Tuesday, January 24, 2006

Pregnant Patent Pause Passed

I just got notification that the first of a whole suite of patents that my team filed when I was working at Sprint on the Mobile Portal and Wireless Application Manager platform was approved last week - US Patent 6987987. The listed inventors are me, Martin Geddes (before the Telepocalypse their was hope that the rebels would deliver the platform for future profits), Daniel Vacanti (the guy who demoed the working prototype live in Las Vegas in October 2001) and Todd Conley (the man behind the curtain). This patent defines a viable micropayments (transaction and subscription model) for Internet and Web content. It’s very Web2.0 and yet it is four and half years old. I really believe that if Sprint had pursued this technology they’d be in the frame as a serious Web2.0 player with eBay, Yahoo!, Google and Microsoft but they didn’t and they’re not. I wonder what WAM would have done to their stock price five years down the road? [rhetorical] Now that the patent has been approved, it will be interesting to see if the do anything with it. Will it stimulate a WAM revival as they realize they have some Web2.0 intellectual property? Hmmm. When the current advertising driven Web2.0 expansion starts to reach its limit to growth, and the market turns to transactional and subscription models to supplement the go-to-market strategy for web services, who ya gonna call?

Monday, January 23, 2006

The Number

It’s 1998, Singapore, 4pm on a weekday afternoon. It’s hot - damned hot. The oxford cotton shirt is beginning to stick to my shoulders. Today, like any other day, I’m taking my afternoon coffee break. Jeff De Luca is standing with me on the South West corner of the Tiong Bahru Plaza building. We are watching the traffic going by on the Tiong Bahru Road heading west towards Red Hill MRT station as we gaze through the palm trees towards the HDB blocks of Buhkit Mehra. It’s starting to rain but we’re protected by the overhanging mass of the retail and cinema complex above our heads. We’re talking about The Number. How much would it take? What’s the exit point, so that we could live comfortably and give up this miserable excuse for a profession? The number is in the mid 7 figures. Neither of us have that number. The topic has come up before. We do this coffee thing casually. It is never scheduled. Perhaps once every two weeks, we find ourselves together having this discussion, same place, same weather, same view, same topic, always the same conclusion. Jeff thinks he might open a chip shop in a beach resort near Melbourne. I like the idea of running a mountain bike touring business and maybe a pub that dedicated to bikers and bike races back in the mountains of Scotland. To anyone like me who grew up on the West coast of Scotland it seems perfectly natural that an Italian immigrant should open a chip shop. [Years later I might come to realize that this is an unusual perspective. Many of Scotland’s Italian immigrants in the early 20th Century actually got on boats that they thought were going to America. When they arrived in Scotland they disembarked only to find they were on the wrong West coast. But they stayed and open restaurants, many selling fast food, fish and chips, others with ice cream parlors.]

The Number is something we all think about even if we never talk about it to our spouses, friends, family or colleagues. Lee Eisenberg argues in his new book of the same name, The Number, that we should all seek to understand our own number and the thinking behind its calculation. [Thanks to 800-CEO-READ for sending it to me.] Then Eisenberg observes that many (most) Americans are simply not saving enough to ever achieve their Number. He encouraged me to redo my own sums. There are wild assumptions built in but I used conservative historic data like house prices will continue to rise by 4% per annum on average, just as they have done for the past 40 years. The bottom line is that after my kids graduate college, I’d have to downsize the house to see the journey out with the standard of living my wife expects. The difference between selling the house or staying put in the home we raised our kids is a few thousand dollars per year in additional savings. It’s worth thinking about.

Eisenberg fills the second half of the book with imaginary case studies. He invents people and circumstances and analyzes what they need to do to realize their Number. As a social commentary on the state of American household economics, he calls these imaginary people - crash test dummies - because he is going to show how their life is on crash course and despite their fancy educations, powerful jobs and six figure salaries they face ruin in later life. They don’t have a handle on their Number. Eisenberg calls them crash test dummies in order to do his analysis work. We’d call them personas. It is amazing the power of the imagination we’ve all been provided with. While others in the Lean community talk about Genchi Genbutsu (“go to the source”) which can involve re-enacting the pain of the actual customer in order to find the facts e.g. by driving the same car down the same roads, it is amazing what is actually possible without leaving your desk. In MSF we’ve adopted the use of personas. It’s a bit of “getting back to our roots” in the Visual Studio group with Alan Cooper’s company famous for developing the concept of personas and Alan also responsible for creating Visual Basic - the first Visual tool from Microsoft. There has never been a good book written dedicated solely to the concept of developing good personas until now. The Persona Lifecycle by John Pruitt and Tamara Adlin plugs that hole and I’d highly recommend it. Not least because I wrote a sidebar on the use of Lifestyle Snapshots based on my earlier work published as uidesign.net. Personas enable us to imagine rich context and to derive powerful requirements in the form of usage scenarios based on the personas. No need for real people - crash test dummies will do.

Thursday, January 19, 2006

About This Naked Conversation We’ve Been Having

I want to talk about us. Where are we going with our conversation? Have you thought about that? Are we getting serious or is it just flirtation? No obligation. Casual. Naked. Interrupted. Fleeting. Just 15 minutes in the study stolen while no one else notices.

I guess my status as a blogerati has been confirmed. I’m invited to the launch party (Saturday night) for Robert Scoble‘s new book Naked Conversations. Apparently it is all about how corporations can get intimate with their audience by getting naked with their conversations via blogs. Even the book has been getting naked with its own blog.

Did you see me naked on Channel 9 last summer? In the raw so to speak. It wasn’t scripted. It was impromptu, in my office, with all my clutter around me. How many more of you will be getting naked on the web this year? Will it be in your next job description or terms and conditions of employment? Will we see job postings stating “must be willing to converse naked on the web?” “previous naked conversation experience preferred.”

Hey, it’s a frothy, bubble web2.0 world out there. When the bubbles burst and the froth dissipates don’t get caught blushing!

[So, do I introduce myself as David or Agile Management Blog? Hmmm. Will people recognize me with my clothes on?]

Wednesday, January 18, 2006

More on the Impact of Metrics

A few thoughts to continue from my post last Friday...

Note that the lowest cost per change request is achieved in the 2nd quarter of 2005 [Figure 1]. Increasing the headcount in the following quarter increased costs because the ratio of dev:test went from 4:2 to 5:3. It was necessary to spend the 2 headcount equally in dev and test to maintain the slack capacity in test and keep the constraint in dev where we’d been managing it. There is no such thing as “2/3rds of a person” and so the cost metric shows a rise.

Note that the throughput metric peaks in the following quarter at 56 [Figure 1]. This is sufficiently high to break the bottleneck and leads swiftly to the elimination of the backlog. The result is that the throughput falls back in the following quarter.

The shortest (optimal) lead time [Figure 3] is achieved in the most recent quarter with sub-optimal throughput and sub-optimal cost. The slack capacity is what delivers the best customer service. Running the system fully loaded - where supply meets demand - or at the lowest cost - where resources are fully loaded all the time - leads to sub-optimal behavior. The optimal values for each metric do not coincide in the same quarter. It doesn’t work that way. The optimal system performance is most likely to occur when individual parts of the system are working sub-optimally. We can read about this in academic texts on queuing theory. However, these charts are from a real department, in a very real Fortune 100 company. [To repeat the sub-heading for the web site: Management Science for Software Engineering - where theory is delivered in practice.]

Tuesday, January 17, 2006

How to Communicate with Me

It’s important to develop a good reputation for up-management - to show bosses that you can handle stuff on your own and only escalate matters when you truly need a more senior manager to help. It’s also important to show that you can take something to the right level for resolution. Don’t bother a GM when a Senior Director will do. Don’t bother a SVP when a plain old VP will do. But that’s a topic for another post.

I developed the following template while I worked for John Yuzdepski. He understood that service goes downward in management and he encouraged us to communicate to him, how he could be of service to us. I’ve used the template I developed for communicating with John as a way to train my staff to better up-manage. It’s important not to expect people to do this intuitively. Generally, their only up-management training came when as a child they learned how to manipulate parents to get what they wanted. Manipulation isn’t the result we’re after. Understanding the correct level to make decisions and how to ask for senior intervention, is what we are looking for. Here is the template…

To: John
From: David
Subject: An issue I’d like your help with

Background

A paragraph (or two) explaining the background. Focus on information the executive doesn’t already know.

Issue (or Problem)

Describe the precise issue or problem. There may be a web or linked issues or series of cause and effect that needs description but try to keep it brief and to the point.

Proposal (or enumerated options)

Describe how you’d like to solve the problem, or a set of alternatives. Describe the pros and cons to the proposal or set of options.

Action (or “What I’d like you to do for me today”)

Describe precisely how you would like the executive to help you. What action do you want them to take, or what decision are you asking them for. Make the request actionable, or explicitly state that you need the executive to escalate the matter and help them identify which level is appropriate.

When I first hand out this template at a team meeting, it is amazing to watch the faces of the team members. They don’t quite believe it is real. I then wait for the first one brave enough to actually try it. It can take weeks but eventually someone will try it. And then the relief on their face when they realize that it actually works. Gradually the word spreads. Using a formal template makes the sender think harder. It makes them realize that this is an official memo. It also cuts down on over-communication and noise in your inbox. Too often team members will copy you on every email - stuff you don’t need to read. Sometimes, something important will get missed. Seize such an opportunity to educate people that they have to tell you when they want your help - your not psychic! Introduce the template. Noise in your inbox will drop off.

[Warning: And finally, don’t try this on bosses who fundamentally don’t get the idea that the service model goes down. One guy I worked for would yell, “Stop telling me what to do!” You have been warned.]