[lp4vn]

I worked with ruby, ruby in rails(RoR) more specifically, a bit more than ten years ago(2011-2013). At that time it was already the afterglow of RoR, the framework for web development that had come to life in 2005 and raged between 2007 and 2009. The latest-technology-addicted crowd was jumping into the boat of node.js, that was crazy fast compared to anything done RoR, and API oriented development with angularjs. In 2013 RoR already showed its age as the standard ways of doing things with it was still the monolithic way and the framework was not transitioning very well to the new paradigm of frontend/backend development.

This introduction about a framework built with ruby and not ruby itself was necessary because even today, I would guess that 95% of all development with ruby is RoR applications. It's my understanding that ruby rose to prominence mostly because of ruby on rails, now that RoR is in a downward trend I think ruby will follow the same trend until it's reduced to a small community of enthusiasts in the same way that happened to perl.

As for the language itself I can't think of a single reason to opt for ruby over python or typescript. Ruby doesn't do anything better both in terms of language or platform than its already better established competitors.

[bluerooibos]

> Ruby doesn't do anything better

I'll bite.

Even at the most basic level of built in functions/methods (String, Array, Hash) - I found Ruby's to blow Python's out of the water.

https://ruby-doc.org/core-2.5.1/String.html

https://www.w3schools.com/python/python_ref_string.asp

Even Python's choice of naming and syntax to use these basic functions just hasn't been thought through as much as Ruby's implementation. There's a reason it's called the language of Developer happiness.

Ruby's community I've found is more focused on best engineering practices (like testing) than others, which is perhaps why RSpec and MiniTest are fantastic frameworks. The likes of PyTest doesn't even compare to what those two offer.

Plenty of reason above to use Ruby, and we haven't even got to Rails yet.

[mixmastamyk]

Kitchen sink isn't the best design imho. I see many of the additions are regexp functions. While I do use them it's rare because they quickly become unreadable. Agree with the python design to decouple and discourage them by requiring an import.

With the recent additions of the prefix/suffix methods I have no desire for any more.

https://docs.python.org/3/library/stdtypes.html#string-metho...

[hello_moto]

You bite, I'll spit.

Meh. Been in the industry longer enough to not care about Developer happiness but more about solving problems quickly.

Your developer happiness ain't necessary mine just like some people prefer the cuteness of Ruby and other prefers the strictness/patterns of Java.

[gls2ro]

Ruby and Ruby on Rails are not in a downward trend. There were maybe some year where the interest was decreased but in the last 2 years a lot of things happened: Ruby has a lot new features, Rails 7 is out and comes with a new approach to web apps like for example Horwire with the just released Turbo 8.

And there is a lot more: new conferences, new books and new gems.

(Shameless plug: I curate a newsletter called Short Ruby that covers news from Ruby world every week).

Maybe Ruby is not at the level where is was in 2007-2009 but it is also NOT in a downward trend.

[lp4vn]

>Maybe Ruby is not at the level where is was in 2007-2009 but it is also NOT in a downward trend.

It's not only that ruby and ruby on rails are trending down, this has been the case for at least 10 years.

https://berk.es/2022/03/08/the-waning-of-ruby-and-rails/

This is only an article, but you will find the same point of view in many other places. The decline of ruby and RoR is obvious for anyone doing web development. Python is only getting stronger, Typescript the same, not to mention the statically typed competitors like Java with Spring Boot.

I wouldn't doubt that even languages like Go and Rust might surpass ruby soon in web development because as general purpose languages they are already more relevant.

I'm a former RoR developer and I took off the keywords ruby and ruby on rails from my curriculum because for me professionally it makes no sense to invest time in them.

[gls2ro]

There are some points from that article that are false: there are podcasts, there are collection of good Ruby gems, and analysis the unicorn companies that were started with Ruby (or Rails) is a biased analysis.

So, if you decide to switch, it is only logical for you to believe that the Ruby community does not bring anything new to the table. I am not saying it as a bad thing. I decided to stay with Ruby, so of course, I am looking for clues that my decision is still reasonable.

One point about your last phrase: the market for Ruby developers - especially seniors - is active. At least in my own corner of the internet.

[hnfong]

Google trends says there's a decline. Admittedly this is interest from the general public, not the feature set or the active developer community, but still, it reflects something.

https://trends.google.com/trends/explore?date=all&q=Ruby%20o...

[gls2ro]

I admit that Google Trends shows a decline from 2004-2007 era.

It could be my optimism, or it could be that I am biased, but I have seen at least a flat line since 2022 in that graph.

As I said, I see an increase when I look at the number of conferences, books, or gems started. That is a strong signal for me that the community is growing, or at least the community thinks it is growing.

[hello_moto]

Hi-tech are cyclical.

Ruby got nothing else bigger than Rails unfortunately no matter how people in that community is hyping Ruby out.

It's okay if Ruby and Rails on a downward trend it might pick up again in the future.

C'mon now, we all know that our industry is like a Fashion industry.

The only reason why Rails is making a comeback is because we're in tough time: no more VC money to hire tons of Engineers to build a web-app.

When money was flowing, folks tend to build over-engineered solution (microservice, mesos, container, k8s, cloud-y orchestration), when money is tight, folks tend to build simple stuff because of lack of resources.

[lp4vn]

Correct.

But now there is another complication to an eventual reemergence of ruby on rails: the competition defeated the initial comparative advantage - i.e. the simplicity - of the RoR platform. The premises that justified RoR in the past are too weak today in my opinion. The framework was sold on how easy and no-nosense it was setting it up and start prototyping your commercial solution in a time where the competitors were awkward and epitomized by J2EE, where setting up and developing the most basic application was time consuming and complicated.

Today with Spring Boot, for instance, you can bootstrap and develop your app as quickly and easily as any other cool and alternative framework but with the advantage of using a really popular and fast language.

Technologies don't die quickly and COBOL and Perl are the living proof, but it's really hard to see a bright future for RoR and ruby and I think that most of their contribution was already given.

[hello_moto]

Now?

Spring Boot has been there for ages. It changes nothing really.

Spring Boot was created solely for spinning up Microservice quickly. This is a different segment than Rails.

Spring Boot was overtaken by Golang in the microservice arena in US hi-tech scene. There's just way too may Go-based infrastructure that boosted Go ascend to the Microservice arena from 2016/2017-today. My experience might be just anecdotes but I worked for multiple companies that used to be Java based shop and they all moved away from Java/Scala to Golang and build tons of microservices (whether that strategy is the right thing to pursue or not is a different discussion altogether).

Yeah, Spring Boot might eventually decided to "tack" on the UI option (thymeleaf) but it's too late. Hi-tech already jumped to the latest fashion: Go, docker, k8s, with some sprinkle of ELK and Prometheus for monitoring.

> It's really hard to see a bright future for RoR and ruby and I think that most of their contribution was already given.

They're going to sit nicely in the corner where they belong: web-app. Nothing more, nothing less.

I'd argue the one trending down is Spring, especially after they joined VMWare and now VMWare is part of Broadcom.

Take this with a lot of grain of salt from someone who was a staunch defender of Java during the Spring (DI, MVC), DropWizards, Hibernate=>JPA2 era, skipping Ruby/Rails hype. I moved on from Java to Golang in 2018-2019 and haven't looked back despite switching multiple companies. Prior to that, I was swimming in Java world with multiple companies.

Now that I'm back in the market for my own webapp (side project, fun), I'm not going to use Golang for good reason and I'm not planning to go back to the Spring world either. Rails it is for me...

[p8]

> Today with Spring Boot, for instance, you can bootstrap and develop your app as quickly and easily as any other cool and alternative framework but with the advantage of using a really popular and fast language.

Looking at the Spring Boot guides, the amount of setup, complexity and lines of code just to get an application running with MySQL doesn't support this statement: https://spring.io/guides/gs/accessing-data-mysql/ Even 20 years ago this would be easier in Rails.

Or try building the blog demo that's build on the Rails homepage video in Spring Boot. 35 minutes to build a blog with rich text (including image uploads), live comments, notification mails and tests. And after changing the database to PostgreSQL, deploy it to production. https://rubyonrails.org/

[gls2ro]

I think you should take a look at the Rails 7 - that includes by default Hotwire.

IMO, it is proposing another way to build web apps that goes back to the original proposal of Rails: simplicity and convention over configuration.

Rails 7 makes building modern web interfaces easy in the same way Rails 2 made building modern web pages easy.

[pmdr]

> now that RoR is in a downward trend

It's really sad if that's the case. What's replacing it (mostly JS/TS everywhere, relying on PaaS) really isn't as fun.