Software At the Coal Face

Software should be for the people who do the work

I’m not a technologist. I’m a software person, but not in the way people mean when they say that. I believe in the power of software to improve lives in big and small ways, and I have a hard time imagining doing much except through software.

I’m less fond of the software industry itself. I think it is too slow at solving problems that matter to people, too myopic to see how much it could really do.

I ran a software automation company for twelve years, and when I explained it to executives or salespeople, they would respond, “Oh, so you can fire sysadmins!” (Draw your own conclusions as to why these are the ones who consistently said this.)

In fact, our strategy was the exact opposite: I set out to turn sysadmins from tactical tools into strategic assets, to make the individuals more capable, and thus more valuable and harder to fire.

It was always easy to show my strategy was the right one: Let’s say I can produce software that can either reduce cost (fire some part of your team) or improve service quality (help that team do better work), which do you pick?

Nearly every time, people pick a higher service quality. “Wait, that’s an option?”, they would say. (The rare exceptions to this response usually came from organizations I did not want as customers.)

Software sucks. It really, really sucks. It’s so bad that when people look at their software investments, and especially the infrastructure layer my previous company lives at, they really only know how to talk about cost. Sure, they know the software is critical, they can say how many nines they have in uptime, or what their NPS score is, but they can’t measure quality in the way, say, a car maker or theme park would. Turns out, though, it’s easy as all get out to talk about cost. So, people naturally gravitate from the stuff that matters into the stuff they can measure. It always reminds me of an old joke:

A policeman sees a drunk man searching for something under a streetlight and asks what the drunk has lost. He says he lost his keys and they both look under the streetlight together. After a few minutes the policeman asks if he is sure he lost them here, and the drunk replies, no, and that he lost them in the park. The policeman asks why he is searching here, and the drunk replies, “this is where the light is”.

It might not matter, but it sure is easy.

You could argue that what Puppet did was weird, was special; that I found a seam of critical work that was undervalued, incorrectly measured, and ripe for both automation and an increase in human capital.

I don’t think sysadmins are that special.

There’s been this wicked, stupid, and brutal trend over the last few decades to devalue people. I don’t just mean, to replace people with automation; I mean the practice of reducing the value of individuals in service to financial strategy. It might be my age, but I assume this all started in the 1980s when corporations were compensated for having fewer costs on the books (which also led to idiocy like building an office tower, selling it, then leasing it back; some realtor makes bank, and your stock goes up because your balance sheet looks better according to the skewed spreadsheets of some hedge fund manager). The claim of many companies that their people are their greatest asset is contradicted by their behavior.

This becomes more true as companies get bigger. Most large corporations have spent decades finding ways to rely less on their people, and then use the results to demonstrate the people deserved it. Under-train them, then over-supervise them because they don’t have the skills. Don’t give them any power, then consider them easily replaceable because they can only do small things on their own. Refuse to invest in better tooling, then complain about their productivity.

I have a different perspective. You might might think I’m antiquated, or silly, or just naive. Meh.

My perspective is that people have always mattered, and that they always will.

My perspective is that there will always, forever, to be money made by making people more valuable.

Further, just as a nice bonus, I believe that the things that make people more valuable are generally congruent with the things that make them happier, which means that success for me helps the company and the individuals both financially and personally.

Don’t worry about people stealing an idea. If it’s original, you will have to ram it down their throats. — Howard H. Aiken

Considering that it’s this perspective that’s led to me to what success I’ve seen in life, and I’ve done enough work now to realize it’s also the only thing that I care about enough to put another decade into, it’s worth considering hiding this little gem, this little kernel of wisdom that others can’t seem to spot. I’m not worried. At the least, it goes against decades of corporate training: “People don’t matter, you should replace them, expect them to fail, they’re just cogs”. That kind of mental conditioning is hard to resist, especially when it’s imprinted on our financial system itself.

Beyond that, it just sounds too nice for others to see its value. Modern finance inherently distrusts anything that’s good for people, because they figure if you can do this thing and help people, you’ve got to be able to make more money by doing that thing and not helping people, right? Right?

No. In many cases, the thing that makes it valuable is exactly that it helps people. If anything, you could argue the entire software industry is about helping people spend more time on what makes them special, rather than helping people do something they could not do before.

Photoshop didn’t enable people to make drawings they could not have before; if anything, in many cases it’s less functional than using pens and pencils. It just enabled people to be far more efficient with the time they spent.

The difference between a great editor for coding and a bad one is not that you write better code with a great one, it’s that you can write good code that much faster and thus get a lot more high quality work done.

Often, the difference in speed is so great that the quality of work is actually dramatically better, but it’s a result of the user having far more time to spend on the problem, rather than the software itself directly improving the work.

It’s not a coincidence that boring, low-value work is easy to automate, or that people enjoy their jobs more when that work goes away. (Of course, this is assuming the job sticks around; in too many cases we use this as an excuse to fire someone rather than promote them.)

This was exactly Puppet’s goal.

Spend less time firefighting and doing menial work, and more time shipping great software.

I bet you can think of an industry or two whose people could stand to spend less time doing menial work, and more time on the things that matter.

I am not naïve enough think that every role a human plays today could be improved by better productivity tools; some roles become obsolete over time (not a lot of chimney sweeps these days), they really are just menial work or in an industry that cannot remain relevant. (Although, even in those cases I’d argue that both the people and the industries would be better off trying to find a way to work together on the transition, rather than the companies seeing the people as a sacrifice to the gods of productivity.) At the level of industry rather than individual roles, though, everything could be better through this kind of investment.

And in the end, I don’t have to believe every industry, every role is better off.

I just have to be able to find enough people, companies, and industries that are better off that I can keep doing great work.

And that’s what I plan on doing.