The Agile Gymnasium

January 12, 2016


I used to be a weightlifter. All through college, and for much of my adult life I have been in gyms exercising in one form or another. I’ve had some modest success. The experience of joining a gym goes along some standard lines. You’ve probably done it yourself. You show up and they take you around the facility and orient you to the equipment. They may even go so far as to give you some very basic training. You get an introduction to circuit training and then they slap you on the butt and tell you to “go be awesome!” You can record your exercise sessions on this little card over here…

That’s pretty much it.

As you might imagine, the success rate with that sort of system is fairly low. A lot of people never come back (although many continue to pay their monthly dues). Those who do come back typically have no idea what modern exercise programming looks like and simply go through the motions: they ride the stair master, do a few sit-ups, and maybe do some curls. That sort of exercise has some marginal utility – you get some small amount of aerobic benefit, but it’s a far cry from exercising a meaningful percentage of most people’s potential.

Most people stop there, but there are a few who have a more ambitious goal in mind. They may be trying to improve their tennis game with better conditioning. They may be looking to build massive pectoral muscles (like most teenage boys). They may be trying to maintain their conditioning in the off season of their sport, perhaps like cycling in the winter. In other words, the purpose of their exercise is to improve their performance in some sort of real world scenario.

I’d like to pause for a moment here. I was listening to a discussion with some folks who owned their own gym and they had an interesting model. It had three tiers to it:

  1. Gym Work: Work in the gym is not like the real world at all. It is where you go to prepare for the real world. The gym is a safe place to work to the point of failure (that’s important) and to learn.
  2. Expeditions: Expeditions are adventures in the real world that are guided by a coach. So it is real world experience, but with someone there to guide you and help if you fail.
  3. The Real World: This is where it all comes together. Ultimately, this is where the training in the Gym and the experience in the expeditions pays off in terms of improved performance.

As a model for the role of training for high performance, I thought this made a lot of sense. There was one more thing that they added to this: They were capturing data on the entire group’s performance and analyzing it in order to provide better training for individuals in the future!

So when you join the gym, you use a training program that is similar to what others in the gym are using. Your performance of that program is measured and metrics across the entire population training in the gym are measured. Then experimental changes are made to the training program and their benefit (or lack thereof) is measured across the group. Gradually their training program improves over time. But the training isn’t just tested in the gym. They also track the performance of their members when they go on expeditions. This measures the effectiveness of their training program in the real world.

OK, enough about this gym. What if we could use the same metaphor for the way we train our development teams? Training would be a weekly thing. Something where you go in for training on a periodic basis to firm up your skills. There might be repetitions (pair programming, mob programming, etc.) and there might be coaching (coaching circles, etc.) and there might be someone who is coordinating the training program and measuring the performance across the entire group of trainees.

There could be expeditions from time to time. Hackathons where people get to try out what they have learned in the gym out in the real world. You know: build a real project, maybe deliver something over a weekend. Test out your mastery of your skills in the real world – with a coach there if you need it.

Then there is game day – the real world. You take what you have learned and join a team. You get to flex your massive coding and collaboration muscles and help build something challenging – something amazing. What a great model for development! But I’m not done yet…

Let’s take this model, we’ll call it the “gymnasium model”, and apply it to something like Certified Scrum Master Training. Right now, there is two days of class time and exercises and then they slap the CSM on you and send the newly minted CSM out into the world. It’s a hauntingly similar scenario to the average person’s experience at the Gym: welcome to scrum, now “go be awesome!” Maybe you do a few sprints, do a few standups and off you go. That’s about as agile as most people get. Seriously. You get some marginal benefit, but that’s about it. It could be so much more.

But what if we did things differently? What if instead of signing up for a 2 day class, you were to join an Agile gym. Maybe twice each week you go into the gym to “work out”. A coach would give you a workout, perhaps something like this:

1. Dysfunctional Standup
2. 3 Reps in the coaching dojo
3. 2 Sets of mob programming
4. 2 reps of code katas
5. 1 cool down with a retrospective

That’s just a sample workout. The Agile Gym is a safe place to try out new skills and to push ourselves. The coach would be responsible for measuring the effectiveness of the workout and modifying it over time. Experimenting with new techniques and combinations of methods and evaluating the outcomes. Of course, this is just training in the gym. From time to time we are going to need to test our our competence in the real world. The coach would provide some guided expeditions (perhaps twice a month). For example:

1. Participating in a Hackathon
2. Participating in a Startup Weekend
3. Participating in a Maker Fair

These are events in the real world that are important places to evaluate the effectiveness of our training in the gym. If our coding skills have improved, then we should do well at these events and build confidence in our ability to use our newfound skills in the real world. Speaking of the real world, hopefully now we would see the agile behaviors that we have practiced being manifested in useful ways in the actual projects that we are running from day to day. Our collaboration skills should be tight, our planning impeccable, our retrospectives revealing. And if we find any weak areas, then it is back to the gym for more training.

In this model, the gym is always open. You actually practice your skills and see improvement. What an amazing way to learn about agile!

It’s not a bad model really. Actually, it’s a really darn good one. Who wants to start a gym?

Ripping the Planning Out of Agile

October 10, 2014


Recently I was following some twitter feed about #NoEstimates. I’m no expert, but it seems to be a conversation about the fundamental value, or lack of value, that planning provides to teams. What they seem to be arguing is that planning represents a lot of wasted effort that would be better spent elsewhere.

Fundamentally I would have to agree. I’ve wasted a tremendous amount of time arguing about story points, burning down hours, and calculating person days – all for what seems like very little benefit.

What I would rather do is spend more time talking about the problem we are trying to solve. I really value a deep understanding of the system and the changes that we intend to make to it. If I have that much, then I’m well situated to deliver fast enough that nobody’s going to give me much grief about not having estimates. That’s my theory anyway. The sooner you can deliver working software, the sooner people will shut up about estimates.

But often we never do talk about the problem at anything other than a very superficial level. We spend most of our time trying to size the effort according to some artificial schema that has nothing to do with the work or any real empirical evidence at all.

So what if there were no plan? What if we took Scrum and did everything but the planning? You show up Monday morning and you have no idea what you are going to work on. The team sits down with the customer and talks about their most pressing need. They work out what they need to build, make important design decisions, and coordinate among themselves. At no point are there any hours, or points, or days. What would happen to the cadence of the sprint if we removed the planning? Basically, we would have our daily standup, and then we would review our accomplishments at the end of the sprint and look for ways to improve.

That sounds pretty good actually. Like anything else, I’m sure it has pros and cons:

Pros: Save time and energy otherwise wasted on estimation, and use that time instead for important problem solving work.

Cons: Stakeholders really like estimates. It’s like crack. They start to shake and twitch if you take their estimates away. Not many will even let you talk about it.

It might be worth a try sometime. It would certainly make an interesting experiment for a sprint or two. What if the sprint were focused entirely on the improvement cycle instead?

This is the Way Scrum Ends

September 30, 2014

Processed with VSCOcam with x1 preset

This is the way the world ends
This is the way the world ends
This is the way the world ends
Not with a bang but a whimper.

– T.S. Eliot

Did you ever wonder if this is the future of Scrum? Will it eventually go out with a whimper? I think a lot of people fear this fate for everyone’s favorite framework. Go to a conference or follow your favorite luminary on Twitter and you hear a chorus of “That’s Scrumbut!”, “It’s FrAgile”, or “Welcome to Scrummerfall!” And maybe that’s the way it has to be. Perhaps all great new ideas eventually become diluted in a sea of mediocrity.

I think I hear a longing in some to fight such dissolution. To resist the forces of corporate entropy. Rather than try to fit in, they urge us to confront and overturn the system. You know, subvert the dominant paradigm? Confronting this dissonance is the difference between making a living and actually living.

I wonder if that’s the difference between those who “fire” their customers and those who stay and work within the system. Are those consultants who give up and declare, “These clowns aren’t ready for Scrum.” going out with a bang? And what about those who stay? Are they afraid to make the big moves and just content to fit in? Whimper. Or are they more subtle than that? Can you embrace your client and still change them? Perhaps the “bang” approach is quicker, and more decisive. And maybe, just maybe, remaining engaged is very, very hard, but yields results in the end.

I know, I know…why so bleak? Well, I feel this tension a lot in our weird little community. I’ve been on both sides of the engagements where a respected consultant has tossed their hands in the air and walked away from the engagement because “They just don’t get it.” or “They’re not ready yet.” And I’ve been that poor fool, laboring away within the system, living on a meager diet of optimism and the occasional conference, trying to make change happen. I won’t pretend to know which approach is right, or even when to use these strategies, but I think it would be worthwhile to understand this issue better.

Team Genetics

September 28, 2014


Today I was listing to “The Splendid Table”, a great cooking show on NPR. They were talking about variation in growing heirloom tomatoes. Somehow, that got me thinking about agile teams (probably time to see the therapist again). It occurred to me that ideas like Agile are memes.

Richard Dawkins defined a meme as “an idea, behavior, or style that spreads from person to person within a culture.” and Agile certainly fits that definition. Agile has spread from obscurity to worldwide acceptance within 20 years. In another 20 years I fully expect that waterfall, plan driven methods will be nothing but a footnote in the history books. Despite some early prognostications to the contrary, Agile has grown at a very healthy rate over the last several years.

“Richard Dawkins invented the term ‘memes’ to stand for items that are reproduced by imitation rather than reproduced genetically.”

While much of the credit belongs to the teams that actually do the hard work of making a new process work, there is also the business that has arisen around evangelizing and introducing Agile to companies that deserves a great deal of the credit. Agile training and consulting has done a remarkable job of spreading the meme throughout the software development world.

I think there are characteristics of Agile training that have made Agile “sticky” as a meme. Much of the Scrum certification is based on plenty of hands-on exercises. Training and certification has yielded a decent business. I’m not sure if anyone has the numbers, but I’d be surprised if it wasn’t a multi-million dollar enterprise worldwide. Strangely enough, much of that spreading has been through imitation. There is no shared agenda for the training, much of it is simply imitated from trainer to trainer.

When trainers and others spread the meme they are like Johnny Appleseed sowing Agile ideas across fertile corporate soil.

Genes change with each generation, and so do ideas. They go through a mixing and blending each time they are shared. Parts of the idea are forgotten, other new ideas are grafted on. Soon the original idea is unrecognizable. I think that’s kind of what has happened with XP. Extreme Programming originally contained a collection of ideas that were very potent. Things like pair programming, continuous integration and others all served as core ideas within XP. Over time, those ideas have been co-opted and found their main expression in Scrum. Today, almost no one trains teams in XP, Scrum is the dominant process that is trained and introduced to teams.

“Memes do this through the processes of variation, mutation, competition, and inheritance, each of which influence a meme’s reproductive success.”

So too does Agile. In recent years methods like Kanban and ideas like No Estimates have arisen and are becoming a meaningful part of the software development landscape. These are evolutions of the Agile meme. Agile is evolving, I wonder where it will go next…

Broadcast Communication

September 27, 2014


In the agile development community, we are all hip to the notion of two way communication. It can be via any mechanism you choose: email, instant messaging, smoke signals or the hands-down, all-time winner – face to face. That’s fantastic, but there is another form of communication that we can develop: one-way communication.

What’s the value of that, you ask? Isn’t two way communication a lot better? The answer is yes, two way communication is great and has it’s place, but one way communication has a different purpose – one that agile teams should learn to develop as well. In fact, most agile teams don’t do very well at the one way communication beyond the team at all.

Let me explain: One way, or broadcast communication doesn’t require any response. You just shout out the news and go about your business. Now of course if there is no one to hear the news, then it really doesn’t make much difference (if a tree falls in the forest…). However in the case of working on a team, there is usually someone around. Broadcasting simply shares information with absolutely no expectation of any information or reply in return. It’s all giving and no receiving. Others can get the information and then act accordingly without ever engaging in dialog.

Some examples of one way communication include: status reports, information radiators: including burndown charts, kanban boards, etcetera. There are tools that promote one way communication such as Twitter and Yammer. I suppose even a wiki or blog qualifies too.

There is one other thing about broadcast communication that I like, especially when it comes to swarming. One way communication removes any expectation of compliance. When you broadcast information, the receivers get to decide what they want to do with it. There is no expectation of any sort of action. Commands are weakened or non-existent with this type of communication. That’s a good thing if you are swarming.

A few sentences back, I made the claim that Agile teams aren’t very good at broadcasting information beyond the team. Many of the teams that I work with tend to be very inward facing. The communication is rich between team members, but it’s very sparse if you are outside the team. This may also be a reflection of the hierarchical nature of many of the companies I’ve worked with. Teams need to take advantage of every mechanism they can find to radiate information outside the team. Some opportunities include:

  • The Scrum of Scrums or other program or portfolio meetings
  • Information radiators OUTSIDE the team. Broadcast doesn’t work if everyone has to come to you to get the message.
  • Attending other forums, other teams status meetings
  • Status reporting – yes, status reports are the root of all evil, but they are a form of one way communication.

If you aren’t using one way broadcast, give it at try. It’s a powerful communication tool – and essential to promote swarming.

The Grumpy Scrum Master

September 17, 2014

grumpy dwarf

“Going against men, I have heard at times a deep harmony
thrumming in the mixture, and when they ask me what
I say I don’t know. It is not the only or the easiest
way to come to the truth. It is one way.” – Wendell Berry

I looked in the mirror the other day and guess what I saw? The grumpy scrum master. He comes by sometimes and pays me a visit. Old grumpy looked at me and I looked at him and together we agreed that perhaps, just this one time, he just might be right.

We sat down and had a talk. It turns out he’s tired and cranky and seen this all before. I told him I can relate. We agreed that we’ve both done enough stupid to last a couple of lifetimes. No arguments there. He knows what he doesn’t like – me too! After a little debate, we both agreed we don’t give a damn what you think.

So we decided it was time to write a manifesto. That is

We grumps have come to value:

Speaking our mind over listening to whiners

Working hard over talking about it

 Getting shit done over following a plan

Disagreeing with you over getting along

That is, while the items are the right are a total waste of time, the stuff on the left is much more gratifying.


Role != Job

September 16, 2014


When I talk to folks about Scrum, one of the points I make sure to cover is the holy trinity, the three basic roles in Scrum: Product Owner, Scrum Master, and Team. I’m starting to think I must be doing it wrong because when I talk about roles, somehow that role manifests itself as a job. Let me back up a step and see if I can explain what I mean. To me, a role is a transitory responsibility that anyone can take on. It’s akin to what actors do. Actors take different roles all the time. But when an actor takes a role, say as a teacher, they act in every way like a teacher, without actually being a teacher. They do it and then leave it behind and move on to the next role. They may perform the role so well that you can’t tell the difference between the actor and the teacher, but to the actor teaching is still just a role.

Now there are people for whom teaching is a job. A job is very different from a role. You are hired for a job. A job is something that you identify with and are assigned to. A job, at least for some, becomes something that they identify strongly with (i.e. “I am a teacher.” or “Teaching is what I do.”). A job is a very different thing than a role. A job comes with identity, some feeling of authenticity and permanence. Typically we hire people to perform jobs.

According to this definition, jobs and roles are very different beasts. However, people have a hard time keeping this distinction in mind. We tend to take roles and turn them into jobs. That’s unfortunate, because a role is meant to be something transitory, something that is filled temporarily. It is meant to be worn like a costume and then passed on to the next wearer. When you turn a role into a job, you risk perverting it’s purpose. When you turn a role into a job, you make it very difficult for others to share it – it’s hard to swap back and forth. When you make a role into a job, people get surprisingly defensive about it. It becomes something that they identify with very closely. If you try and tell them that anybody can do it, they tend to get all fussy and upset. They start to try and protect their job with clever artifacts like certifications – they’ll do anything to make themselves unique enough to keep that job. It’s an identity trap.

Here is how I see this problem manifest itself with Scrum teams: You sell them on scrum and teach them how it works. Every team has a Scrum Master and a Product Owner. So what do they do? They run out and hire themselves some people to fill the jobs of Scrum Masters and Product Owners. They get their teams sprinting and start delivering quickly – hey, now they’re agile! Only they’re not really. You see, as you face the challenge and complexity of modern day business, the team often needs to change. That person you hired as the Scrum Master? You may be best served to swap that role with somebody else. Maybe a developer or QA on the team. The ability to move that role around to different actors could be very useful. But you can’t do that now because it’s no longer a role, it’s somebody’s job. And you can’t mess with their job without seriously upsetting somebody. The end result is that your organization effectively can’t change. You limit your agility.

The bottom line is that I believe that the roles in Scrum were never intended to be jobs. To make those roles into jobs risks limiting your agility.