DEV Community

CodeBase Journal
CodeBase Journal

Posted on • Originally published at Medium

Log 01: The Legacy Break | Switching to Docker for Data Science


In 2024, my setup was comfortable.

I had SQL Server Management Studio (SSMS) running on my laptop, primarily through Microsoft 365 Business Central. It was a stable, professional environment. It never caused me an issue, and for a long time, that was enough.

But as I move deeper into Data Science and Analysis, that comfort started to feel like a liability.

I’ll be honest: I was terrified. There’s a specific kind of anxiety that comes with trying to repurpose a "business tool" for a new, technical journey.

  • Will I actually utilize this properly?
  • Is this tool too rigid for data science?
  • Will it even work once I step away from the Business Central ecosystem?

The "Of Course It Failed" Phase

My fears weren't unfounded. It didn't work.

I can’t tell you the exact technical reason why the native installation hit a wall, and honestly, in the heat of the moment, I didn't care. I was just stuck at a crossroads: spend days debugging a legacy configuration or find a modern way forward.

I chose the forward path.

The Accidental Hero: Docker

I ended up turning to Docker—a tool I had completely ignored until now. It was always in that "I’ll learn that eventually" pile. But out of necessity, I pulled a sqlexpress2022 container, and suddenly, SSMS was back online.

It’s a strange realization when the tool you never paid mind to becomes the very thing that saves your workflow. Docker just... allowed me to work. It removed the friction I didn't even realize I was fighting.

Focus Over Curiosity

There is a massive urge right now to stop everything and do a deep dive into Docker. I want to know why it’s so famous and how it fixed a problem that a native install couldn't handle.

But I’m practicing discipline.

The Docker deep dive has earned its place in my journal, but that’s a story for another day. Right now, the mission is Data Analysis. I’m keeping my focus there, utilizing the containerized environment I’ve built to actually crunch the numbers. The "why" behind the container can wait; the insights in the data cannot.

This is Log 01. The lights are back on. Let’s get to work.

Top comments (0)