I like to write about the software engineering stories that sit on the margins.

We've all been asked about the Strategy design pattern during interviews, but few of us know that the inspiration behind it didn't come from the software engineering field.

It came from an architect.

The only non-negotiable thing I do every day is write.

8-9 every morning, bebop jazz, good coffee

50% of what I write turns out to be pure slop, and I throw it out.

30% has potential and can turn into something good with heavy editing.

And 20% is actually good writing that I’m proud of.

Been a while. Here's a write-up.

fragmentsofcode.substack.com/p/the-intro-...

I need to start posting here more frequently

The term "clean code" leaves the impression that we're looking for an aesthetic quality in the code we write.

But it's always been about maintainability, not beauty.

Let's create something today.

You can read the first 25% of most non-fiction books and get the point.

After all this time I still mostly use 3 React hooks in 99% of cases - useState, useEffect, and useContext.

To reach for something else it has to be a super specific situation. Also, no performance optimizations unless I can measure an actual problem.

@alexanderkondov.bsky.social has some great books on React and Node.js and they’re 50% off right now with code BLACKFRIDAY. Check them out: alexanderkondov.gumroad.com

Took me 10 years to figure out that the challenge is not building the thing but running the thing in production for a long time.

When I inherit a legacy project I focus on two things:

1. Can I run it locally?
2. Can I deploy it?

If the answer is "yes" to both, every other problem becomes much easier to handle.

I’m not a 10x engineer. Not even close. Maybe above average on a good day.

I just want to build things.

I write simple code so I can maintain it. I structure things so I don’t have to remember where they are. I follow principles so I have solutions to common problems.

Early in my career I had the constant fear that I'll hit a problem that I won't ever be able to solve. But every time I hit one I managed to find a solution.

With time the fear that you don't know enough gets replaced with the confidence that there's a way around even the gnarliest of problems.

I did a review of blusky's react app structure last year. Don't know if it's still relevant but it was very interesting at the time. It was (and still is) the only mobx codebase I've looked at.

alexkondov.com/reading-code...

Obligatory hello world post.