2

To measure or not to measure or how to measure

I already answered a few questions about my work on this paper in a Cell Press Q & A but there is one more point that I want to share: Easier and less direct measurements that appear to save time and money often take longer and end up being more expensive than the best measurement one can do.

Science is fundamentally an empirical pursuit relying on measuring and observing nature. Thus the question of what to measure and how to measure it is central to research. This question was in many ways defining for my work on this paper since most measurements of respiration and fermentation rates are rather indirect and my observation happened to be utterly surprising for me. Before I did the experiments, I expected that even if respiration per cell changes during the first growth rate, it will gradually increase, not decrease. Similarly, the assumption that exponential cell growth at a constant rate indicates a steady-state was so deeply ingrained in my thinking that no indirect measurement could have changed this assumption for me.

More generally speaking, indirect measurements can hardly support a surprising discovery. Data incompatible with expectations bring both excitement from the possibility of a discovery and discomfort from the possibility of errors and artifacts. The extent to which we choose direct and reliable measurements determines which possibility dominates. As the availability of data increases, their informational content and reliability become increasingly important. Nowadays there is no dearth of data; there is a dearth of data with useful and reliable informational content.