In Part 1, I described my Sunday morning Cleaning the Bathroom problem and how I think about the tools I'm using, the way I use them, and why. In particular I talked about using a credit card as a scraper for the grotty build up around the sides of the bath and sink. On the particular Sunday that kicked this chain of thoughts off, I noticed myself picking the card up and using a corner of it vertically, rather than its edge horizontally, to remove some guff that was collecting around the base of a tap.
This is something I've been doing regularly for a while now but, before I got the scraper, it was a job I used an old toothbrush for. In Part 1 I recounted a number of conscious decisions around the way I keep the littlest room spic and span, but switching to use the card on the tap wasn't one I could recall making.
Observing myself taking a tool I'd specifically obtained for one purpose and using it for another put me in mind of this old saw:
When all you have is a hammer, everything looks like a nail.Until I looked it up1 just now I hadn't heard this saying called The Law of the Instrument nor come across the slightly subtler formulation that Wikipedia attributes to Maslow:
I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.Given the first of those two variants, it's easy to imagine that the universal application of the hammer is a mindless option - and we've all probably seen instances of that - but I think that, generally, tools are used in places where they are inappropriate or sub-optimal for a variety of reasons, and temptations, as Maslow would have it.
There are three explicit variables in play in this space: the problem, the tool, and the person using the tool. Here's one way I explored it, by considering the possible scenarios involving the tool and choice of tool, and trying to understand how the person might have got there:
I recognise the shape of the problem, and I have a tool that I know fits it
- ... but I use my favourite tool instead because I'm more familiar with it.
- ... but I use something else because of politics in the office, my boss, my colleagues, ...
- ... but I use something novel because I want to own this problem and be the expert in it.
- ... but I use something else to prevent an increase in our already large tool set.
- ... but I won't use it because of ethical or moral issues I have with the tool vendor.
- ... but I won't use it because of previous bad experiences with the tool, or others similar to it in some way.
- ... but the context changed since I last looked, and I didn't notice, so I'll continue to use the existing tool.
I recognise the shape of the problem, but I don't have a tool that I know fits it
- ... so I'll use the tool that I have invested heavily in anyway because sunk cost fallacy.
- ... so I'll use the tool I do have that looks closest.
- ... so I'll use the tool that I have in my hand right now because there's no context-switching cost.
- ... so I'll continue to use the tool I am using now, despite its flaws, because I believe there is some benefit.
- ... so I'll use a tool I do have because there's no time/budget/desire to look for others.
- ... so I'll use something I do have because I'm uncertain of my ability to choose a new tool well.
- ... so I'll continue to a tool I have because I'm worried about the cost of getting a new tool wrong.
- ... so I'll use whatever is to hand because I don't really care about doing a good job.
I don't recognise the shape of the problem
- ... so I'll try the tools I've got and see where they get me,
- ... or make a tool,
- ... or use no tool,
- ... or try break the problem down into pieces that can be attacked with tools I do know.
The latter class can be particularly galling because it contains two sub-cases:
- I don't recognise the shape of the problem, and - even if I did - I don't have a tool that fits it.
- I don't recognise the shape of the problem, but - if I did - I would find that I have a tool that fits it.
And much wailing and gnashing of teeth have been caused by the hindsight searchlight focusing on the second of those. Your wailing and gnashing of teeth, right? And the same is true of these scenarios:
I don't, or am not prepared to, recognise the existence of a problem
- ... so I make no decisions about tool use at all
- ... which means that I might stay as I am or unconsciously drift into some other behaviour.
I recognise that there is no problem
- ... but I have an agenda that I am pushing and so force tool use anyway.
- ... but I just love to try new things so I'll go ahead and use a tool for its own sake.
- ... but I'm putting off some other work so I'll do needless work here.
- ... but I haven't got enough to do so I'll try this tool out to look busy.
As I enumerate these cases, I begin to think that they apply not just to the person with just the hammer, but to all of us, every time we do or not use any tool for any task.
In using any tool at all you are making a decision - implicitly or explicitly. When you enter three commands into the shell to get something to run you are either accepting that you will not use a script to do it for you, and avoid the typos, being in the wrong directory and so on, or you are missing out on the knowledge that a script could help you, perhaps because you don't care to avoid that time being spent on typing and typos.
In choosing to use the same tool that you always use for editing a file you are missing out on the chance to learn that there is something better out there. But also avoiding paying the costs of learning that new thing. Do you do that consciously?
I started trying to map these kinds of observations back onto my own exploration of ways in which I could satisfy my bathroom mission. As I did this, I came to realise that I have mostly cast the problem recognition and tool choice as something that is done from a position of knowledge of the problem. But my own experience shows me that it's less clear-cut than that.
In this area, I love Weinberg's definition of a problem:
A problem is a difference between things as desired and things as perceived.I love it not least because it reminds me that there are multiple levers that can be pulled when confronted with a problem. If I am struggling with the shape of the problem I can change it, change my view of it, change my desires about what it should be. Opening out this way in turn reminds me that exploration of the space is an incredibly useful way to begin to understand which of those levers I can and/or should be pulling: perhaps if I can remove the things that look like nails, I can even put down my hammer.
Sometimes I find that I can learn more about the shape of the problem by using the tools I have and discovering their weaknesses. Sometimes I can only imagine a possible solution by attempting to resolve the problem the wrong way. If I do that tacitly, deliberately, then I'm here:
I recognise the shape of the problem, but I don't have a tool that I know fits it ... so I will explore the problem space with the tools I have, deliberately experimenting and hoping to learn more about the tools, the space, the problem, myself.And I might find that I'm then saying "aha, if only I had something which could ..." or "oh, so perhaps I don't really need ..."
But this means deliberately deciding to spend some of whatever budget is available for problem solving on the meta task of understanding the problem. Stated as baldly as this it seems obvious that someone with a problem might consider that, doesn't it? But how many times have you seen something else happen?
How many times have you seen only a tiny fraction of the capacity of some tool being exploited? For anything more complicated than a hammer, it's easy not to know that there are capabilities not being used. The Law of the Instrument can be applied within tools too. If I don't know that Word can do mail merge, I might find myself buying a new tool to do it, for example.
On the other hand, creative reuse can be a good way to get additional value from an existing tool. A hammer can be used for things other than hitting a nail - as a door stop, as a lever, to encourage some seized machinery to become separated, as a counterbalance, as a pendulum weight, as a goal post, and might be a sufficiently good substitute for the "proper" tool in any given situation, at any given time. And, in fact, imagining ways to reuse a tool can itself be a useful way to get creative juices flowing.
But contexts change - the problem might alter, views of it might alter, the available tools might alter. Being open to reconsidering decisions is important in getting a good outcome. Doing nothing, reconsidering nothing - that pretty much guarantees at best standing still or perhaps applying a solution to a problem that no longer exists or applying the wrong solution to the problem that does.
Tool use is inherent in software development and the kinds of choices I've described above are being made all the time for all sorts of reasons, including those that I've given. It was interesting to me, as I enumerated these thoughts, that although in my bathroom cleaning example I have no reason to be anything other than on the level - there are no bathroom politics in our house and the stakes are not high in any dimension - and despite doing my best to be as clear to myself about what I'm thinking and trying at any given time as I can, I still proceeded to make choices unconsciously, to not take account of useful evidence, and to continue with one line of exploration past the point at which its utility was exhausted.
In Part 3 I'll try and recast these thoughts in terms of some practical examples from the world of work rather than bathroom cleaning.