To begin with I’ll skate around the edges, even look for things to distract me! Once I’ve forced myself to get into the problem I won’t stop for lunch, and will keep going on and on. I’m much more likely to be “in the zone” several hours in rather than at the start.

I work mainly from home, so the arrival of my wife from her work always stops me getting daft, but when I was young, single and working in an office, if I didn’t have anything in particular planned for the evening I’d just keep going, sometimes past midnight, and right through the night a couple of times.

When you view it from the perspective of the insurer, insurance is about spreading risk. You don’t know if one particular car is going to crash in the next year, and motorists can’t afford to take the risk it will be theirs. If you’re insuring 100,000 cars then you’ve a pretty good idea that 10,000 will be in an accident and you can make money from that. The risk is spread across the whole portfolio, but from the point of view of an individual car owner the risk is transferred to the insurer.

Testing isn’t about spreading risk or about transferring risk either, so from my pedantic viewpoint it’s not a form of insurance.

The owners of applications are carrying the risk, just like uninsured motorists. The owners lack safety in numbers There’s just one application (at least there’s just one at a time) and there’s no insurer to transfer the risk to.

You can draw an analogy between insurers and application owners, however. The insurers underwrite risks based on their assessment of the risk. In the case of cars they’ve got a huge amount of detailed information about the population of cars and motorists, so it’s not economic, or necessary to inspect individual cars.

For large commercial risks the insurers use specialist underwriters to look at each separate risk before taking it on. In the case of engineering risks they employ highly qualified specialist engineers to inspect the risk. Then they know what they are taking on, and they will quote an appropriate premium, or decline the proposal.

Owners of applications likewise have to decide whether they are going to take on the risk by launching the application, or decline the risk by binning the application, or sending it back for remedial work.

If you’re looking for an insurance analogy for testers we’re the specialist engineers who inspect the risk to let the insurer know whether it’s going to be acceptable or a ticking time bomb.

Insurance companies wouldn’t dream of underwriting large risks without inspecting them, and cutting back on the inspections so you can close deals faster is comically unprofessional . Owners of applications do it all the time when the launch applications after skimping on the testing, however. Perhaps they need to think of themselves as insurers who’re recklessly taking on a risk they don’t understand. Then they might value us as inspection engineers a bit more highly.

I think it will always be a struggle to sell testing to people who have either little knowledge or interest in the subject. I think I’m prepared for that.

What I struggle with is that in the effort to clarify and define what testing is, some of the the language and terms that no-tester people would relate to become harder to use. Just an observation. I could be wrong.