Gossip-Driven Development
An honest expression for what motivates software efforts
The list of software development patterns called “X-Driven Development” is long. “X” can stand for Test, Acceptance Test, Behavior, Spec (although the latter is an antipattern). A couple of years ago, a then colleague jokingly mentioned “Jesus-Driven Development”: You sit in front of your computer and just pray that it will work.
I have a new entry to the list. I got the idea for it when listening to Paul Krugman’s “The Chips Are Down”. (Yes, I admit that I am about two months behind with catching up on all his videos.) The video/article is worth watching/reading as a whole.
Paul Krugman: The Chips Are Down
The Nobel laureate in economics had criticized the buzz around AI before. As his profession as an economist requires, he needs to separate rumors from facts and create educated guesses that can be proven wrong. And yes, from time to time he gets things wrong and then later admits this, as any serious scientist will do. So what’s new here?
On the surface, it is impressive that he repeats the exact same argument that has come from agile practitioners: If AI only improves one step of the overall process or increases the output, what is won?
AI lets you do much more, but how productive that is in terms of the ultimate goals of a business, let alone economic growth and quality of life is much more doubtful.
Of course, you do not need to be an agile practitioner or an economist to come to that conclusion. Nevertheless, it is refreshing to read it outside of a small professional circle.
However, it is the later part of the video where the insight hit me. He argues that companies keep themselves busy because their leaders believe that this is what the market expects. No business or customer success attached, it is only about activity!
For me, this is a new level of escalation. There have been plenty of hypes, fads and crazes that infected management thinking, all pushed by “if you do not do this, you will fall behind”. There has also been “fear of missing out” (FOMO) — “I do not understand why I should do it but it might turn out very bad to miss the train”. But this one is not even FOMO: People know that there is no return on investment (ROI) for their AI efforts (yet, if ever). They are fully aware and still pushing hard. The motivation, in plain old English, can be summarized as follows:
You do it because you fear that others will talk badly about you behind your back.
This is what I call Gossip-Driven Development. You act based on what others might say about you. You know it’s not true but you do not feel powerful enough to withstand. Of course, these social mechanics at the workplace are not limited to AI. This is just another example that AI is not revolutionary but only reenforcing the old — actually the long-outdated thinking.
Get Gunnar R. Fischer’s stories in your inbox
Join Medium for free to get updates from this writer.
It breaks everything usually preached in the context of Product:
- Your products get bloated with features of questionable value.
- It does not fit to any strategy or higher-level plan.
- It does not solve a real problem or need of the users.
But it is typical for going with the gossip: Doing things that are useless, are not a fit for you, cost you more than they give to you.
It is also interesting how this means giving up on several principles:
- “rational” behavior, open communication, insight-informed decisions, “we do what works”
- meritocracy, “the best idea wins”
- we do things differently than the others, innovation, “the market gives incentives to those who differ”
Instead, we are back to schoolyard mechanics: Do not stick out or you will be punished!
As much as I like to display idealism, I even found myself picturing a way of how to do this properly, without me getting angry or feeling disappointed. Actually, a very pragmatic approach is possible: Stay active, not passive. Be honest internally. Communicate clearly: “We do it because people expect it from us, not because we are convinced it works.” Under this condition, it is ok to reverse the problem — tool causality and start with the tool, looking for a problem to solve: “We need to show that we are investing enough to explore this tool.” It can make sense to invest reasonably into something where the belief of success is low. Keep the option open to be surprised!
About 20 years ago, a friend told me the story of when faking activity and artifacts made sense from a professional point of view. He was working for a company in the chemical industry. Some investors were visiting the facility. Unfortunately, the real production was very abstract and looked boring. Therefore, his manager and he agreed that he would set up something complicated and impressive-looking — “eye candy”. It did not cost a lot of effort and was enough to keep everybody happy.
If this is what people expect to see and what they need to see to trust you, give it to them! Only make sure that you have something real under the cover and that you are honest with yourself.
Gossip: Heavy Cross
