Adam Lambert or Kanban… America Has Decided
Reading:
Adam Lambert or Kanban… America Has Decided
The American Idol season finale starts in a few minutes and I am really hoping that Adam Lambert pulls it out tonight…. that guy can really sing. Okay… okay… this is not a blog about American Idol… just had to get that off my chest. But in all seriousness… that is what I am planning to do with my evening… watch Adam Lambert take home the crown. Just had to pound out a quick post to get a few questions out into the blogosphere before the Adam Lambert victory party begins.
Okay… really…. no more Adam Lambert… I want to know what you guys think about all this Kanban stuff and how it fits with the rest of agile…
Over the past few days I have been dipping in an out of the Kanban boards and paying a little extra attention to what David Anderson is writing over at his site. David did a nice piece this week asking if Kanban is just a tool? The conversation has been great and I appreciate the time and energy everyone is putting into the discussion. I am learning a lot. That said, there was one particular quote in David’s last post I would like to discuss:
“So Kanban (pull) changes the underlying paradigm from project-centric to flow and value-stream centric.” – David Anderson
I’ll admit, I am still getting my head around all this Kanban stuff. If we are using Kanban within the sprint to put visibility around the flow of work within the iteration, I get it. If we decide the iteration is waste and decide to go to a total pull system… cool, I get it. If we are able to increment an existing application little bits at a time, I pretty much get it. Are we going too far saying that our industry is moving away from the project-centric paradigm? If that is the case, we need to at least establish context.
Are we working on incremental changes to an existing product? Okay, I get Kanban. Are we working on a support queue? Okay, I get Kanban. I can see how Kanban is more than a tool in that context… it embodies the values of Lean thinking… pull, flow, value, waste elimination, and continuous improvement. It becomes not just a way of measuring and limiting work in process but a way of thinking about the work itself. In that context, projects don’t make a whole lot of sense.
What I don’t get is how we are going to do pull and flow and only work on a single slice of the application at a time when we are doing a large scale system implementation. When we are building an application from the ground up and there is no foundational architecture. When it takes time to established a shared understanding of the overall product vision to converge on an acceptable technical solution. What about when we are working on a system of systems? What if some of those systems are external suppliers of systems that have to integrate with our system?
What about when we have incremental investment and have to be done at some given point in time? No release planning? No iteration planning? No product backlog? Have we moved past the triple constraints when our customer expects delivery at a certain time, at a certain cost, with some idea of what they are getting for their dollars? Does Kanban work here too… if so, is it the same Kanban we were talking about before?
I think that in some ways Scrum is hitting a wall because some of us think that Scrum is the answer to everything. Is Kanban the new answer to everything? Is it going to replace Scrum and XP or DSDM or AUP or Crystal? I can’t even imagine how we can have these conversations without understanding where and when is the best context to apply these principles. I feel like a broken record but there is a time and place for all these techniques.
There has to be a way to make use of this entire body of knowledge. However you slice it… it is up to us to know how to use all this stuff and choose the best practices for the situations we find ourselves in. Once we see things through that lens, it becomes less about who is right and who is wrong… and more about how to apply this stuff… and when to apply this stuff… and more about how we can move the industry forward the best we collectively know how.
Adam Lambert or Kanban… you decide?Addendum 5/21/2009: I was really bummed that Adam didn’t win… oh, well.
Comments (8)
Mike Bria
Absolutely. Sometimes, you need more long-term vision & planning, more "lock & load", more stability. Sometimes, you don't. Most times, somewhere in between.
I've had the luck to help people apply "agile" at various companies and with various projects – each with VASTLY different *contexts*.
To believe there is "one true way" is short-sighted, novice.
At my own company, an "XP shop", I suppose we internally do more of Kanban-like thing – no estimates, little long-term commitment, no official time-boxes or iterations (we actually release "every few days, whenever we have something cool"). Whatever's hot, we bum rush it and "pull" it through the pipeline. This works for us, in our context.
Do I teach this to all my clients? Absolutely not. Some get some bits of it, because it fits their context. Others get other bits, because it fits their context. Still yet others get close to none, cuz that's what their context wants.
A time. And a place. As they say, anyway.
So, in more important news – sorry, Adam lost!
Cheers,
MB
aaron
Kanban over American Idle, for sure. I think that if we can really get a handle on rolling wave planning, shouldn’t we be able to establish lead times at the level of objectives/goals? There’s some blogs out there about Kanban for the portfolio. I’ve also worked with teams on doing relative sizizing for projects within a portfolio and it seemed to work fairly well.
Project Management Tools & Techniques
What I don’t get is how we are going to do pull and flow and only work on a single slice of the application at a time when we are doing a large scale system implementation.What Kanban does here is actually pretty clever: it forces you to think in a modular way, understanding each slice in context of the finished product. Yeah, it adds overhead: if you need to do A, B, and C, and A is having problems, Kanban calls on you to make sure that the folks handling B and C know about it.
But if one group is having problems, and the rest aren’t, that is a problem. Kanban allows resources to flow to where they can have the most positive impact.
You’re right that it’s not a total solution. But it’s definitely a great framework for getting closer to one.
Mike Cottmeyer
Mike… great comment, thanks for sharing… I could not agree more. I just wish that folks would set the context for where and when things work before the discussion starts. I am sure I’ve been guilty of not taking my own advice on this one over the past few years… I want to be able to take advantage of Kanban without it having to be the next big thing…
Mike Cottmeyer
Aaron… I think the idea of a “flow” of smaller projects through the enterprise is a great idea. I talked about that a bit in my ‘Throttling the Agile Enterprise’ post. Maybe that helps break the habit of creating monolithic 18 month project deliveries and opens the door for Kanban at the enterprise level… and then Kanban at the team level.
Mike Cottmeyer
PM Tools & Techniques… thanks for the comment. You are talking about the flow of value across the enterprise. I really like that concept and am exploring some of those ideas now in a paper I am writing. Kanban inside the team… Kanban across the enterprise value stream. My only problem is that I don't believe the project metaphor is going to go away. Could project and MMF begin to converge????
Kevin E. Schlabach
Mike-
Good point, one which I was starting to come to myself as I’ve seen that last few weeks of discussions.
I’ve realized that I picked up Kanban due to the limitations of Scrum compared to how my company works. Reflecting on this, I also struggle to see Kanban working better than Scrum in my prior organization. I couldn’t put my finger on why until I read your proposal here. I definitely can confirm your ideas with my experiences.
Project Management Tools & Techniques
Mike (C): I expanded on my earlier comment in this post about the Kanban debate that’s been flaring recently. I think it clarifies a lot of your concerns: yes, Kanban helps you work on a single slice of a large application, if you apply it with people who can creatively analyze the big picture. It gives people the chance to contribute, which is a great idea to the extent that they’re ready.