When I start to feel overwhelmed, I just go check the Days since the last new JavaScript framework counter.
Usually calms me down.
this post was submitted on 07 Jun 2023
27 points (100.0% liked)
Programming
13376 readers
1 users here now
All things programming and coding related. Subcommunity of Technology.
This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.
founded 1 year ago
MODERATORS
About ten years ago, I was getting stressed out about it. Someone told me:
-
Stick to what you know well, if it will do what you want the end user won't care.
-
If what you know won't do what you want, find the thing that will do what you want and just focus on that. No need to go down a million rabbit holes.
-
If your team is using something you don't know, learn that and just focus on that and ask for help if you get lost.
I've found this advice to be extremely useful.
I'm managing some devs right now, one of whom is going really slowly. She's determined to do her part of the project in React. I said to her "How comfortable are you in React." She said "Not really, I'm totally new to it. I've used Angular in the past and React is really different." I said "Why aren't you doing your part of the project in Angular?" She had no good answer.
I feel like when you're managing a team you also have to consider the skills you want future devs to have to have. Not saying this is necessarily the case for you (for all I know you already have a mix of React and Angular), but on my teams we have bottlenecks when we need to do work in certain plugins because only one person knows VB6, or WPF, or has the license for the third party library needed to compile the plugin. The dev may not be available for weeks/months because other teams need work done in that tech. If everyone's using the same stack you can just assign tasks to people based on their availability.
I feel like when you’re managing a team you also have to consider the skills you want future devs to have to have.
I fully agree.
This particular case is a startup in a very sensitive area rushing to MVP that has chosen to imbue it's devs with a lot of autonomy and build a distributed infrastructure where they treat each other as black boxes. This decision was made before me, not that I necessarily agree or disagree, it's mostly working, but I'm struggling to get everyone to document their shit and I'm concerned about bus factor and replacing people / maintaining these systems in the future.
Anyway, I think that was her thinking when she chose React (wanting to learn it) but her progress has been underwhelming at the last few standups (she's built very little while the rest of the team is rocketing along). She's built so little in the last three weeks that she even agreed (today) that pivoting to Angular makes sense. We'll see if she can catch up.
That's fair--the right stack is the one that delivers on time. I wish her good luck and less stressful future sprints
The only ways that this potentially falls short is if you want or need to change jobs and the market is looking for people that know the thing you don't, you're potentially at a disadvantage. Also you may be able to do a thing with the tools you know, but there may be different tools more suited for doing the thing that takes fewer resources or allows resources to be reallocated to other things that customers do care about.
I could best sum up my answer to this question as "lol no"
Not proud of it though!
Man, we just finally got away from supporting Internet Exploder.
Svelte is where the fun is 🙂
We have a guy doing svelte on my team and it looks great, I'm just hoping it's not going to be a blocker down the road when we need more people. Is it popular these days?
Yes, even Apple has been hiring Svelte developers for 4 or so years
I keep meaning to build something interesting in svelte. I was very excited about it a couple years ago and rebuilt a super simple version of the homepage of my primary site, but I didn't really feel like it made sense in the way React does (admittedly, I've worked with React a lot more...). What's your favorite thing about working in svelte?
Personally, I avoid any fronted dev and stick to backed.
I used to keep up but eventually just burned out. I do plenty at work so if I do have a hobby project in mind, I tend to do something other than just webdev, to spite it up a little bit.
It does feel scary sometimes just how quickly you can get outdated, especially with frontend. I want to stick to backend but it feels like all the fancy jobs eventually end up with 90% work being on front, APIs and DB seem like the easiest part of it all (especially if managed properly)
I gave up when back in the day, we had like JQuery, AngularJS, Vue.js, React.js, and so forth and so I just stick with JQuery for better or worse for most of my professional career in ASP.Net Core development. (CDN alleviate the trouble of distributing JQuery and web browser would cache it, so I don't put much stock on people claiming that it's bloated or heavy.)
I often bring up that we just needed better GUI toolkit with a designer and to replace all of HTML/JS/CSS with just WebASM and WebGPU. Rather than supporting legacy crappy unholy trinity languages, we could push for "survival of the fittest" languages/tools to fill into this space.
I do my best to keep up, but it depends heavily on what my day job has me working with and how modern they keep their tech stack.
Does Blazor count? I've done a little with that. Mostly though, no. My job doesn't require much frontend work and even if it did it would be jQuery at best.
I don't think you'd really need to either unless you genuinely enjoy being at the bleeding edge. React, Angular and Vue aren't going anywhere any time soon. That could be laziness talking...
I've been loving Blazor and it totally counts! I'm using it in a hobby project to make mini-games using canvas and integrating with PlayFab (kinda like neopets). It feels so nice to work with and can be so powerful and surprisingly flexible. Angular feels so verbose and unnecessarily bloated any time I even look at my colleagues' code now 😅
As a .NET developer with a strong dislike of JavaScript being able to write a frontend using C# is fantastic. The only complaints I ever hear about it is actually around Visual Studio support and not Blazor itself.
I really hope it takes off.
I thought I kind of was... and then I saw your post! I’m coming out of years writing plugins in large old school open source PHP projects and it’s such treat to write my own thing with a nice shiny stack (Laravel/Vue/Vite). PHP8 and ES6 are so much more expressive than their predecessors as well.
I know very little to none about serverless / edge, but to be honest I’ve made peace with the fact I’ll never know everything about the web dev ecosystem. Trying to improve every day is a more attainable goal.
I work mainly with Ruby on Rails. Picked up Vue. I try-harded Elm, ended up with Elixir on my personal project right now (without phoenix). I gave up on shiny front ends. jQuery is still my favorite go to now. Let's see how long it will last.
I'm in a very similar boat as you, and I often feel frustrated with how much there is I'm not keeping track of. But I don't really like coding side projects in my free time, so I just learn as deeply as I can about the frameworks my works teams are using. It tends to pay off insofar as people can usually tell that I've done research, so at the very least it helps me less less insecure...
I do keep up with modern web development but that tends to be changes to things like CORS, client hints, things like that. I never used JavaScript and am never gonna start.
Development is much more simple and stable, and the user experience is far superior, when you cut out JavaScript.
My current open source project is a programming language engine, which I implemented in Kotlin and compile to the JVM, Native as well as JS. That means I can run the JS version in a browser, but I don't have to deal (much with JS itself). The actual web UI is very minimal, because developing for the web is, as you correctly asserted, a nightmare.
I am, to a degree, but only because it is my job :)
Frontend has changed / is changing much more rapidly than backend, IMO.
I'm not a huge fan of this shift towards eng-driven dev-ops. I get why it's happened, but infra has never been particularly interesting to me personally and I don't enjoy owning that aspect of the stack.
Fireship on YouTube gives my regular dose of new frameworks and similar developments. Apart from that not really.
Yeah I still style my pages with tables 😅