The Golden Rule of Experimental Science

A Golden Rule for Experimental Science: Do it quick: Then, do it right.

I just thought I would share with you a piece of wisdom I acquired around 25 years ago when I worked at Bristol University.

Over tea one day, Don Gugan expostulated what he called the ‘Golden Rule of Experimental Physics‘.  At the time I did not fully appreciate its simple perfection. But now I am utterly convinced of its embodied truth and I pass it on (sometimes repeatedly and tediously) in the manner of an ‘old person’. It is this:

Do it quick. Then, do it right.

The rule encapsulates the simple fact that when one is doing something for the first time, it is hard to anticipate all the things which might happen. Here are two examples:

Example#1 It is common for students in laboratory sessions to begin measurements, carefully noting down everything they anticipate might vary, only to find that that nothing changes!  Or they find that that something they hadn’t thought of changes so much that the experiment is a waste of time. Or they find that they have taken so much data in a ‘dull’ regime that they ran out of time to make measurements in the ‘interesting’ regime.

Example#2 The rule applies to professional scientists (such as myself) undertaking complex measurements. For example before making my recent measurement of the Boltzmann constant, we were able to borrow an apparatus (Thank you Laurent: we remember your kindness) and quickly cobble together an experiment.

Carrying out a ‘quick and dirty’ experiment gave us an idea of the relative sensitivity of a large number of factors that we previously did not have the experience to properly assess. And we discovered several things in the process, most importantly that we had massively underestimated of the amount of water vapour in our otherwise ultra-pure gas.

And reading the excellent ‘Thinking, fast and slow” by Daniel Kahnemann I realise that Don’s ‘Golden Rule‘ is merely a specific example of a wider rule of thumb. Kahnemann calls it seeking an ‘outside’ view when planning an activity, and he recommends it as a remedy for a cognitive bias towards optimism that he calls the ‘planning fallacy’.

This refers to the tendency when making a plan to consider all factors one knows about, but not to actively seek knowledge of other factors that one might not have considered. The Golden Rule embodies an experimental approach towards seeking knowledge of relevant unanticipated factors.

Another way is reading about other people’s experience, but this is hard for experimentalists. Mike Moldover from NIST is fond of reminding people of what he calls Dean Ripple’s guideline that:

Two weeks in the laboratory can easily save a whole afternoon in the library.

This quotation sums up perfectly the fact that ‘spending an afternoon in the library’ can feel harder than ‘getting on with the project’. Reading other people’s work is hard – and it takes time to work out the relevance of their experience to yours.

Happy experimenting. 🙂

UPDATE

Don Gugan denies being the original source of this quotation. He writes

“Not my formulation, though I believe it absolutely. First heard it from Cecil Reginald Burch (FRS and much else besides) in about 1960, and suggested it it as the guiding principle in the “Project Manual” when I was in charge of third year undergraduate projects.

CRB was an amazing fellow, Bob Chambers was/is a great admirer: they don’t make them like him any more.”

CR Burch is recorded as being a pioneer of optical design of microscopes and telescopes. And during this research he was frustrated at being unable to create a perfect enough vacuum to allow the evaporation of aluminium onto a surface to create a mirror.

So in furtherance of making better telescope mirrors, he discovered silicone oils which could be used to make better vacuum pumps. What a range of achievement!