Over 10 years ago, I had this sort of a prediction that, with the massive adoption of a dynamic language like javascript on both client/server sides and test-driven development gaining a lot of ground, the future of programming would be dynamic and "feedback-driven". As in, you would immediately see the results of your code as you type, based on the tests you created. To naively simplify, imagine a split screen of your code editor and a console view showing relevant watch expressions from the code you're typing.
Instead what happened was the industry's focus shifted to type safety and smart compilers, and I followed along. I'm just not smart enough to question where the whole industry was heading. And my speck of imagination on how coding would have looked like in the future wasn't completely thought out. It was just that, a speck of imagination that occurred to me as I was debugging something tedious.
Now, most of the programming language world, seem to be focusing on smarter compilers. But is there some language or platform, that focus instead on a different kind of programming paradigm (not sort of OOP, FP paradigm, may be call it the programming workflow paradigm?). May be it comes with a really strong debugger tooling that's constantly giving you feedback on what your code is actually doing. Think REPL on steroids. I can imagine there would be challenges with parsing/evaluating incomplete code syntax and functions. So I guess, the whole compiler/translator side has to be thought out from the ground up as well.
Disclaimer: There's a good chance I simply don't know what I'm talking about because I'm no language designer or even close to understanding how programming languages and it's ecosystems are created. Just sharing some thoughts I had as a junior dev back in the day.
The main issue I have with this is the exclusive focus on happy path. The point of unit tests and especially mutation testing is to validate that every code path has a defined test and hence defined behavior. I wouldn't want to give that up, myself. I'm a huge believer in type safety and immutable objects and other things that prevent architectural misuse.
Just to share a perspective from erlang/elixir: pattern matching to filter for only happy-path inputs and the principle of “letting it fail” (when the inputs don’t match the expected shape) works really well in some paradigms (in this case, the actor model + OTP, erlang’s 9 9s of uptime, etc). In that kind of architecture you can really only care about the happy path, because the rest is malformed and can be thrown away without issue.
I'd imagine there would be no need to give up type safety, unit testing and all that though. I'm thinking more about language and tool creators' focus and efforts going mostly into compiler and type safety.
I've seen live updates for things like Go, NodeJS, Vue, Flutter, Angular, and you can recompile a Java class during debugging in Java (with limitations as certain things like static variables and methods need a full restart). More support and integration is certainly better but it doesn't feel like it's in a bad place to me.
I wouldn't say it's in a bad place either. Most enterprise grade technologies already have great debugging tools. Sure, those hot reloads, live updates are nice for UI development. But, I was thinking more of something built from the ground up to be, well, "feedback driven" in general. Most new stuffs that came out in the last decade touted their compiler as a killer feature first and rest of the tools are only developed as the ecosystem mature. May be that's just the best way to go about creating new successful language ecosystems, I don't know. Sorry if it feels like I'm being vague about the specifics. That's because I really only have vague ideas about whole the whole thing would work.
This is what Smalltalk is all about, and it has been like that since it’s origin: you basically program in the debugger, you program running, you change something, you proceed the debugger, etc. That’s why technics like TDD, refactoring, and others were developed in Smalltalk and just later translated to other languages (and always lacking, since no one reproduces the live programming experience 100%). As the time passed, attention has moved to other languages and most people not ignores what it was to program like that. But there are still some implementations around: I work with Pharo (https://pharo.org), and I can to say is all what you ask for in this post :)
It's always fun when someone comes up with a new idea of how to write code and it is something smalltalk did in the 80s.
Compose and flutter both have "realtime" previews? And I believe react has something similar? They aren't truely realtime, because usually there is a compile/execute step happening, but they can be fairly good for rapid feedback?
https://docs.flutter.dev/tools/hot-reload
https://blaszcz.uk/live-reload-with-compose-web/
For unit tests, there is https://infinitest.github.io/, which automatically updates your tests as you code. Never used it though.
infinitest looks interesting though. Will check it out, though I'm not much of a Java dev.
I am familiar with hot reloads. What I had in mind was something more fine-grained, not just the UI. A simple example would be that I declared a function signature. Then I write a test. As I start to implement the function, there would be constant feedback visible based on the inputs to the functions from test I wrote. If I declare a variable 'x' by adding function params 'let x = y + z', the feedback view would show a watch expression of x based on the test's input. If I changed it to 'let x = y * z', the watch expression would immediately change. I would be constantly seeing the result of my actions. May be this is asking for too much with the current technology we have. I don't know.
Okay, sounds more like infinitest, but sadly that is Java only.
I've never really played with it, but jupyter might also be closish to your vision?
https://jupyter.org/
Let's you do python stuff and see results of your code near realtime I think?
I've looked into Elixir livebook that's probably inspired by jupyter before. Yeah, something like that but for a much more general use case.
Huh... I've used Guard and TDD to do this in Ruby, works great. Yes you're testing the happy path, but it is easy to define negative tests as well.
I think of this as interactive development, or repl-driven development. You can work this way today in Clojure (frontend, backend, and lately even for scripting via babashka), and with lisps in general - the syntax lends itself to sending expressions to the repl and returning values to your editor.
It’s really the best way (my favorite, at least) to program that i’ve found for exactly the reasons you mentioned - it’s excellent for debugging and ensuring the behavior of small functions with minimal overhead.
Types are frustrating because they lock things up and they don’t guarantee behavior, which is really all a program cares about. I feel similarly about unit tests… it’s extra code locking up your behaviors, so make sure they’re what you actually want! A general problem with types is that you have to commit to some shape early, which can lead to premature design and basically some arbitrary DSL when you just needed a couple functions/transformations. Feels like the problem of OO at times.
On the other side, the trouble (beyond people generally not wanting to read/learn lisps, which is unfortunate) is that repl-driven dev requires that you take care of your tools, which means there’s a tough learning curve and then some maintenance cost for whatever editor you want to use.
At a career-level scale, in my opinion, the investment is well worth it, but it’s a tough thing to figure out early in your career. I expect most devs with a couple years of js/python see types and feel like it’s a huge relief, which is real, and maybe types make sense at a certain team size…
I think people should spend time in several different languages and paradigms - it makes the ones you go back to make more sense :D
I think elixir/erlang is also in the same class of languages as clojure in that sense. A lot of lisp-like languages tend to go into that trend, I guess. I love working in it.
May be my headspace was a bit too much in systems that benefit from rapid prototyping. Other class of systems might benefit greatly from type safety and unit tests. Even though, I still felt a bit iffy about unit tests and almost ideological spouting points of it. I struggled with unit testing for a few years and now I just use them for automation of bigger picture behaviour testing. Call them integration tests or whatever.
Something like this? https://codepen.io/darrengriffith/pen/KpKxqR
I can edit the code and see the changes without rewhatevering.
