Why doesn't everyone do XP?

Hi! Hello! Hey! I'm Nat Bennett, and this is Simpler Machines.

We're getting back on the regular posting schedule – this is officially the start of Season Three of this newsletter. This may change as I get my feet under me but expect to see these come out Friday AM, PST.

I wrapped up Pivots, my pop-up newsletter about Pivotal. If you missed it and you fancy reading ~15,000 words about what it was like to work at a strange software cult you can buy the back issues on Gumroad.

Do you read Lucidity? I know several of you are here because of a link in his most recent essay to "What was special about Pivotal?" (an essay you might which really ought to have been titled "The Island of Non-Biting Possums" since that's how people tend to refer to it.)

I'm always tickled when people who weren't at Pivotal are interested in it. When I write about Pivotal I always think I'm writing for other Pivots. We had this strange intense experience that ended in this weird and out-of-control way^[1] and so I'm writing to try to make sense of that, for myself and for the other people who were there with me.

And then all these other folks show up! And they're like, "woah, that place sounds wild – tell me more!" or "Oh! I did something similar but not for (as long | at that scale | that intensely) and I always wondered what that was like over there." (Or: "Ya'll sound insane, stay away from me.")

Which is great, don't get me wrong. I often feel like I can help folks who weren't there more than I can help folks who were there. It was a consulting company, right? So explaining "here's how to do the thing, here's the parts that were the most important, here's how to get from where you are to where you want to be" – it's part of the lore. There are Standard Answers.

If you were at Pivotal you already know all of that. (I mean, mostly. I find I can still be helpful talking to folks who were at Pivotal but were far away from the consulting parts. Or who were only ever in the consulting organization, but are now in long-term product organizations, and I can tell them about how we solved the problems they're now having for the first time on Cloud Foundry.)

If you were at Pivotal you have my problems.

I've been thinking a lot about what Pivots need. Back when I was at Pivotal I thought of my mission as "make more Pivots."^[2] I tried to pull back from that for a few years after Pivotal, tried to find other things to Do With My Life, but I keep finding myself circling back to it.

"Make the world more Pivot-y, by making more Pivots, and helping other Pivots make more Pivots."

Some of what Pivots need, I can do. "Make sense of this experience" and "figure out how to use what I learned at Pivotal in my career going forward." The writing that I do is useful.

But the thing that Pivots really need is more people and places who employ Pivots. Most of us don't really want to run our own companies! What Pivotal gave us was a place where we could show up, work the backlog, and go home. What we really need are people who both know how to manage Extreme Programming and solve distribution problems.

The question comes up sometimes, "If XP is so great, how come it hasn't taken over the world?

Why do most software businesses still manage software engineering teams by managing the individuals on the teams? If XP is so much better (especially when combined with Lean, Balanced Team, and User-Centered Design) shouldn't companies doing it run circles around the companies that aren't?"

My answer to that is basically:

• Distribution completely dominates product
• Nobody gives a hoot about profit

Companies live or die on their ability to repeatably find customers. Actually giving those customers what they need is important for a business mostly to the degree that it helps with the process of repeatable custom acqusition – customers stick around, and they refer new customers.

And the people actually running companies are not motivated primarily by business success. Business success is part of how they're getting what they actually want – status, freedom, power.

This is part of why you see so many more XP consulting companies that XP product companies. When "deliver a software system on-time and on-budget" is actually the thing that's directly building your company's repuation you're much more motivated to do XP (that is – to make scope flexible, in a way the customer is still happy about.)

XP is also, I think, basically incompatible with individual performance management. It collapses when you start to ask, "Who should get credit for this?" This makes it very hard to run a large XP organization in our culture.

Individual performance management is expected both by central services like Finance and HR and by the individuals doing the work. If you're a manager and you're not managing individual performance, the people you're managing will come to you and demand performance management! (They won't think that's what they're asking, but they are.)

More on this next week I think – I need to talk about Christopher Alexander to really explain what I'm talking about here – and for now I need to sign off.

Until then,
Nat