Software Development Methodologies as Collective Fictions

The failure of methodologies to accurately model reality is a common source of cynicism amongst boots-on-the-ground developers. I can recall many conversations with teammates where we’d end a meeting and then chuckle in hushed tones with each other about how “wrong” we were doing Scrum and how clueless our managers were.

In his article My 20-Year Experience of Software Development Methodologies, Ian Miell makes the point that methodologies are necessary as “collective fictions”—a thought I personally find comforting.

As an industry, we tend to look back on Waterfall as hilariously naive and antiquated, but to Ian…

…the waterfall process was a ‘collective fiction’ that gave us enough stability and coherence to collaborate, get something out of the door, and get paid.

He goes on to discuss his experience with Rapid Application Development at a startup:

We were small enough not to need a collective fiction we had to name. Relationships and facts could be kept in our heads, and if you needed help, you literally called out to the room.

…

We got slightly bigger, and customers started asking us what our software methodology was. We guessed it wasn’t acceptable to say ‘we just write the code’.

…

Turns out there was this thing called ‘Rapid Application Development’ that emphasized prototyping. We told customers we did RAD, and they seemed happy, as it was A Thing. It sounded to me like ‘hacking’, but to be honest I’m not sure anyone among us really properly understood it or read up on it.
As a collective fiction it worked, because it kept customers off our backs while we wrote the software.

Ian then experienced Agile, with all its foibles:

The few that really had read up on it seemed incapable of actually dealing with the very real pressures we faced when delivering software to non-sprint-friendly customers, timescales, and blockers. So we carried on delivering software with our specs, and some sprinkling of agile terminology. Meetings were called ‘scrums’ now, but otherwise it felt very similar to what went on before.
As a collective fiction it worked, because it kept customers and project managers off our backs while we wrote the software.

As someone who’s been making software professionally for over a decade now (not as long as Ian!), I have arrived at much the same conclusion, that software development methodologies have value regardless of their literal reality. Organizing principles and shared terminology alone are worth adopting something...anything.

Accepting the inevitability of these collective fictions is like taking a weight off of one’s shoulders. It just saves so much angst.

Bosses like to give politically-correct names to things that their people are already doing, and that’s fine. The software industry’s current fixation on Agile concepts is a reflection of the way that a distributed workforce would inevitably do things. Ian explains:

If software methodologies didn’t exist we’d have to invent them, because how else would we work together effectively? You need these fictions in order to function at scale. It’s no coincidence that the Agile paradigm has such a quasi-religious hold over a workforce that is immensely fluid and mobile.

We’ll always be working under some “methodology” and that’s okay. Just keep making good software, and don’t worry about what people are calling it.