After a few years (sort) of lull, my weekly working hours have started to spike. Just last 
week, I ended up working 36 hours straight and in office. It’s been a while when I did an all-nighter. While my reason to spend the night in office working was something very different, the night went very different. I was going several WTFs per minute but in the end, when there was dawn I had built a solution that I know will last for long and will make sure that some of my team members won’t be spending their nights trying to solve for the problem. It has been a while since I went that deep on a project, but it was fun – I didn’t even realize how time flew by.
We all have our own definition of perfection… next time we should probably stop and ask what is clients’ definition.
Over the weekend, I was thinking what could we have done differently to avoid this situation. In hind-sight I had 10 ideas of how this could have been avoided. The more fundamental message is not those 10 ideas or the fact that I can still work 36 hours non-stop. The fact is that we completely miss the essence of what is needed. This is just one of several project that I have been on where the focus is on getting “features delivered”. Clients wants us to get stuff done because there are business drivers – someone has committed a vision to someone else; there is paid media that needs certain features to be delivered. Hence, we pick up and run with it and deliver stuff.
Step back a minute and ask how do doctors work – when there are planned and un-planned surgeries. Do they cut corners, do they say “we need this get done for a lower price – so lets not spend time stitching the patient properly and save 1 hour of Operation Theater time and save some money for patient”. It costs what it costs – there is only one way to do it.
Then think where they really spend the meat of the time. I still recall when my wife was in labor with my second kid, the actual operating time was about 20 minutes. She was gone for almost 90 minutes but 70 minutes of that was prepping her for the surgery and then keeping her in post-op. thats more than 2/3rd of the time spent in making sure that 20 minutes that were critical were flawless and she is in no danger. They didn’t start until everything was perfect and she wasn’t back in room until they were sure she was stabilized. My kid was in Neonatal ICU for 48 hours – not because he was ill but because doctors wanted to make sure that he got the best care and was free of any infection in the hours when he was most vulnerable.
Only if we would spend 50% of time (not 70%) in before and after coding, we would have much better products. Yet, we choose to dive into implementations off of the bat, and we put things for later. We all have our own definition of perfection… next time we should probably stop and ask what is clients’ definition.
