Ask HN: Will Ruby Get a Comeback?

I've been writing ruby since 2008. It's been the core language in my career since. I think that languages/frameworks are more or less popular based on your location. For example, Utah was an early adopter of Ruby/Rails and has become prevalent in many large companies here (Silicon Slopes). There were (probably not anymore) 4+ ruby groups that were within driving distance to me that ran each week.

Companies who choose Ruby/Rails I believe do so because they want fast iterations, agile development, and small dev teams. I think this also applies to Python/Django and PHP/Laravel as well.

I am in the same boat, but jumping back. Folks at the non ruby company all had fond memories of rails apps and missed the simplicity. The tangled mess of TypeScript was a disaster codebase with massive tech debt.

If OPs concern is only availability of cheap talent, there are only two choices: nodeJS and python. Which is what most code academies, the main source of cheap ready-to-grind talent comes from.

Ruby used to dominate this space; in fact, I think the whole online code academy surge was lead by rails, although someone more in the know might point out I'm wrong. But market forces have changed the situation and favour nodeJS and python. For some reasons. Both are easy to setup, easy to learn, interpreted, with an integrated REPL setup, integrated packaged management. Just like Ruby. Where they differ? They dominate a valuable niche.

* JS owns browsers, and that strength creates other potential value points, starting from "what if I can use the same code in server and client" all the way to "what if I can use the same code for web and native". * python owns ML, probably the most on-demand niche in the market now. It has the most complete ecosystem of ML libraries even your mother can use (R was competing until 3 years ago, I don't think it's close at this point). This opens up a lot of possibilities, such as "what if I can deploy my ML model along with my web service", or "what if I can use the ORM layer that I use for the web directly to feed data directly to my data analytics toolchain".

These questions make CTOs and Eng Managers everywhere very excited, as the thing they hate the most is "technology drift", i.e. betting on tech which eventually loses community mindshare and inevitably rots to the point it gets rewritten in PHP.

I love Ruby, but it doesn't own a niche. It's particularly strong in the web application space, mostly due to Rails, but so is PHP, so is Java, .Net, and probably even VisualBasic. It's not backed by any megacorp, so it can't even claim that fake sense of validation entitlement Google's go has.

Developers love it, and it has a lot of value and success stories going for it, specially considering how 80% of Y-Combinator most valuable companies are built on it. So it'll never disappear, and someone's always going to bet on it. But it'll never be 2011 again, so expectations of cheap labour have to be aligned with that.

Choose ruby because you like ruby. Choose ruby because you feel a creative release with it. Choose ruby because it's not just about delivering customer value, but having fun cranking out features that look fun, and your users might think so too. Github wasn't the first CVS server with a UI, but it became Open Source's social network because they used ruby.

And goddamn it, make more ruby libraries. Eat some of those niches. Make some ruby ML. Compile ruby to webassembly. Add value to ruby. If you add it, they will come.

I don't see any advantage over pascal vs C++. I already know C/C++, so I'll stick to that. But great to hear that you are also using a compiled language that is not typically used for web development.

You mentioned 2 essential things: 1. You say it's hard to find good RoR developers 2. Good RoR developers have become super expensive

Ad. 1. To have more Ruby or Rails developers on the job market, you need to train them. Ruby is not a good language to start with because it offers many ways to solve a given problem. Juniors prefer to hear "do it like this" and not "you can do it like this or one way or another". The responsibility lies with the companies, they should patiently train and encourage developers in Ruby because of its advantages. The Python Foundation does a good job because they always say that Python has the best web framework, has the best X and Y (although that's not true). But it works for juniors. In contrary, Rails is a good framework for beginners, as it allows you to hide a programmer's shortcomings through built-in "good practices". It allows you to quickly create a relatively safe advanced product, at the same time giving developers some time to make up for gaps in their skills.

Ad. 2. The developers you knew 10 years ago are now expensive because they gained a lot of experience and are in a different position than they were 10 years ago. New developers are not "good" yet, they have to gather this experience. It has nothing to do with language. Hire less good RoR developers in your startup and let them grow under the watchful eye of 1-2 experienced (expensive) RoR leaders.