7 ms·
I'm a a typescript/java engineer. I would say my lifetime code stats are probably around 40% JS / 35% Typescript / 15% Java / 8% Python / 2% other. That is to s
by aerovistae 5y ago
I'm a a typescript/java engineer. I would say my lifetime code stats are probably around 40% JS / 35% Typescript / 15% Java / 8% Python / 2% other. That is to say, I've barely ever written any Ruby outside of the occasional little script. And yet it's secretly my favorite language.
I wish I wasn't afraid that demand for it would vanish over time, which makes me reluctant to invest skills in its ecosystem. It seems so small and uncommon today compared to others.
- mooreds 5y agoWhat would convince you? * top companies using it? (shopify, github, stripe) * number of stars on github for projects? https://github.com/rails/rails https://github.com/rails/rails has 50k stars * place in the Redmonk language rankings? https://redmonk.com/sogrady/2021/08/05/language-rankings-6-21/ https://redmonk.com/sogrady/2021/08/05/language-rankings-6-2... shows it at #9 * ability to get a good paying job? https://marketing.dice.com/pdf/2020/Dice_2020_Tech_Salary_Report.pdf https://marketing.dice.com/pdf/2020/Dice_2020_Tech_Salary_Re... has the ruby salary at 114k, more than typescript/javascript I am an active member of the ruby community so am definitely biased, but I'd love to know. Sorry if that's combative, but I always wonder how folks choose languages based on popularity. If there's anything folks can point at, or if it is more of a gut feel. (Unless it is javascript or C, which are always going to be around because of their huge deployed base.) Edit: added place in the redmonk ratings. Changed to be less combative.
- seiferteric 5y agoThe usual reason given is lack of libraries compared to python. I started a project a while ago with the idea that you could create a ruby library that would implement the C python extension API using the ruby C extension library... I got some basic stuff working, so you could "import" a basic python C extension in ruby (after recompiling it against my headers/library). Then you can use a python -> ruby transpiler (which does exist) to convert the python parts of the code to ruby and maybe have stuff "just work". The more I got into it though, it seemed pretty daunting to get something like numpy working... so I have stopped for now.
- axelthegerman 5y agoI'm not sure what Python library you couldn't find an equivalent for in rubygems. I never had a problem finding what I needed. (Maybe some ML or related fields?) Depending on the size of the library it might make sense to rewrite it. Using C extensions is more popular than Python or other languages
- mooreds 5y agoAh, good point. Domain does matter. But that didn't seem to be what the original poster was pointing to. Based on my ambient knowledge, I wouldn't want to write anything in the machine learning space in ruby. Or for that matter in any language other than python (or maybe scala/java, if you are using spark). Or are you suggesting that python has more/superior libraries compared to ruby across the board?
- chucke 5y agoWhy not just use numo, Ruby's own numpy?
- mekster 5y agoWhat you say means it's worth learning if you can land in the right company which is quite risky. Besides, unless a language has a strong point compared to others, it's a hard sell, like PHP is good on web, JS is known by everyone and ecosystem is huge and python is good for ML and scientific stuff but what's the selling point of ruby besides rails? With that said I do like the language itself (but not the typing system in 3) and would like to use if there's a place for it.
- mooreds 5y ago> what's the selling point of ruby? I'd say it is custom webapp development (with rails) and scripting on the command line (anything bash can do, ruby can do better). PHP is good on the web with the huge installed base of wordpress, but when it comes to custom web app development, ruby (well, rails, really) holds its own with laravel, symfony and the other MVC frameworks.
- l30n4da5 5y ago> scripting on the command line (anything bash can do, ruby can do better) completely agreed. on a linux system, most of my shell scripts are written in ruby at this point, simply because it is much better to work with than bash.
- girvo 5y agoAs someone with years (oh god, so many years) of experience in Laraval and Symfony, I think Rails is superior to both. They're all excellent mind you, I just think Rails is even nicer in usage.
- brigandish 5y ago> scripting on the command line (anything bash can do, ruby can do better) That's true but you could say that about zsh too, and any number of other languages. Usually, the important reason that shell programming is used is the lack of need to install anything to run a script, which makes it easy and portable, and hence, reliable. I can't write a script in Ruby and share it safe in the knowledge it will work on another system, especially if it relies on a gem or two. I'd much rather go with something that compiles and move a binary around if I'm dropping dash/bash/zsh. If I was forced to choose an interpreter to do that kind of work, perl would be much safer than Ruby, Python or Javascript.
- aerovistae 5y agoIt didn't come across as combative at all, no worries. I find it pretty convincing to be honest. If anything I think the biggest hold-up is that I can get jobs as a senior front-end engineer but if I looked for Ruby jobs I'd kind of be taking 2 steps back down in terms of what I can contribute. Frankly I'm not a fast learner, unfortunately. Although it's possible that I've convinced myself of that because I hate the Java ecosystem so much that my brain just doesn't want to absorb information about gradle and intellij configuration and java dependency management and so on.
- scrose 5y agoI went from being mostly frontend focused to working almost exclusively on backend and while I was able to get a generous pay bump, I found that I was generally much happier to just be working with Ruby/Rails almost full-time. Even though I was not as proficient right away. I guess what I’m trying to say is if you end up focusing on working only with what you’re proficient at, you may be missing out on other opportunities that you may end up enjoying more long-term.
- phaedrix 5y agoIt's not you, it's java.
- lawik 5y agoThe market is quite good for people who want to try a move or two. I'm not into Ruby but I see people hesitate similarly about my preferred lang and mostly think they should just take that leap.
- sunsetSamurai 5y agowhat's your favorite language?
- jambalaya 5y agoWe're hiring for Ruby/Rails. We're also hiring JS. I've used Ruby/Rails my entire career and don't plan on switching. Pay seems good. Definitely more JS jobs of course but I'm happy with what I'm doing and making. I bet you can pick it up in no time.
- Lapsa 5y agoI made a switch from .NET to Ruby. Initially it's surely a downgrade in terms of salary and whatnot. In a long term - just about everything seems bit more sane over here. Ruby is my fav, don't mind Javascript (although I find its quirks annoying), secret crush on C (you clog it up w/ billions of calculations and it just goes: "yeah, here you go...").
- cutler 5y agoHere in the UK most Ruby jobs pay a lot more than .Net. In fact .Net, for some some reason, is second only to PHP in terms of low-pay.
- jstummbillig 5y agoThere is no need to sway anyone because there is nothing to sway. Javascript is eating the web and interest in Ruby/Rails is declining, measurably: https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0505cl https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0... That is not to say that you can't get a job and you might win the market by doing something increasingly rare in relatively high demand. Or you might not. Or you might do ruby for a multitude of other reasons but relative popularity better not be one.
- pizza234 5y agoFun thing (or not, depending on how one sees it): the trends of "Ruby" and "Ruby on Rails" match almost exactly: - https://trends.google.com/trends/explore?date=all&q=%2Fm%2F06ff5 https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0... - https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0505cl https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0... But this is well known :)
- krainboltgreene 5y agothis is exactly how i would expect hackernews users to do analysis when the question is "should i get a ruby job"
- weatherlite 5y agoAs you said everyone and his sister knows javascript, JS devs are a commodity. I don't think anyone can expect to secure his future by picking up something like JS or Python since there are tens of millions of people learning it right now. So the popularity of the language means just that - everyone knows it. There are more jobs but way more people applying. Nor is Ruby a secure bet, nothing is really. Ruby is niche but it still has a very strong community, there is almost nothing web related Rails and the community haven't solved. I think people overrate the importance of community size. Does it matter if JS has 10 million devs or 20 million? Not really imo. Maybe even 10 is better than 20, less decision fatigue.
- BilalBudhani 5y agoJust to add another great resource * Job board for Ruby https://jobs.gorails.com/ https://jobs.gorails.com/
- axelthegerman 5y agoI understand job security and demand is important. But for me the most important has always been to enjoy myself as much as possible. That's also how I can do my best every day. If that means having a 10% lower salary or half the number of job postings I'm fine with it. If I really needed to find a non Ruby job I'm sure I still could, there's enough demand in general.
- filoeleven 5y agoThis is similar to my take, only I never ever want to work with Ruby, since it just rubs me the wrong way. This is a throwaway comment unless you care about the following question. What is the the fundamental difference between Ruby and Lisp enthusiasts? The easy answer is OOP vs FP, but as soon as you dive into the details the distinction becomes more difficult. A thread [vs] between Alan Kay and Rich Hickey is perhaps informative, but not illustrative. I know that I lean towards lisp, but I cannot articulate why lisp usage fits my mental model more readily than smalltalk/ruby does. Is the gulf between Lisp- and Smalltalk-inspired languages that meaningful? [vs] https://news.ycombinator.com/item?id=11945722 https://news.ycombinator.com/item?id=11945722
- vinceguidry 5y agoBefore embarking on my programming career, I went through HtDP while I still had time to do stuff like that. So I got a bit of grounding in Scheme. Regardless, when I discovered Ruby, I jumped on it and to this day I prefer to use it over any flavor of Lisp. In the end, I really think it comes down to syntax. I much prefer a normal, imperative kind of workflow. You lose a little bit of power and flexibility over Lisp but the friendliness of having verbs, nouns and diverse semantics wins out over what feels like a monotonous landscape. As you say, the underlying language semantics are largely similar, expressivity is just very similar. So the only thing left is syntax, and when you talk about syntax, you're talking about the most aesthetic part of the language, the part that's hardest to describe and reason about.
- tarellel 5y agoI've been doing Ruby development for about 12 years now. I can honestly say I've had more job offers this last year that ever before and the money is honestly pretty great. With the release of Ruby3, Rails7, and many people having JS fatigue it seems like the ruby community has had quite a bit of growth within the last year or two. (not HUGE growth like 10 years ago, but enough that it's still pretty noticeable throughout the community)
- lyime 5y agoI am always looking for senior ruby engineers.
- dghlsakjg 5y agoAs a ruby specialist it has never been hard to find a job including in positions that aren’t in ruby.
- heshiebee 5y agoHey, the company I work for is hiring engineers for a Ruby On Rails app. My info is in my bio.
- dcchambers 5y agoGive it a go. Ruby (and Rails) aren't going anywhere. There are far too many companies that are perfectly happy and successful with rails-based apps. Name a tech unicorn from the last 15 years, they probably have some ruby/rails code that is foundational to the company. Many have adopted microservices and newer languages like Go, but the core Ruby codebases aren't going away any time soon. Even outside of the rails ecosystem, ruby is a wonderful scripting language for writing small tools with. I prefer it to Python when I need something more powerful than bash/sh.
- pqdbr 5y agoI started a new company at the end of 2021 and I chose Ruby/Rails ... once again.
- thejosh 5y agoOver time = how many years? You can still find work in pretty much any language. Think COBOL is a dead language? :)
- exdsq 5y agoCOBOL 14 has introduced OO, it’s very much alive and developed!
- jnathsf 5y agoWe're hiring Ruby engineers. Check my bio to drop me a note.
- ornornor 5y agoI couldn’t find any job opening or description on cityinnovate.com — would be nice to know what you’re looking for, salary range, location etc.
- fishtoaster 5y agoI'll be honest, I'm worried too. I'm a fullstack dev, but I consider myself a decent rubyist. I've been writing it for about 10 years and have spoken at rubyconf a few times. All else being equal, I just like the language itself. The points others make here are true, there are a ton of ruby shops, and demand for someone with a couple years of rails experience is still quite high. But the issue is not where ruby is now, but where we see it in the long term. My predicted timeline is something like: 1. Right now, plenty of companies are still being founded on rails. The demand for ruby devs well outstrips the supply. I don't think it's the clear favorite anymore, though, even among trendy SF startups. 2. 5 years from now, far fewer companies will be being founded on ruby. It'll be rare, and mostly just when the founder happens to be a seasoned rubyist. 3. 10 years from now, the supply of rails devs will continue to trend upward, but the demand will have been trending downward for a few years and the good jobs will start to be harder to come by (relative to other tool chains) 4. 15 years from now, Ruby will still be going strong at large companies, but few or no early-stage startups will be using it. It will be seen as stagnant and enterprisy. 5. 20-30 years from now, Ruby will be kind of archaic. The demand for seasoned devs will still be there, but it'll be specialized roles to manage a legacy codebases that drives a lot of revenue and isn't worth replacing in a modern language. There will be few jobs, but also few senior rubyists so the jobs will pay well. So, honestly, you could pick up ruby today and make a whole career out of it. If you're 25 years old, it might be getting a little tough as you approach retirement, but otherwise you'd be fine. That said, a lot of people don't really want to work in a language that occupies stage 5. I know I don't want to be limited to only working on ancient codebases. And so I, like a lot of people, am hesitant to go all-in on ruby right now. It seems like it's a little too far down the hype curve to base the next 25 or so years of my career on. On the other hand, it's certainly possible I'm wrong here. One thesis is that JS fatigue will drive a new wave of developers to Rails. Rails has been betting on that with a bunch of features that, theoretically, allow you to skip a lot of what you'd otherwise need JS for. Personally, I'm unconvinced. I think the future lies in JS-native tech. Specifically Typescript and an all-in-one Rails-like tool like RedwoodJS or something similar. But then, few of us can really guess how these trends will turn out over the course of long decades. ¯\_(ツ)_/¯
- Lio 5y ago
- JamesSwift 5y agoWhen I was contracting full time, a _very_ large portion of my time was paid for in Rails work. Something like 90%. I market myself as a true generalist, basically a CTO-for-hire, and still the contracts I landed almost always involved Rails in some way.
- ryanmt 5y agoTo me that is sensible in that a strongly opinionated framework is conducive to hiring temporary talent into. Not to disagree with your conclusion just that it seems likely to bias in favor of ruby and rails.
- JamesSwift 5y agoThat wasnt it at all. It was a lack of available talent. So many rails devs have "moved on" and theres a real supply/demand problem in the market in terms of running software vs available devs.
- l30n4da5 5y ago> I've barely ever written any Ruby outside of the occasional little script. And yet it's secretly my favorite language. I'm in a similar boat, with the caveat that on linux, I use ruby as my main scripting language rather than bash, mostly because I find bash to to be a terrible language for anything remotely complex. Ruby has great integration for running shell commands, while also allowing you to write your scripts as an actual program, which is great. For me, the language I have written almost nothing in, while it is secretly my favorite, is Elixir/Erlang.
- tomc1985 5y ago10-year Rails dev here, there's still plenty of work to be found if you want it. (Full disclosure- I am currently not employed and working on a project not involving SWE.) Come on in the water's fine.
- fuzzythinker 5y agoUnlike other languages, ruby is mainly driven by one application - Rails. As newer web FWs come, it's hard to see Rails keep up. The trends for Rails does not look good for the past 5 years [1]. I am hopeful that natalie [2] will work. If it does, it will remove the biggest downside to ruby. [1] https://trends.google.com/trends/explore?date=today%205-y&q=ruby%20on%20rails https://trends.google.com/trends/explore?date=today%205-y&q=... [2] https://github.com/natalie-lang/natalie https://github.com/natalie-lang/natalie