I come form a long line of people (men mostly) who worked with their hands. I like to think I have my own mechanical inclinations, even if most of my work is done on a computer screen. One of the things I and my forebearers share is a love for toolboxes. Most things are there because they are always useful, tools like a screwdriver, a wrench, and a saw that we are always reaching for no matter what the project. Others are there because they were once necessary and may be important again. The 12″ ratchet extension or the orbital sander.
The same seems true for data science. There are the basic tools, namely programming fundamentals, SQL knowledge, Excel experience, and R understanding. Others are more specialized but still useful. It depends on ones particular line of work. My father the air conditioning technician always had his liquid coolant gauge and would consider that standard while my grandfather the welder wouldn’t have a need to even own one, but one was no less a craftsman than the other or skilled in their particular task.
So too the tools of data science. We each have our own tools. Some are more useful than others and we’re all at varying levels of competency in using them to get our jobs done. What’s important is that our respective toolboxes are always growing. Just like metal tools that aren’t taken care of, our skills and competencies rust with disuse, and easily break if taskedwith something they can’t handle.
And it’s never the size that matters but how we use it, especially how well we’re able to perfect the tools we have. I’d rather be highly skilled with a few tools than a hack with many tools. I have a feeling my father and grandfather would feel the same way.