Pure-Impure Segregation Principle

[giscus[bot]]

Pure-Impure Segregation Principle

Two months ago I published an article detailing why I think that Unit Testing is Overrated, which seemed to resonate quite a lot with readers, prompting very involved and interesting discussions. And although most commenters mainly shared their personal…

https://tyrrrz.me/blog/pure-impure-segregation-principle

[ploeh]

Thank you for the compelling example about song recommendations. I know it's years later, but I finally managed to write a series of blog posts that engage with that problem in order to show alternative architectures: Alternative ways to design with functional programming.

To be clear: I'm not disputing what you write here, and I'd probably write it that way, too. Still, I found the example so inspiring that I wanted to riff on it.

[Tyrrrz]

Thanks a lot for saying that, Mark!

I remember mentioning the song recommendation problem on Twitter quite a long time ago, never considered that we'd come back to it after so many years. I appreciate that you highlight that this discourse is not about providing the "most correct" solution, but about finding the limits of a given approach (and ways to work around it).

Indeed, I was initially intrigued by the whole concept of purity in code thanks to your talks and articles from almost a decade ago. That led to a significant change in how I design software -- and eventually inspired this article. So I appreciate your work a lot :)

Will definitely check out your series of articles on this topic. I also find it amusing how that simple "song recommendation" example ended up becoming kind of a kata in and of itself, eh?

[Tyrrrz]

FYI, I've added a link to your series at the bottom of the "interleaved impurities" section

[ploeh]

I also find it amusing how that simple "song recommendation" example ended up becoming kind of a kata in and of itself, eh?

A bit more than a kata! Although I've been doing a lot of other things in the meantime, this article series has been almost two years in the making. It's no secret: For anyone who cares to look at the author date on the commit that added yesterday's article, they'll see that it was written in 2023 😅

[Tyrrrz]

It's no secret: For anyone who cares to look at the author date on the commit that added yesterday's article, they'll see that it was written in 2023 😅

I really admire your perseverance! I have some pieces that have been in progress for over a year, but only because I haven't really been writing them in most of that time 😅

[ploeh]

Well, I think the project lay dormant for most of a year, so that sounds comparable with your situation. I could imagine that you have more important things to look after than writing blog posts.