On learning new tools

It strikes me that it is better to sit and think about what you need to do, rather than endless try out new tools and hoping you end up with a use for them. I’m not saying that innovation is bad, or that you shouldn’t stretch yourself into some new spot once in a while, but you will become more valuable to others if you can identify what it is you are supposed to be doing and then use your existing skills to get that job done. Don’t go looking at galleries of tools and techniques in service of finding something to do. Trust me, I know from experience that it is a great way to spend a hell of a lot of time not getting anywhere in particular.

You might not get the buzz of doing something mundane in a new and exciting way, but you can perhaps do something that is needed in an optimised way. You can get the job and use it an exercise for working out what new skills you need. And yes, it might be a bit boring, but it won’t be boring once it has been done, because you won’t even have to think about it anymore.

New software and new paradigms are a way of shifting your own lassitude on to other systems (or worse yet, other people). You can blame your delays on the fact that other people don’t share their vision. But ninety percent of the time, you already have the tools and skills to get things done. The way to make things easier is to note down the pain points and work to smooth those. A whole new way of doing things is rarely the answer.

Ways in which things can be made ‘better’:

• make existing code run faster
• make sure your code is easily adaptable to other data sets in the future
• make your code more accurate
• write documentation (for your future self and to train others)
• producing additional outputs and/or more understandable/usable output
• expand existing code to more data sets by using loops and larger data sets
• refine your outputs by asking people who have to take action on the data what they are going to do with it

• Procrastination
• Work
• Software
• Twenty Three