2 ms·
I half agree. There is the notion that I actively don't like most of what is on the brown list. And I'm old enough to remember most of those things being new an
by jillesvangurp 6y ago
I half agree. There is the notion that I actively don't like most of what is on the brown list. And I'm old enough to remember most of those things being new and have used a few of them. Also, I've done maintenance on a lot of code over the years (and still do). Including my own code. If you work products rather than projects and stick around long enough, this happens. Some stuff I've written in one language, I eventually rewrote in a different language. E.g. a lot of my early Kotlin code replaced Java code I wrote before. Kotlin made doing that particularly easy. These days, I rarely deal with Java code bases as I default to Kotlin for the last three years.
I've also done Ruby, Php and declined to ever touch VBA with a ten foot stick even when it was brand new. Just no. Php I never really liked but I appreciate what it does. Ruby was kind of alright but a bit limited for the things I do with it. I've deleted quite bit of it over the years. Some of it written by me.
There's a pattern that the author overlooks: the green languages address issues with the brown languages. E.g. Rust provides meaningful memory safety and other zero cost abstractions relative to C++. That's why a lot Github projects exist that essentially are about building better versions of classic unix tools in Rust. Also see the discussions about allowing Rust in the Linux kernel.
Kotlin was an obvious attempt to improve on both Java and Scala. Scala purists may disagree with that of course. The key issues there were lack of expressiveness in Java and the relative verbosity of that language. The issue in Scala that Kotlin tries to address is the complexity of that language. Jetbrains actually considered using Scala as an alternative to Java and in the end decided to create Kotlin instead.
With Ruby, people loved meta programming but the lack of typing made it less suitable for larger code bases. I've seen enough messy code bases to agree with that. There are several languages that you could name that ended up with a lot of former ruby-ists that basically feature powerful type systems, similar levels of expressiveness, etc.
So there's a progression of languages and programmers. Of course you use a green field language for a greenfield project. Why wouldn't you? Brown field projects are more common in this industry and rewriting them tends to be costly. So, that tends to not be a thing. Particularly successful things last a long time.