Garbage in, garbage out
“If you put into the machine wrong figures, will the right answers come out?” The members of parliament asking such a question clearly don’t understand, thought Babbage. Charles Babbage was a mathematician, philosopher, inventor and mechanical engineer. He was the originator of the concept of a digital programmable computer. He knew that poor inputs led to poor outputs.
A hundred or so years later, William D. Mellin and his fellow US Army mathematicians working with computers had the same insight. Put nonsense into a computer programme and you’ll get nonsense out. ‘Garbage in, garbage out’, they called it.
Computers and programmes have changed a lot since then, but GIGO, as it’s known, persists. In fact, we recognise it outside of computers too. Garbage in, garbage out affects every process. Anything that takes an input, transforms it, and produces an output is affected by GIGO.
Cooking a meal, manufacturing a chair, calculating a budget, making a decision. All will give the wrong answers if you have the wrong inputs.
Why do we tend to focus on the process, then? Better recipes, faster production methods, more advanced spreadsheet formulas, specialist analysis tools. However efficient and effective the process, poor inputs lead to poor outputs. Better processes create garbage better, with less waste, but still garbage.
So, we also need to focus on the inputs. Fresher ingredients, more robust materials, better information. Better quality inputs create less garbage. In cooking food or making chairs the benefits are obvious. And so are the costs. Better inputs cost more. But is the same true for information?
Reading a good book costs the same as reading a bad book. Listening to a well-researched podcast takes the same time as listening to an uninformed opinion piece.
The problem with information is there’s no way to judge its quality until after we’ve consumed it. And by then, if its bad, its too late. Perhaps Mellin and the other mathematicians had the same problem. They only knew that their computers had produced garbage outputs after they had processed the garbage inputs. We face the same problem.
How do we improve the quality of the information we consume, how do we get better inputs so that we can get better outputs?
Perhaps only through active intentional learning so that we consider the usefulness of what we read and watch and listen to. Perhaps only by honing our ability to pre-judge similar things in the future. Perhaps only by questioning and critiquing our information inputs. Mellin’s computers couldn’t do that. Computers don’t care about their outputs. But we do.
It’s an unreliable, flawed and fallible way of improving the quality of what we consume. But it’s worth doing. We need good thinking processes, using good information inputs.
Good in, good out. GIGO.
Weeknotes 333
This week I did:
Traffic management
Our quarterly planning went really well. The next bottleneck to improve how our work flows is in how we move from uncertainty to (some) certainty about what the work involves. It’s going to be an interesting challenge as there are lots of variables, lots of different perspectives, and no single definition to be reached. Need to think about some experiments to try next quarter.
Humans out of the loop
This week’s Irregular Ideas was about generative AI and what it might do to human culture. “Culture builds on what went before, but when humans build culture they do it in messy, tangential, reactionary ways. When AI builds culture it optimises for efficiency, sameness and incremental change.”
Playing with GPT-3
I asked GPT-3 a few product management questions and got some answers.
I read/listened to this week:
What a Social Systems Perspective Teaches us About Change
I’ve been listening to The Liberators podcast. Lots of the episodes are really good as they are based on research which gives them more validity. This episode is about the social systems view of the company, and in particularly the idea that people don’t make the team, the team makes the people. I’m interested in how to create the right environment for the team, and this is an interesting theory to build on.
I thought about:
Single and double loop learning
I thought a bit about how we might use single and double loop learning to identify how to tackle problems. Single loop learning is about improving within the system, double loop learning is about changing the system. Perhaps even by simply asking ourselves the question of whether we should tackle problems from within or from outside.
Less like geese, more like sparrows
Teams might work better if they behaved like sparrows, which respond to each others behaviour to adjust where they’re going, than geese, which require a leader to follow and alignment.
Letting go
Had lots of opportunities to practice letting go this week. Not sure I’m doing very well.
How do you know if a product launch is successful?
A successful product launch can be measured by a variety of metrics, including customer feedback, sales figures, and customer retention. It can also be measured by the amount of media attention or social media engagement the launch has generated. Ultimately, a successful product launch should be judged by how much it contributes to the company’s bottom line.