It's hard to make sustainable software development practices stick. The larger an organization is, the harder it seems to be.
Why?
There are many possible answers and probably all of them have at least a little validity.
- It's way easier to learn a computer language than it is to learn how to properly attend to design.
- Refactoring (true refactoring) is hard to learn, too.
- TDD requires significant effort to figure out how to apply meaningfully.
- BDD requires organizational changes to make its biggest impacts.
- Larger organizations have politics.
- Larger organizations have communications and handoffs.
- Larger organizations have often deadlines that aren't linked with capacity or value.
All of those are at least somewhat true, some of the time, but none of them smells very much like a root cause. Maybe there isn't a single cause and that's why it's hard to pin one down. In fact, that's probably true, but I still feel like the above list (and its ilk) is a list of symptoms of some other problem, not a list of first-class problems.
Yesterday's blog entry helped me make a little progress in the form of a new hypothesis we can at least invalidate.
As I said, I automate everything because I know I can't afford to do otherwise. In fact, that's the reason why I apply modern software design principles. It's why I refactor. It's why I have a test-driven mindset and why I write tests first whenever I can. I'm on a shoestring budget and I know I can't spare anything on wasted effort, so I work as efficiently as I can.
What if the reason why larger organizations ostensibly tend to struggle with code quality, design patterns, refactoring, TDD, agile workflow, and lean product definition is as simple as the inverse statement? I know I don't have the budget to work inefficiently, so I work efficiently. Larger organizations have the budget to work inefficiently, so they don't work efficiently?
It sounds crazy. At least, it would have to my younger self.
"People are working in an unsustainable way just because they think they can afford it? What?!?" That's what young Max would have said. Today Max really only has a resigned shrug to offer in dissent.
So, because Old Max put up such feeble resistance, let's explore the idea a little more.
A small organization is usually comprised of* a tight-knit group of individuals. While they may not all be experts in every area of work, the impact of any person's decisions can be plainly seen in day to day activities. This means that the costs of bad decisions are not abstracted debts to be paid someday. They are real problems being experienced at the moment.
Pair that with the tight budget that smaller companies usually have, and you get a recipe for action: the plainness of the problem helps people know what should be done and the necessities of a small budget provide the impetus to do it.
Contrast that with a large organization.
In a large organization, consequences or often far-removed from actions. If you create a problem, it may be weeks, months, or even years before you have to pay the cost. That's if it doesn't become someone else's problem altogether, first. Fixing a systemic problem such as, say, not being test-driven can be imagined as this high-minded "nice to have" that probably won't really work because you'd need everyone to be on board, which will never happen because everyone else feels the same way.
At the same time, pockets are often quite deep in large organizations. While a show may be made of pinching pennies in the form of (for instance) discount soap in the bathrooms, they tend to spend a lot on wasted effort. They are able to do this because they have some array of already-successful and (typically) highly profitable products they can exploit to fund new efforts. Furthermore, in addition to being very large, corporate budgets seem like they are usually very complex. Large costs can sometimes seem smaller than they really are because they are impacting many different line items.
Pair those two together and you get fatalistic ennui making everything seem academic with a budgeting apparatus that consistently says "we've got bigger fish to fry".
I'm pretty sure this is one piece of the puzzle but I think there's more to it. For instance, there are many small organizations with shoestring budgets that still make bad decisions about sustainability. There are also counterexamples in the form of large companies that tend to make decisions that are good, or at least better than those of their competitors.
However, this writing is now quite long. So I'm going to end it, here, and discuss another factor in tomorrow's entry.
*: That is one of the correct usages of "to comprise". Look it up.
