You don’t actually have to setup your web-development environment like the hype dictates. I mean, work in a Danish municipality with more than 5k employees and 60k citizens who directly use our, mainly web-based, solutions daily. We build around 50 a year, and maintain a few hundred of them, and almost none of them are build with a JavaScript MVVM framework.
Not because Vue/React/Angular aren’t nice, but because we don’t have to. If it doesn’t need to run offline, then a MVC framework with Ajax will do just fine, and they are extremely productive. Both because errors are server side, but also because asp mvc hasn’t really seen radical changes for almost a decade. We also rarely put things in containers, we don’t do automated CI/CD and we certainly don’t orchestrate things with Kubernetes. Not that there is anything wrong with doing that, but we’re not Netflix, we can publish a new build directly to an IIS during a slow period and no one will notice the service being gone for 5 seconds.
Sometimes I think it pays to do a little JOMO instead of all that FOMO. Especially if you don’t want to burn out. Changes are a constant in our business, but you need to make absolutely certain that adopting those changes make sense, and it has to make sense in the real world, not on hacker news. Our business doesn’t care one bit about the front-end tech stack, as long as it performs like they want it could be written in ASP webforms.
What do JOMO and FOMO stand for? I think of FOMO as "Fear of Missing Out", the controversial motivator driving retention for free-to-play games like Fortnite, but that doesn't seem to be correct in this context.
It’s completely anecdotal but I do hire people and I work as a part time examiner for CS students, so I do come into contact with what motivates people. Unfortunately I see a lot of fear driving people to push forward, picking up technologies because they think they have to, not because they actually need them.
I mean, if I do a quick search on the Danish job agents, no one is looking for someone who knows GraphQL. Yet I see people desperately trying to learn it in their free time because they are afraid of becoming obsolete.
I wonder where this fear comes from. It’s a fairly recent thing in my experience, and it’s not like it’s a buyers market. We still desperately need more developers.
It probably comes from the internet and how the web development media is heavily dominated by FAANG companies and trendy startups.
If your reference point is these companies and their developers, then it can appear like you're getting more obsolete by the day, since they're always talking about the hottest new tech and you're seemingly not using it.
It's turned web development and software engineering into a game of keeping up with the joneses. And I suspect just like with social media as a whole, it's made a lot of people feel like they're inferior to the rest of the population/their peer group.
But I've got an article about that planned at some point. About how the Silicon Valley focused tech media and internet has given people the wrong impression about what most such work is like in most industries.
They’re productive, because they’re stateless, isolated and have a super short edit-compile-run cycle.
Depending on the architecture of your react/angular app, things become too big, fragile and slow
But you don't work for a business, you work for the government. I think that makes a big difference in terms of what's expected and also what the consequences of non delivery are.
Having worked both in the public and private sectors I think it’s actually the exact opposite. If we miss a deadline, or if a system goes down when it really wasn’t supposed to the consequences can be as severe as someone dying because life-necessary data wasn’t available. The only sector which is as focused on security, on-time-delivery and risk-management that we are is banking, which is frankly rather similar to the public sector, except their business focus is much more narrow and they generally have a lot more resources.
This is actually part of the reason we try to think about what technology we use and how we use it. We have too many critical systems and too few people, so we need to be extremely efficient in our choices. If something is cool, but adds complexity, then it’s just not worth it.
Because we are the public sector we do a lot of benchmarking though, and we actually have ASP WebForms legacy apps that are rated as high as your favourite mobile app by our citizens. Which was my point, your users don’t actually care about your tech-stack when it delivers.
Yes I know he says that but in general if they're are no real incentives to achieve delivery and no consequences for non delivery, as is usually the case for public servants, then the results tend to be not very good.
Yeah, I don't know, I feel like the constant framework churn has sort of died down and javascript is no longer changing the way it was a few years ago. It essentially moved from being a scripting language a decade ago to the 'full fledged' programming language it is now. Evolution during that time period was necessary, and a lot of frameworks and trends turned over pretty quickly. But now I feel like the JS ecosystem is on a much more stable juncture and not evolving as much. The most popular MVC frameworks are 5 years old and show no sign of going anywhere anytime soon. Best practices for tooling have mostly been decided. I think the web should continue to evolve but I just can't imagine it being as dramatic as it was over the last decade. Or maybe, I'm just a dinosaur too.
Yeah I find it really weird that he'd write this _now_, when most of the technology he's referencing is 5+ years old, and there hasn't been that much churn in the last few years. About 4 years ago I started learning React sarcastically wondering what the next thing I'd have to learn was would be, and four years later I'm... still using React, and it's still pretty cutting edge.
Yeah, I agree. Strong declarations about what the future will look like usually are laughably incorrect in retrospect; but, well, without some major paradigm shifts in how browsers work, I don't really see React/Vue going anywhere. The only thing I could maybe guess would be some AI that abstracts the programming away.
The real danger is working 14 Hours/day instead of 8.
Clearly is a problem of toxic environment in the office (none notices that you're there for so long hours every day?).
It's not a problem of tech, which is fluid, but a management problem. You don't always need the last fancy technology or tool, but pick what you might think will help you and go for it.
Being pragmatic should be the 1. Point in a developer's career
> I work 8:00 AM - 5:00 PM. Then 9:00 PM - 2:00 AM. And in between, I'm "on call."
Yeah, don't do that. Primarily, make sure you get at least 10 consecutive hours off every day, to wind down and sleep usually 8 hours, some need more. And get some exercise every week, and don't forget to eat healthy. And make sure you are not so stressed that your primary social relationships involve yelling or worse.
That will go a long way to prevent burn-out, no matter what industry you are in.
Yeah, this has nothing to do with what he is doing but with how much and how long.
There is a reason we fight for the 8/8/8 standard and why we arestarting to demand more free time instead of more work.
I'm not a web dev but try to extend back-end to full-stack especially on personal projects which are mostly to try out something new. I was impressed by what React achieved but not by codebases using it. Vue was my go-to SPA. Now it's Svelte with the v3 release. I've yet to use it for anything substantial but that's my best hope. In the past I've used RoR and even Yii/php which are impressive. For non-SPA I'd now choose Phoenix/Elixir.
How I stay sane is not to go 'all in' on any framework. My pick for mobile is Flutter but that could easily change to Kotlin Native or Svelte Native if they turn out better than the game engine rendering mode. Declare this is my fave 'for now'.
I'm sure this is much harder when you've committed to a stack for a good part of your career. I tend to grow more tired of companies not adopting new tools--the pragmatically good ones not just shiny. I'm also always looking for my next language, maybe Clean or Pony but happy to use Kotlin, TypeScript, or Dart in the meantime.
Only one of the complaints is technical related, the rapid pace of front end development to someone who’s “skipped Node, RoR and Python”.
As a “web dev” for the past two years, previously elsewhere, who has never worked in an agency, none of this applies. The front end pace absolutely would have two or three years ago, but things have generally stabilized.
Not because Vue/React/Angular aren’t nice, but because we don’t have to. If it doesn’t need to run offline, then a MVC framework with Ajax will do just fine, and they are extremely productive. Both because errors are server side, but also because asp mvc hasn’t really seen radical changes for almost a decade. We also rarely put things in containers, we don’t do automated CI/CD and we certainly don’t orchestrate things with Kubernetes. Not that there is anything wrong with doing that, but we’re not Netflix, we can publish a new build directly to an IIS during a slow period and no one will notice the service being gone for 5 seconds.
Sometimes I think it pays to do a little JOMO instead of all that FOMO. Especially if you don’t want to burn out. Changes are a constant in our business, but you need to make absolutely certain that adopting those changes make sense, and it has to make sense in the real world, not on hacker news. Our business doesn’t care one bit about the front-end tech stack, as long as it performs like they want it could be written in ASP webforms.