In his Oral History, Nandan Nilekani shared his experience with programming at IIT Bombay.
These were the days when computers were new and slow. You don’t have the freedom to write bad code and fix it on the fly.
He had to submit the code to the computer centre and wait a day to see the result. Buggy code is a day gone waste.
As a result, programmers were careful with writing the code.
Thinking preceded doing.
A similar story is from the Physicist Richard Feynman’s childhood experience of fixing radios. He already made a name for repairing radios when he got an assignment to fix a more complicated radio. He looked at the radio and walked around, thinking. The radio owner was perplexed, “What are you doing walking around? Go fix it!”. Feynman told that he was thinking about the solution. And he eventually thought a simple solution to a complex looking problem. For which the radio owner exclaimed, “He fixes radios by thinking!”
I’ve also read this quote recently,
“First, solve the problem.
Then write the code.”
Simple principle, but we are not following it these days. Is it because doing has become easy? Technological advancement made doing easy. We can find the right solution by trial and error method easily today. And many are just doing that – Trial and error – and present the reasonable solution. There is less thinking on why this trial worked over others. That’s where the science of the solution lies. But it is overlooked.
Trial and Error method became Hit and Run.
Of course, thinking alone won’t help. There is a famous case study shared in “Atomic Habits”:
Two groups of budding photographers is given an assignment to submit a photograph – one group is asked to click only one photo; other group is allowed to click as many as they want. The second group performs better. This case study is used to promote doing over thinking. Second group did well because they did more, while first group thought more.
But I think the second group also thought more. After clicking each picture, they would think where they went wrong and learn from mistakes. That’s how they would improve and eventually click a killer picture. If just doing works, you can even train a monkey to click 100s of photos. That monkey would never learn though.
So, I think we should leverage our ability to do more trials today to think better.
Sai Krishna 