Before we begin, Ordinary Analysis is totally free, and I’m happy to share it with you. If you want to receive every new post in your mailbox, you can subscribe by clicking on this cool button below.
Around 10% of Indians are trilingual. In most cases, it is English, Hindi and the native language. I belong to this category. My native language is Bengali. And it is not only my native language but also my first language.
I grew up in a household where 99.99% of conversations were in Bengali. Not only that, my medium of study was also Bengali until I joined college.
In college, the official medium was English. I didn’t have any problem writing papers or reading books. Some terms were different, for example, “Jaddo” (জাড্য) became inertia and so on. However, I was already familiar with most of these terminologies because I had studied most of them in higher secondary.
The part where I faced some challenges was speaking. I was forming correct sentences (because I knew the grammar, parts of speech etc.) but often they would be very slow. I figured out why this was happening. It was because I was converting Bengali to English in my mind before speaking.
So, I figured there are two ways I can improve: either I have to make these translations instantly or I have to start thinking in English. The second one seemed like a more reasonable path. During the same time, I got into Beatles and Rock music so my transition was rather easy. Over the last 9 years (I started college in 2014), English has become the “2nd 1st language” for me. Due to heavy usage, I often think in English too.
This is something I realise is happening with programming languages too.
I have mostly coded in R. Even to fetch data, in my previous organisation, we used to write very complex SQL queries in an R wrapper. SQL queries themselves have very long and rigid structures, so scripting languages like R (or even Python) make life easier.
But this is not the case in my current organisation.
Here, I need to look at SQL queries written by others. Sometimes debug them. Often make changes to pull different data. So far, on many occasions, I took help from Chat GPT and converted the query to an R code. Then debugged, and made changes.
Even when I have to do some data wrangling, I start thinking in terms of R functions, then google the equivalent Python code for that and proceed if I am writing a Python script for the same. R has basically become my Bengali. The good thing is, that the syntax and code flow for R (base) and Python are not very different, but SQL is a different sport altogether.
The challenge here is a bit different. The Beatles have never written an SQL query.
Thank you for reading Ordinary Analysis, a blog-newsletter where I write about different topics that intrigue me. If you enjoy this post and want to see more, you can subscribe by clicking on this cool button below. It’s free, easy and rewarding.
Alternatively, you can also share it with someone you think will like it.
A good read.. thanks
So true. Being an R programmer and user, I have faced the same in my day-to-day data science job. Thank you so much for sharing these wonderful insights.