Ask HN: What cool development languages/tools changed your career?

I've used bootstrap in the past, but not anymore.

Tachyons is different, it doesn't give you components but single-purpose classes and a font-relative layouting system:

http://tachyons.io/docs/layout/spacing/

It takes a while getting used to the names, but after that it's just awesome that you don't have to think about padding, margin and font-size values anymore. Just slap "pa1" or "pa2" there and everything looks pretty.

I have the same problem, and I've yet to find an acceptable solution.

Another problem I've encountered: functional CSS (such as Tachyons) is only good for minimal layout. I was recently converting a mobile+tablet+desktop PSD design to CSS with very complicated reordering rules (this <div> is full height on the right on desktop, floated bottom of the page on tablet, etc.) and all my elements had 20+ classes attached to them.

So I went back and converted regular 60+ lines of CSS. Still, for some reason, working with Tachyons was for some reason faster and "easier" to reason about.

I think it's a step in the right direction, but we're not there yet.

you reach for sed/awk ;)

That's a problem for your text-editor. Use search and replace.

Also, if you're using something like React, you can share the style strings in variables.

Without a benchmark I remain unconvinced. PHP7 performance was shown to have doubled since 5.6 in most cases, including with frameworks like Laravel. Prior to PHP7 it wasn't noticeably slower than Python.

yeah defo, learning my second language was big step in my development

My experience is the opposite, interestingly. I used to know and use a wide range of languages. Being able to say "6 years of professional Scala" has opened more doors, and focusing on a single language has allowed me to learn techniques and algorithms rather than worry about the differences between languages so much.

It is not like studying a course in a organised way.. Luckily we switched to microservices architecture in my company and thus many stuff had to be rewritten. The old monolithic application was PHP and PHP was deprecated for new projects company wide. So I and my team picked GO, JAVA and Node.js for different microservices we are responsible for. (Mainly JAVA and GO, we deprecated then Node.js unofficially in our team because we did not like it at all.) So as you have a well defined project and you have to finish it in the language you are new with, you start to research and learn a lot. Tutorials, blogs etc. And when I had a problem and I could not solve the problem because of my unfamiliarity in the new stack, I started to read also after work and I started to apply what I learnt the next day. Every day I started to have less problem or finding solutions was taking less time because my involvement in that ecosystem increased. Specifically for JAVA, probably still I do not know many stuff, especially I don't have so much idea about the abandoned enterprise technologies, or legacy build tools like Ant etc. but still our applications have good quality and they are working well and my team did not receive any bad feedback from JAVA-expert fellows from the company, it means we adopted well the technology and caught the train.

I've to open a new paranthesis about Node.js. If you've never worked with Node.js before and you start in 2016 you're kinda f*cked up. The ecosystem is changing a lot and maybe 70% of the resources you can find over internet are already outdated. You start the day happy and with full batteries and you start to use some frameworks etc. The next day you realize that people already abandoned it. Beside all the disadvantages of Node.js, it's also so annoying and make the adoption process so hard. How such a simple language like Javascript could become so complicated?

GO is very cool. Relatively easy language and also very very powerful, it has a high quality community and the resources you can find are very clear. If you are trying to pick between GO and Node.js, go with GO. Your adoption will be much easier.

Or you could have tried Vaadin (no connection to the company other than as a satisfied/impressed user)

https://vaadin.com/home

Sticking to one language is like being a carpenter and only having a hammer in your toolbox.

Any advice on increasing rates? 90$ to 150$ is rather drastic.

If you spend extra time after work building personal projects, contributing to open source community etc. you'll not stuck in nothing. Don't let your workplace limit and dictate the technology that you know and use, feel free to develop yourself. Then it'll be very easy to jump between different workplaces until you find the best for you.

With the way the market is, unless you stay at one job forever, its actually hard to stay with the same language... the world changes, opportunities change, your interests change...you'd have to try very hard to stick to one language.

Now, on that topic, what's much harder than learning multiple languages, is learning multiple language paradigms. If you have a C++ background, learning C#, Java, Swift, JavaScript, whatever is pretty easy.

Learning Haskell, ML, Elm, F#, Clojure, etc...now that will make your mind bend and spin for a bit. Those aren't nearly as common in a work environment, but you can gain a lot from learning at least one.

This is kind of similar to how if you're an english speaker, learning another european language isn't nearly as interesting as learning a CJK.

Same story: C++ at school and VB then eventually C# at work. About 7 years into my career, I started feeling as though I had stagnated. Over the past 4 years, I took the time to learn Ruby (RoR), Scala, some Clojure, Nim, and now, some F#. Each of them has, in some way, made me a better C# programmer. I can't go out and interview for a position programming any of these languages, but they aren't blindspots for me either anymore.

I would suggest picking a language you have an interest in and a project you have an interest in and having a go at it. Then repeat the process with a new language and a new project. At the end, you will have some working software and an expanded perspective.

It's still evil - what if the input program doesn't halt, or allocates too much memory?

> You need to allow a user to run half a dozen functions, with a syntax.

You could do this with Javascript 'eval' too and I still wouldn't recommend it. Usually when this sort of requirement comes up the answer is actually visual programming, where the user just has to join blocks up and get's some sort of immediate real time feedback [1]. It's the only time VP is a good idea as the problem is in an extremely extremely constrained environment. Putting Lisp in it's place is really just an all round awful solution. Look at how the AutoCAD Lisp experience turned out [2].

[1] http://acegikmo.com/shaderforge/

[2] https://www.youtube.com/watch?v=LM1RMgIdgR8

    (eval (read (current-input-port)) 
    (let
      ((add (lambda args (apply + args))) 
        (null-environment))) 

Yes, the Plone CMS still runs on Zope2. https://plone.com

The ZODB (Zope Object Database) is also still going strong: http://www.zodb.org

Absolutely, though I wouldn't choose Zope per-se for a new project.

However, for the use cases where Zope shone (complex security requirements, extensible content types nested arbitrarily, etc.), there are a number of Zope-inspired frameworks that are reasonable choices today, such as:

http://www.pylonsproject.org/

http://www.substanced.net/

http://morepath.readthedocs.io/en/latest/

Yes, really. You can write Scala in a very Python-like style, but "_.foo" is much more concise than "lambda x: x.foo", and class declarations are much more concise too ("class Foo(x: Int, y: String)" rather than "class Foo(object): def __init__(self, x, y): self.x = x; self.y = y").

Compared to Java, the inherent safety level of the type system is very similar, but since it's much lighter weight you can do a lot more of it; "newtypes" are much more concise, and for/yield makes it possible to treat errors as values in a lightweight way rather than using the second parallel type system that is Java checked exceptions.

Opinions on Scala aside, it should be said that Python isn't the epitome of conciseness that a lot of people think it is. There are more concise languages out there and languages that are as concise as Python that end up offering more elsewhere.

We just switched some of our back-end code to Phoenix in our farm management web app (https://www.harvestprofit.com/).

Your username intrigues me....what are you working on?

I would almost always use Elixir these days, the advantages are just that good. I don't see anything python or ruby can give me in a server context that Elixir doesn't bring. If I need to go for speed or write a CLI tool, then I'd probably pick Go.

For frontend work, clojurescript. So Elixir+CLJS for almost all web-based apps (or, you know, react native).

There's no need to involve Go in that stack if you're building on Elixir. If you absolutely need interop with something else it'd be better to go with something more focused (either on speed, correctness or something else, as Go doesn't really focus on much of anything, except design-wise) and use it through ports or the like, from the BEAM.

Go as an addition to a BEAM app won't add anything as it mostly does nothing much better than the BEAM and what it does better will be better handled by other languages. Some people would attempt to make a case for it being able to spawn threads, etc., but its scheduling makes in inherently as incompatible with the BEAM as any other language, so it's not a reasonable point to make.

I think the borrow checker is actually pretty mind-blowing. In general, I'd say that the combination of modern language features with a real systems target (i.e. low-level, no GC) certainly has the potential to be game changing, in the sense of making it that much less scary to write things in a systems language.