It's also worth considering that kbin currently has around 600 stars on GitHub, Lemmy has 11k. Kbin is written in PHP, Lemmy in Rust. PHP is older and more mature as an ecosystem than Rust, but Rust is really popular. I've heard few people say nice things about PHP.
Take from all that what you will, but to me it says kbin will grow more slowly. Also if you use an app to browse you'd hardly know the difference between Lemmy or kbin anyway
Github stars is not a good metric, firstly because KBin is hosted on codeberg but mainly because a healthy project has lots of unique contributors and regular updates/enhancements.
KBin has 79 open Pull Requests, while Lemmy has 29. From a visual check PR's seem to be older than 2 weeks. Its hard to say one is "healthier" than the other, without scraping information into a spreadsheet.
Secondly Rust is new and has a lot of hype surrounding it, as a result you get a lot of people using it on random projects.
Languages have strengths and weaknesses and developer ecosystems build on the strengths.
For example if I was writing a web application with a database backend I would choose C#, Java or Node.js because there are loads of libraries, tools and frameworks to make it really easy.
Rust is gaining a lot of adoption by embedded system users (replacing C mostly). Lemmy is the only Rust based web server project I am aware of. Which means the level of work to do anything and to keep it updated falls on the Lemmy devs rather than spread out amongst a larger community.
Everyone loves to insult PHP but it has a niche in webservers and won't disappear anytime soon. KBin effort will thus be spent on KBin.
It's relevant in open source because folks like you who are proficient in PHP are less likely to contribute to the project because it's just not fun for you to work in that language.
PHP also seems to be really unpopular in general. I'm not insulting it as I've never used it but no one that has seems to have many nice things to say about it in my experience
Is there an app for kbin? I've been using Jerboa for lemmy and like it. Browsers and web pages don't feel right on mobile. But I do like the idea of kbin being connected with mastodon.
PHP has a lot of rough edges and somewhat archaic paradigms (OOP), and thus not liked by many, especially those who got used to more modern languages and its features.
Also PHP lacked type safety for a long time, which is almost only disliked by young inexperienced programmers, who think typing int instead of var is scary because their teacher said so.
Pure OOP is now archaic, but the principles are still used. More modern OOP languages incorporate functional programming concepts and are more of a hybrid. C# is probably the best example of a language with strong OOP and good functional support. Java is also heading that way slowly.
Functional programming is gaining steam again, but in its pure form, it's not as useful in many domains. A hybrid approach of take all the best ideas and use whatever fits best with the problem at hand is going to become the next paradigm, but I don't think it has a name.
"Pure paradigm" programming languages are really just toys for research and experimentation, IMO. I remember taking various courses on these sorts of things and, for example, the OOP prof would say "one of the cool things about pure OOP is that there are no loops or if statements. Now, here's how you go about faking a loop or if statement using pure OOP, because it turns out you really need that to accomplish anything."
In the real world you'll want to use whatever works best, which often means a language that contains a bunch of features from different paradigms merged together. Ideally in a rational and well-structured manner, but given how much usage Python gets that's clearly not a fundamental requirement.
You also have to balance the ability to get shit done makes a language more widely used, see perl. Eventually things also need to get modified, so languages that are easy to follow become popular, see perl losing popularity.
Archaic by the sense of some people really want to replace it with something new, which will be as groundbreaking as structured programming was. While functional programming has its merits, it's not the catch-all solution as its evangelists claim. It's more of a tech hype around a pretty useful paradigm, but unfortunately people want to use it like a Swiss-army knife, while it's more like a hammer or a screwdriver. I personally prefer multi-paradigm programming languages like D.