams. Last three years studying biological teams and have developed 12 things human teams might learn from them.

All of this is on the blog, I’m quite prolific, 350 articles. If you can’t sleep, there is plenty of material.

Why virtual teams need to look to nature, introduce bioteaming, and then experience it. Better than explaining to a Martian how to eat an orange, better to hand him one. Learn from ants, birds and learn how to swarm. Food webs and team networks. Tech teams are tightly bound and ignore their external networks. Sales teams form only loose networks. Have to manage both. My favorite animal, the stickleback. A fish with a tit for tat collaboration strategy. If you are married, you probably use it but play it wrong.

I used to be a software engineer and used to be in terrible teams. Lots of boring stories. Went to blog and pulled out stories.

Freeloading – dead wood colleagues. 80% of staff in UK are not happy with their team members. Parasites that survive off of but don’t contribute to the work. But people won’t tell you directlyu, being British.

Motivation – ¾ people in ICM survey said they would stop working if they won the lottery. Must be real fun in their workplaces

Leadership – American managers no longer see mentoring and coaching as part of their job

Overwork – Fortune Magazine – when vacation is simply a change of work place (did not capture quote – need to pull from the posted slides on the CTC site.)

Ignorance – quote from – It’s amazing how much easier for a team to work together when no one has any idea where they are going.

Virtual teams – much worse. Research it is easier to break a virtual commitment to someone you rarely meet than a physical commitment. That’s why people lie more on the phone than F2F. Can get let down if you don’t meet the people you are working with.

Virtual teams lose their focus easily. 40% of UK web activity non work

Technology – executive managers exaggerate their use of PDAS, only 1 in 7 use them. Research reports conclude that managers of flexible workers need to manage in a different way, but virtual teams are still led by traditional managers.

The science of learning form nature is biomimicry, from the great bios, meaning many life, and mimesis, imitation. Taking nature as a model, looking at the geckos’ fate, looking at Velcro. Nature has had 3.8 billion years to get it’s designs right. They have survived the test of time, so nature is a great place to look, all the bad stuff is gone. It is not a resource to be quarried but something to teach us.

Humankind did not invent team. Nature invented them and we copied them. Social biologist Carl Anderson has ob served four types of teamwork in biological teams.

Doing different things at different times, solo work, team of one. Teams of one are brilliant, but can only do limited things.
Crowd work, doing the same thing at the same time. Ants pushing a rock, in a crowd, singing the same song.
Same thing different time – Group work, partitioned work, workflow
Teamwork – different things at the same times. Division of labor. Allocation of roles
Carl has multiple species examples. We didn’t invent collaboration and teams, we learned it from nature.

A whirlwind tour of biological teams. This could go badly wrong

The cell is a self organizing system, unique porous membrane. Let’s good things in and keeps bad things out. If you look at today’s blog you will see an IBM research article urging self organizing with porous membranes.

The nervous system. All nervous systems are distributed interplay of component. Ants occupy 10% of the biomass. If you go outside anywhere in the world you will see an ant. Tiny intelligence, but their root foraging is better than our best computers

Bees use the dance to communicate. The waggle dance. Once a bee finds honey, it does a clever dance. The orientation of the axis of the dance points to the honey. Then everyone heads out to find it. Different dance for close and far. Waggle dance beer – tried to get them to sponsor my blog and give out beer at conferences.

Termites build mounds that in relationship to their size is three times larger than our tallest skyscrapers with built in temperature relations

Dolphins use acoustic signatures, individual recognition which was thought to be only human. Dolphins recognize individuals.

Wolves – an amazing lady called Temple Grandin who is autistic and she has a book about animal behavior. She suggests that humans and wolves co-evolved. Wolves taught humans how to hunt in packs, before we were not quite as social as we thought we were. “Animals in Translation.”

Geese, we all know geese can fly 50% further by flying in slipstream. When a goose gets sick, two stay with it till the goose gets better or dies.

Songbirds – blue tits taught the whole species how to open milk cartons. Socio propagation to teach the group. Rapid evolution requires learning within the generation. Normally evolution is generational. No giraffe can increase the length of it’s own next. IN teams, can we get learning within the team or does learning only happen with new teams. Learn from the Bluetit

Sticklebacks have a complex system of collaboration. Simple, but effective.

Crocodile bird – the bird cleans the crocodile’s teeth, symbiosis. Evolution has been rewritten by Lynne Margulies who suggested that almost all evolution is triggered by symbiosis. Survival of the best pair. The theory of evolution has been altered in recent years. Teams can be quite lonely places. You have to have someone on your side.

Ecosystems show how different species sometimes compete and collaboration.

Gaia – the mystical believe that the whole world is an alive, complex adaptive system where no one is planning it.

It’s all on the blog. Biomimicry in the area of team dynamics. All free. 350 articles. Not for profit kind of blog as most blogs are. I’ve published a manifesto from Want to create a bio-friendly ecosystem of tools. The blog is a way to create this ecosystem of tools. Many of the tools today could be bio friendly. - you put up your manifesto proposal, people vote on it and if you are selected, it will be published.

The manifesto identified four zones and 12 principles

Leadership zone – every team member is a leader. Not no leader, but everyone is leader in different domains

Connect to self, partners in symbiosis and networks

Execution – bio teams execute by experiment, cooperate tit for tat, swarm and learn fast

Organization – bioteams are sustainable, self organize without constant management interventions. Autopoesis – self organizing networks. A definition of life.

The two lower zones, leadership and connectivity are fundamental. You need to get those in place first. I’m only dabbling in the higher two zones. Bits of theory could turn out to be totally wrong.

What is it like to be an ant? It’s a lot of fun. California company has produced a simulator. I’m going to play it now.

Ants communicate via smell, Pheromones. They leave scent trails. Ants are aggressive creatures. If you give them nuclear weapons, the world would be over in a week. Ants hate other ants, just like us. Ants love brownies. Going to put the simulation on. Write down one thing about their communication process as you watch. Can’t be smell. More abstract, clever stuff. Foraging is a form of swarming. Swarming has a bad reputation as aggressive. When a bee stings it send out a pheromone that signals other bees to sting. That’s attack swarming. The other form is going somewhere. Where no one knows where to go. Foraging is another form of swarming. Animals play quite a bit. Starlings just having some fun. I’ve preplanned this. I’m kind of mean, once the ants found the brownie, I moved the stone. Ants are brilliant at route scheduling. The red trail is an alarm pheromone. The other trails are food trails.

Can you tell me what you have learned about how ants communicate.
• Immediate trail to food
• A woman comes into a room and says to a man, this perfume is only for you. It doesn’t work like that. They are broadcasting, not concerned about precision. One to many.
• All are busy. Every member can broadcast. No reply. Just get the message and do it. Don’t have permission structures. When they see it, they do it.

Key bioteaming Principles
1. Information not orders
2. team intelligence, one to man broadcasting
3. permission is granted, immediate action, no check and approval
4. always on, don’t go into hive for messages, messages go to them
5. symbiosis – transparency and pervasiveness rather than privacy and precision. Over-broadcast and multiple channels. Does the message get trough
6. Swarm, forage for useful material

What does it mean for technology. E-pheromone group messaging systems. Here is what I discovered about a year ago. Pheromone for a team is messaging: IM, text messaging and RSS. It maps so well. Wherever a team member they can broadcast and if needed, receive a reply. A new type of SMS social group tools let you do a helluva lot from your mobile phone, commands to join and leave a group. Some problematic java issues. All come out in the last two months.

Open IM tools, java based solutions like meebo, and when mobile IM becomes predominant on phones without a download. Me being a business man, after I did this research I started to think that there was a business opportunity. I started to build a toolkit for those who wanted swarming in their applications, Swarmteams. Any to many in situ message divider, a plug in for creating swarm applications. Swarm participation by being invited or subscribed. Key the word “join ctc” and you would join that swarm.

Natures teams are peer swarms. In bossiness no one liked that so invented star swarms with owner having more rights, then migrate to peer or hybrid swarms

Interswarm linkages, discretionary bridges between swarms. Bond where a whole bunch of swarms need the same message. Create an ecology of swarms in organization. That is how it is done in sales. Being used in two applications, you can see a number orf swarms, ad hoc collaborations. I asked “who will win the world cup” and people can answer by sms or IM. You decide how you want to subscribe to the swarm and I can send from my phone. That’s in beta, coming out in August

Swarm club. Children don’t want to pay for their messages. May take a commercial message a day, competitions, to allow IM.

That’s enough about swarms and messaging. So what is it like being in a bird flock.

Swarming is sophisticated team behavior by simple team responses. Craig Reynolds, creating boid, showing complex bird flock flying behaviors could be simulated in 3 simple individual member rules.

Don’t collide with each other
Go in the same basic direction
No leader
Stay close together

Zaira developed this simulator. There is a group of bumblebees going around. Slide size, separation and alignment sliders. Increase alignment and they group up. Then there is a leader bee who should not bee there. Reduce separation, then they go around in circles a bit. Then increase cohesion and now they move together in a swarm. Incase flock size and it becomes kind of interesting.

What can bird flocks teach us about virtual team operations? I corrupted the word biod to Biods – virtual team members who consistently and naturally swarm

What if you had 7 automatic behaviors a team member would do in a crisis. What would you like built into every team member

Who thinks humans don’t swarm. Nobody. I went to the world cup match last week. Started constant Mexican weaves. Total collaboration between two opposing teams until Ecuador scored. Then the Mexicans would start the wave, then the Ecuadorians would stop it.

1. Outgoing. Talk to the other team members. The ones you don’t know.
2. Recruit – if you see someone useful to the team hook them on
3. Go –everyone go and forage, find interesting things, build network
4. Ask for help. Give help.
5. Note – keep yourself aware of priorities. Everyday.
6. Investigate –when you see something interesting
7. Collaborate – get working with people, particularly people you don’t know.


Not saying those are the right behaviors. If birds can do it in three, what are the seven you would expect to have in your team.

• Respect, listen to others views
• Devils advocate to see where we are going wrong. Not over attached
• To be present – polite – a protocol for intrusive technologies and how they might be used in a less intrusive sense.

What to bird flocks teach us about virtual team technology

Group messaging, activity tracking and reputation management
You should be able to track the level of messaging between members. No useful outputs are produced without the right amount of signals happening between people. You don’t know what they are talking about but know who is talking to who

Self management through enhanced reputation. Learned from Open Source. Used well reduce management overhead.

Allow team members to rank the value of interactions. If something is perceived as spam, mark it.

What is it like to be a new species entering the food web?
Clustering – team managing their internal and external networks. Software engineering teams are often poor at managing external networks. Other teams don’t manage their internal networks – essential for getting things done. Team social networking software
Teams need to find their niche
Porous membranes – dynamic team boundary. Not according to a plan. Emerge. Grow naturally. Opportunistic

Ecological language, to survive a new species must find it’s niche =- missed some stuff

Relationships between who eats who. Predator. Cats consume mice. A predator for me is competing for customers. Sometimes compete for resources. Should distinguish. Symbiont – birds coordinate with ticks. That’s why you don’t’ stroke ‘em.

What does this teach us about teams. Know position on web…

Marco Inansit at Harvard – the Keystone Advantage

Focus on collective networks
(Lots of stuff I did not catch – see slides0

Three levels of commitment – inner, mid and outside part time experts and hangers on. Every team you identify your place. All are vital.

Jessica Lipnak talks about the three rings. You and they decide where people are and structure communication to reflect that.

Aggregate individuals into composite groups. Use SNA to determine. that analyze vs phone and email connection.

This is a much more important idea than showing structure and org charts. Show the interactions.

Loose boundary systems. Accept our partners have their own systems. Get them into our team with a single log in.

What is it like to be a stickleback. Tit for tat. Concept of win win is too complex for nature. Tit for tat has three rules

1. Never be the first to defect
2. Retaliate only after partner has defected
3. Be prepared to forgive after carrying out just one act of retaliation

Milinksi (1987) showed that stickleback fish work in teams – they continually adjust. For predator inspections. Natures form of collaboration.

The prisoners dilemma (Axelrod) : Tit for tat is the best long term strategy for long term human cooperation. TFT constantly wins over other strategies. Most teams don’t know TFT. They say “I play win win.” But what happens is someone else is not playing Win Win, you get hurt, cynical, stand off and reduce trust. If you start with that style, expect to be betrayed, you will be betrayed.

Win win is not a good strategy because there is no mechanism for checking a non cooperating partner.

MIT games for teaching this, Barry Klopfer

Play on PDAs. (DEMO OF GAME)

Don’t bioteam without addressing the X factor. What makes human teams special. Intelligence, autonomy, motivation and belief. Insects are equally motivated and competent. Humans aren’t
Learned optimism – Seligman. Optimistic teams rebound better from failure. Did research project with software teams to identify top beliefs

Clear accountability
Trusted competency
Give and take
Outcome optimism

A project I’d love to do is to explore this before the project starts as a predictor of outcome.

• Virt teams need tob ecome more like bio teams
• Messaging
• Ability to swarm
• Personal cooperation strategies TFT
• Networking

But remember not everything they do is worth copying. (penguin clip)

Masking in biology, protect from predators. ON net, anonymity, avatars. What about authenticity. Could reputation management handle that anonymity factor. Or does it not even matter with TFT? Key thing about bioteams is they over-communicate. No secrete. Communication is eavesdropped by collaborators and predators. Not precise communication. Can’t always identify the author of the communication. Good evidence of thick communication.

With ants and bees there is not gender like we do, run the risk of stereotype, men and women in collaboration and competition sometimes have different approaches. Haven’t dared go near the M/F but have looked at different cultures. IN science community women are starting to rise due to different mentoring patterns in female centric groups. Different rules. Maybe a symbiosis of the sexes in teams.

What about personality types. I’m on software development. Introversion. The social networking aspect is challenged. Small teams work well, but interactions with other teams, product managers, working with more extroverted cultures. How do you account for that? I don’t think you can make introvert software engineers network. You have to put someone else in the team who acts as a hub to keep them connected to the bigger environment. Challenging teams to keep connected. The other principles can be very strong.

I have another presentation for software teams. One of the areas is getting them focused on different things. Not just on outcomes, but on signals. How do you conceptualize?

We don’t talk in the hallway and offices, but we talk online. Introverted in one environment and open in another. Between IM and a CoP discussion forum – very verbal, but not F2F. Harder to engage.

Certain teams may not lend themselves to bioteaming. Routine, unmotivated, not focused teams. I was seconded to the public sector and it is difficult there. You use individual techniques as they fit.

Tags: , ,


Post a Comment

Links to this post:

Create a Link

<< Home

Full Circle Associates
4616 25th Avenue NE, PMB #126 - Seattle, WA 98105
(206) 517-4754 -