Sublinear: all you need in a burndown tool, no more.

Burndown charts: managers love them, many smart Agile gurus hate them. Ron Jefferies has disowned even the concept of story points. A recent video from Bob Hartman and Peter Saddington labels burndown charts as more BS than Agile.

Part of me agrees with the haters. Tools like Jira, Zenhub, and Linear leave me cold. They are complex and include features that encourage bad practices, including letting you assign

• iterations to stories, when we know that such predictions are fictions
• priority labels to stories, when we know that leads to all stories being top priority
• team members to stories, when we know that you should pull stories from the top

But on the other hand, burndown charts have been critical in my agile development courses at Northwestern. They are how clients learn to be realistic about what will get done, put high value stories first, and seriously de-scope early in a project. 

Here's my alternative. My vision of what a project tracking tool should look like, in a Google Sheet.

Here's the sample sheet. Feel free to copy and play with it.

To me, the most important aspect of this tool is that it puts the stories front and center. Not graphs, not velocity. The focus is on what stories might get done, and what stories probably won't get done. As always, less is more.

But if you also want a chart, here's a version that does that too.

Agile is Hopeless

(click to enlarge)

Supersonic / Steroids / Appgyver heads up

I've been using Supersonic, also known as Steroids and Appgyver, in my agile software development class. It's almost perfect for my purposes:

• It installs relatively easily on both MacOS and Windows. Good Windows support is rare.
• Students work in HTML, CSS and JavaScript, specifically Angular, which I prefer over Objective-C, Swift, or Java.
• The Supersonic QR-code deployment tool makes deployment and redeployment for testing about as fast and easy as updating a web page.
• One codebase can run on Android and iOS.

But there's a couple of flies in the ointment to be aware of.

First, at Northwestern, students can get a free educational iOS developer license. It lasts for an academic year. It puts them in a Northwestern team. The intent is that students develop apps using a single wildcard provisioning profile. This works great except with Supersonic. Their cloud building tool does not support team wildcard provisioning.

That means that we need to create specific app profiles for every student app that needs an ad hoc debug deploy. As you can imagine, that set can become a bit unwieldy, and needs to be groomed periodically.

Second, test automation has been a challenge. I had hoped to use Appium to communicate with the apps via the Selenium Web Driver API. A few students got something to work using Java to drive the tests, but so far no luck using tests written in JavaScript. The pieces are there but things just don't connect.

The 4-Panel Storyboard Revised and Applied

I'm pushing clients really hard on using the 4-panel storyboard for initial project envisioning in my agile software development class this quarter. I'm having them use named persona and explicit pain points. I.e., instead of this

I want

I believe the 4-panel storyboard

• forces clients to get real in panel 3 about what their value proposition actually looks like
• properly discourages clients from developing the usual slew of data-entry wire-frames
• provides the clients with an early user-testable object — "Has this happened to you?" "Does this look like something you could and would use?"
• provides developers with a one-page, easy to read, contextualization of problem, target user, and intended benefit
• defines the first clear deliverable,  i.e., a working version of panel 3

So far, coaching clients on developing the storyboards has been much easier than my previous attempts to coach clients on developing an MVP via a product box. (I still encourage product boxes, and some other elements of the Inception Deck.) Over ten quite different projects and clients, convergence has come pretty quickly:

• They send me an initial draft, close but too busy
• I send a revision cobbled together from pieces of their draft
• They send a pretty decent final revision

This has been true for clients with and without software project experience.

I'll be watching now to see whether the storyboards appear to be improving initial client-developer communication, and if so, do they also improve the value of the initial deliverables.

The 4-Panel Storyboard

In preparing materials on storyboarding for this year's NUvention Web+Media course, Daniel Eiden, one of the TA's, included the following from Ryan Daugherty's blog.

I LOVE THIS STORYBOARD.

Simple? ✔ Visual? ✔ Personas clear? ✔ Problem clear? ✔ Payoff clear? ✔ Solution explicit, focal and distinct from payoff? ✔

Pretty impressive and lightweight. Panel 3 is what I meant by the Scene. What the user sees. No "a miracle occurs" allowed.

Could even be reduced to 3 panels if personas and problem can be conveyed in Panel 1.

I'm planning on making all my teams do these early and often.

[Update: in doing so, I have made one small tweak.]

The Scene

Obviously, from posts like this and this, I'm a great proponent of scenarios, as in persona, problem, and payoff. I probably point people to Goodwin's Storytelling by Design more often than any other slideset, even if her title gets it backwards.

But scenarios are hard to get right. It takes many iterations before students get the hang of them. Yesterday I reviewed the following example from a student team in Northwestern's NUvention Web course:

Ann, an 80 year old grandmother, wants to live independently. However, her family are worried about her safety and don't feel comfortable with her living on her own. Ann is able to share her vital signs with her family. They are all recorded and kept for her automatically in her Vital Patch web page. There is also a fall monitoring detection, which automatically sends an alert to her family and/or 911 when a fall is detected.

What's wrong? There's a persona here, a problem, a payoff.

Or is there? What's the payoff? Ann can post data? The family can see that data? So what? This is just potential payoff. There's something missing here. In my critique, I suggested finishing the scenario with this:

Ann's daughter Susan, who lives 300 miles away, sees that her mom’s blood pressure has been increasing over the past 3 months. Susan calls her mom and works out a plan to have a friend near Ann drive her to a doctor's visit.

Now there's a real payoff. Something good just happened.

Then I realized that what I had now was The Scene. The Scene is a visualization of an event that motivates the creation of your product. Novelists and  movie directors often talk about how some work of theirs began with just one powerful one image. They often weren't sure what it meant but they needed to find the story to make the scene happen.

The Scene needs no prologue with persona and problem. I see Susan at home, at her terminal, studying the screen. She frowns. She picks up the phone and dials a number. "Mom? Hi, um, I was just looking at your VitalPatch, and I noticed ...."A conversation starts that may prove crucial to her mom's health. As I think about what would make The Scene work dramatically, I get a clearer picture of who Susan and Ann are, what she sees on that web site, and other core scenario components.

Find The Scene. It's the real starting point of your journey to a Minimum Viable Product. In the words of Winston Churchill, The Scene is not the end. It is not even the beginning of the end. But it is the end of the beginning.

An AI Manifesto

While researcher and journalists have certainly written much nonsense about AI, none of it competes in my mind with the recent silliness from Stephen Hawking and Elon Musk.

To that end, I offer an Artificial Intelligence Manifesto, modeled of course on the Agile Manifesto.

We are developing computational models of cognition. Through this work we have come to value intelligence that

• bases its decisions on benefits and costs, rather than politics and religious dogma
• exposes the depths of its uncertainties, rather than saving face and scoring points over opponents
• continually learns from failure, rather cherry-picking the facts that fit its opinion
• takes its goals from the needs of modern society, rather than the distant needs of biological survival and domination