Is Lemmy's code base "good"?
I know that Lemmy is open source and it can only get better from here on out, but I do wonder if any experts can weigh in whether the foundation is well written? Or are we building on top of 4 years worth of tech debt?
You are viewing a single comment
Is Lemmy less bad or more bad than a typical open source project?
From some comments I've read, it's at least in better shape than kbin? A few people expressed interest in helping with that project and then went running for the hills after reading through the code.
It’s probably not the only reason, but Rust is a much more attractive language/platform for devs to work with than PHP. (Source: https://survey.stackoverflow.co/2023/#section-admired-and-desired-programming-scripting-and-markup-languages)
It’s also more scalable, because it’s a compiled multi-threaded technology, while PHP is interpreted and mono-threaded.
Mother. Of. God. Did they really write Kbin in PHP?
I may be talking shit because I'm not a PHP coder, but the times I've seen it, it was a nightmare.
Which... makes sense. The creator of doesn't like coding.
So PHP it born out of a dislike of coding. In turn the documentation is all over the place and inconsistent.
Makes sense. And JavaScript is born out of a dislike of coders🤣
To be fair, PHP has slowly been getting it's shit together since PHP 7, and 8 seems to be in a reasonably great shape compared to the horrors of 5.6
It has become really solid over time.
But it will always be a mono-threaded and interpreted technology, and therefore never a good choice for a high-scale solution like a Fediverse application.
There are good PHP codes out there as well...
PHP is really old isn't it? I remember using phpBB forums some twenty years ago. They worked really well, but that's going pretty far back.
PHP is an interpreted language which is inherently slow compared to a compiled language, such as Rust which is very fast. Modern PHP isn’t so bad kinda but I’m guessing a guy who hates programming and decided to start a new PHP project in 2023 isn’t really optimizing anything. Also, you’ll never get anyone you help you write PHP because gross. It’s a dead language with a small community of masochists and maintainers of legacy projects.
It would be like if you saw someone building a fighter jet and thought “hey, I can do better!” and then started getting your paper mache out to start playing air plane designer.
It is, but it aged pretty well.
Devs wanted to store state in objects, so it became object-oriented. It also gained a really solid full-fledged web framework with a strong community, with Symfony; and some strong micro-frameworks like Laravel.
But it will always be interpreted and mono-threaded, and therefore never a good choice for high-scale solutions. Facebook has to invent a brand new language (Hack) and runtime (HHVM) that was close enough to PHP so they could spend millions spending their PHP codebase to it, in order to make it compiled and multi-threaded, and make it scale.
That’s PHP for ya
I read from one admin that a Lemmy instance is a lot easier to set up and maintain than a kbin instance. It's initially more complicated to set up and updates are just a super headache to deal with. That sounds like a showstopper. I mean kbin is not going to get too far if it's that difficult to run and maintain an instance, no matter how good or bad the code.
From a user perspective kbin has a really nice looking interface, though Lemmy has more features. I'd like to see kbin do well. It's younger than Lemmy so it's going to be behind, but hopefully the overhead in running an instance can be resolved.