• Question: when you find a error in your work how do you solve it?

    Asked by anon-201742 to Sophia, Meirin, George, Emily, Andy on 13 Mar 2019.
    • Photo: Sophia Pells

      Sophia Pells answered on 13 Mar 2019:

      Sometimes I’ll realise that I just made a typo somewhere in my code or something. Those ones are easy to fix but frustrating! I have meeting every other week with my research group where we all present what we’ve been looking at and any errors or problems we are having so we can see if we can help each other solve them. Sometimes I will have been stuck on something for a while and then as soon as I mention it to someone who has been in research for longer than me they will straight away say ‘oh, I’ve seen that before! You need to do this’. I hope I’m that wise one day.

    • Photo: George Fulton

      George Fulton answered on 13 Mar 2019:

      I always think its great to find an error in your work 🙂 If you didn’t find the error then you’d be publishing the wrong results and that would be worse in my opinion.

    • Photo: Andy Buckley

      Andy Buckley answered on 13 Mar 2019: last edited 14 Mar 2019 12:25 am

      That’s a great question. Usually we spot it because the error produces obviously nonsense results somewhere… and hopefully the nature of the crazy absurdity hints at where we went wrong.

      If it’s a programming thing, there are a bunch of standard ways to debug and fix code; maths is tougher and sometimes relies on two independent calculations to check the sanity. We do have certain checks of consistency, though — things like the units always being correct and consistent (or other, equivalent, tricks for unitless calculations), and we use a lot of symmetries, i.e. things that the calculation shouldn’t depend on, in the way that the length of a line doesn’t depend on what way you turn it.

      And of course, sometimes the answer is: tear your hair out for weeks until you finally find it! In this situation it can be useful to have more than one project at once, so you don’t go mad…

      This is for the situation where you realise that you have an error. A lot of the time you’re probably getting away with something… which is ok, until it’s not!

    • Photo: Meirin Oan Evans

      Meirin Oan Evans answered on 14 Mar 2019:

      First of all, it’s quite common that I find mistakes in my work!
      After trying lots of different things on my own, the scientists around me are always so helpful in solving my problems. Sometimes this will be my friends with the same level of expertise, since everyone is working on something different, so will encounter different problems at different times.
      Senior scientists have gotten to where they are since they’ve seen lots of problems and fixed them (with the help of others, of course). So I try tap into their expertise because there’s a high chance they’ve seen a similar problem to the one you’re encountering.
      Sometimes, my students teach me stuff!