loading...

Technologies of future past

rhymes profile image rhymes ・1 min read

These tweets made me giggle a little bit:

and a variation

then in 2008

The chain started because of this serious tweet:

Who knows if GraphQL, WebAssembly and WebComponents will be relevant in 10 years or won't be.

I think the real lesson from the first three tweets is twofold:

  • don't be too attached to a technology (most of us, me included, have a bias when it comes to our favorite piece of tech)

  • "always be learning" (as in be ready to move on when the time comes, and be curious)

Question time

Which three technologies would you bet on in the next 10 years?

Which three technologies you think/wish won't be relevant anymore in the next 10 years?

Discussion

markdown guide
 

You’ll be safe for the next twenty years with:

  • JavaScript
  • HTML
  • CSS
 

Any 3 things you hope or think will disappear?

 

I am very afraid that this can turn into a technology war 😅. We all know how developers are passionate when it comes to preferred tech.
But I will join the play. My 3 bets for the future are:

  • Rust
  • Web Assembly
  • Git

As for what I think/wish would be dead:

  • Java (I don't see it dying, just wish it could to)
  • Pure functional PL like Haskell
  • C
 
 

No hate at all. I really love the FP principles, but i think technology will move towards something hybrid, not purely functional

Ah well, that's fair, considering many popular 'functional' languages are actually just functional style (i.e. not pure). Hell, that includes Haskell without extensions.

I predict the opposite though: we are currently rather hybrid and the future will be pure, simply because making effects explicit makes them easier to deal with by automated systems.

Admittedly, this argument has been given since the invention of FP, but I think it had some catching up to do for practical use and we are just now at the turning point (and have passed it for some cases).

I don't think the future will be (modern) Haskell though.

 

It's best to guess these at a higher abstraction level than just programming languages. I'd include here:

  • Automation (automated tests, CI/CD pipelines)
  • Application containers and microservices (Docker, Kubernetes et al might go out of fashion, but something will likely replace them)
  • Asynchronous programming (Javascript has shown us the light of async / await and I don't think this will be fading in the next 10 years)
 

Sure, these three techs will still be relevant (not sure about async js in 10 years, I'm waiting for when web assembly will get threads support 😁)

My hope is that Docker and kubernetes will be abstracted away in 10 years, basically "make heroku/Google app engine/AWS elastic beanstalk cool again".

Docker and kubernetes make sense in a lot of contexts, they make less sense in others and I feel like some devs are just over hyping the whole thing. The old idea of using a hammer for everything.

 

I don't think any current techs are likely to be relevant in 10 years. The ones that you are devving from 10 years ago will be because it doesn't make financial sense to rewrite them in whatever is then-fresh. The only things that appear to last are the principles behind the implementations. And even those are subject to refinement periodically.

Things that are fine solutions but I wish we don't need in the future:

  • Front-end build complexity
  • Frameworks
  • Component-based everything
 

Can we finally please get rid of Flash? I recently replaced a datalogger that ran Flash with a Raspberry Pi. 😇

 

A data logger that ran Flash? What is this mythical abomination you speak of?

 

datataker.com/DT80.php
see: datataker.com/wiki/index.php?title...
/* dEX itself requires no application for you to install other than a web browser with Adobe Flash 10 (or higher)*/

I bought it for the "makes it's own webpages and displays the data on it's native webserver" feature before I realized it was Flash.

Hardly mythical, but it is their current product, and while they "claim" to be rebuilding the OS to use something other than Flash, there's no actual sign of it.

It is an abomination though! Programming it is an exercise in futility: Hit a key, wait for the echo, hit another key, wait for the echo... Did I mention constantly crashing?

RaspberryPi with RS-485 interface has been a godsend!

 

You want to stay relevant as a software developer for the next 10 years?

These are 3 major things you should focus on:

  • C
  • HTTP 1.1
  • PHP
 
 

Wholeheartedly agree with the sentiment but if Web Components aren't relevant tech in ten years I'll eat my shorts. I'm also throwing in a vote for Rust.

 

if Web Components aren't relevant tech in ten years I'll eat my shorts

Ah ah ah see you in 10 :D (I secretly hope they will be as well)

So, Web Components, Rust and... we need a third one.

What instead would you like to disappear from "the face of the tech earth"?

 

In all fairness I hedged my bets - I do not own shorts.

My 3rd "expect to see a lot of" is C++11, and I can't think of a single tech I know that will be fully obsolete in 10 years. Legacy code endures. You could probably snag a COBOL gig in 2028.

Just don't buy any shorts within the next decade and you'll be safe.

 

I would really love to see Rust overcome C in the next ten years, especially in embedded programming

 
 

Learn higher order concepts and how you can apply them: functional programming, object oriented programming, asynchronous programming (and yes I know that can apply to both).

 

I think we won't find developers in the next 10 years ai will take our jobs

 

i have no doubt one day that AI will be able to do software development on-par with what people do. but not within 10 years. maybe 30.

 

ahahha come on, that's sad :D

I don't believe it will be so dire, but well, we'll find out soon enough

 
  • leetcode easy
  • leetcode medium
  • leetcode hard