Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I feel like I made a mistake investing professionally into Flutter

reddit.com

Post reply on HN
Print view View original

301–308 of 308 posts

Re: I feel like I made a mistake investing professionally into Flutter

#301

Post by pjmlp »

mgkimsal wrote: post #214
ThinkBeat wrote: post #199

Earlier quoted context omitted.

true. But Java 2023 is quite different from Java1. The tools you use now are different (unless vi) You have had to learn a lot of new technologies, frameworks, syntax to keep up. PHP has also gone through so many changes that it is near a different beast these days than when it first appeared. So if you had relied on your skills you learn as PHP 1.0 and Java 1.0 today you would not be very productive.

Much/most of Java 1 will still compile, at least (FWIR). The evolutionary progress of java 1... to whatever it is now (current one I'm using is 17, but I know there's more past that) has largely been adding of new libraries/methods, etc, and the majority of your java skills from 10-15 years ago are still valid (although might not be the most efficient any more). I've watched people go through Flash/Silverlight/Action…

> Much/most of Java 1 will still compile, at least (FWIR). ...

Actually it won't, past Java 9, as after the modules introduction it was also decided to take action on deprecation, now things get really removed after a couple of releases.

Re: I feel like I made a mistake investing professionally into Flutter

#303

Post by pzo »

jcelerier wrote: post #285
pzo wrote: post #130

Earlier quoted context omitted.

I'm wondering about this as well, actually this weekend been doing Flutter 'Test Drive' of their tutorial and experience was not that great. I'm still forcing my brain to parse this complex structure with very simple widget structure from their samples. Its Flutter: ``` class MyHomePage extends StatelessWidget { @override Widget build(BuildContext context) { var appState = context.watch (); var pair = appState.curren…

I'm curious at the point you are trying to make - I don't know any of the three but Flutter looks by far the most obvious and easier to read and understand

I recommend putting those code snippet I shared to VSCode and using default formatting - this will be more visible in IDE. But other annoying things for me:

1) SwiftUI feel more like jquery so that you can chain styles like padding, color, font etc. Because of that :

- code completion just works and is fast you just press a '.' and see all choices

- it's easy refactor the code if you wan't to change styling or layout

2) Flutter abusing too much composition and adapter pattern to the point they using refactoring in VSCode to add simple padding etc. Just have look at this official GIF from 'Test Drive' - how many times 'down arrow' got pressed:

https://codelabs.developers.google.com/static/codelabs/flutt...

Now imagine designer come to you and telling you we change layout and styling slightly - in flutter I feel it's easier to throw away such layout code and start from scratch than trying to refactor it. In swift UI you just change VStack to HStack, change maybe some parameters and move around chain of .padding().color() around and are done with it.

What would improve flutter readability is extending Dart to have something like Swift:

- first parameter can be marked as optional specifing parameters name such as '_ child:'

- escaping parameters in Swift so you instead of doing ... children: []) you can just close previous Widget and make also 'children' parameter name optional so that you can write :

Row(.min) { ElevatedButton( onPressed:() { appState.getNext(); } ) { Text('Next') } }

- similar enum behavior like in Swift when you can just type '.red' instead of 'Color.red' - this is even more useful for more nested enums.

edit: I gave up with proper formatting of code here in HN, super annoying.

Re: I feel like I made a mistake investing professionally into Flutter

#304

Post by sanderjd »

scarface_74 wrote: post #259
sanderjd wrote: post #253

Earlier quoted context omitted.

Sure you would. It's all the same thing, just different details. You'd do the same thing you do with whatever stack you're currently using: you'd read a lot, do a lot of proofs of concept and prototypes, make a bunch of mistakes, but figure it out and do just fine. Sure, you wouldn't be as fast as someone who has worked with these tools a million times, and sure, that would matter if you're being hired on a three-mon…

And why would you waste money on waiting for me to come up to speed when there are literally thousands of people who know react and can hit the ground running? And at this point in my career I’m looking for lead positions. Would you hire me to lead your front end efforts? (my area of expertise is enterprise app architecture/dev + AWS. As a developer I spent most of the modern era of my career working with Node, C# an…

I would not be wasting my money, I would be hiring you for all the other skills and experiences you have, which I would find far more compelling than whether or not you already know something you can learn in a few months. If I thought it was a waste to hire anybody who would take non-zero time to learn the stuff they need to know to be an effective employee at a specific company, it would be impossible to hire anybody (except people who already worked there and only left quite recently). Hiring full-time employees is an investment in their ability to contribute net positively to the company over their whole tenure, not in the first few months.

You are right though that I wouldn't hire you to lead a dedicated front-end team. But I'd prefer not to have such a team to begin with. But I'd be happy to hire you[0] to lead a product team working on a product that requires a front-end to succeed. Either other members of your team will know how to make that part work well, and I would expect you to learn quickly from them, or I would expect you to figure it out and then teach the rest of the team.

[0]: I'm using "you" as a stand-in for someone with your experience, who is otherwise a good fit for the organization. I may well not hire you in particular and suspect you would not take a job from me in particular, since we clearly have very different philosophies on specialization vs. generalization and investment in employee training, such that we may not enjoy working together :)

Re: I feel like I made a mistake investing professionally into Flutter

#305

Post by nailer »

invalidname wrote: post #86
nailer wrote: post #85

Earlier quoted context omitted.

Did gmail migrate off GWT?

It never used GWT. They claimed GWT can be used to build apps like gmail and that was misleading. They didn't write it using GWT since it predated the existence of GWT.

Thanks. Did any well known Google apps use GWT?

Re: I feel like I made a mistake investing professionally into Flutter

#306

Post by invalidname »

nailer wrote: post #305
invalidname wrote: post #86

Earlier quoted context omitted.

It never used GWT. They claimed GWT can be used to build apps like gmail and that was misleading. They didn't write it using GWT since it predated the existence of GWT.

Thanks. Did any well known Google apps use GWT?

I think Wave was written in it... Not sure if there were others.

Re: I feel like I made a mistake investing professionally into Flutter

#307

Post by mgkimsal »

pjmlp wrote: post #301
mgkimsal wrote: post #214

Earlier quoted context omitted.

Much/most of Java 1 will still compile, at least (FWIR). The evolutionary progress of java 1... to whatever it is now (current one I'm using is 17, but I know there's more past that) has largely been adding of new libraries/methods, etc, and the majority of your java skills from 10-15 years ago are still valid (although might not be the most efficient any more). I've watched people go through Flash/Silverlight/Action…

> Much/most of Java 1 will still compile, at least (FWIR). ... Actually it won't, past Java 9, as after the modules introduction it was also decided to take action on deprecation, now things get really removed after a couple of releases.

Interesting... I have a project that was originally done in J8, and... it's still working/compiling as of a couple weeks ago, but it's not necessarily as advanced as it could have been. I think we moved to J11 a few years ago, and IIRC J17 in 2022, and didn't hit any issues other than some dependency versions.

Re: I feel like I made a mistake investing professionally into Flutter

#308

Post by SpaghettiCthulu »

zigzag312 wrote: post #188
tjpnz wrote: post #182

Earlier quoted context omitted.

I don't understand why this language exists. It just feels like a promotion vehicle by someone at Google who had enough clout to then get it shoehorned in to Flutter. If they had gone with JavaScript they could've made the switch to TypeScript later on. I realise this is being said with the benefit of hindsight but Flutter would be in a far better place today I suspect.

It compiles AOT to native code and supports tree shaking. It also has JIT mode with hot reloading/restarting. It has sound null safety and is statically typed. It's not a bad language. It just lacks some needed features. They'll probably get implemented over a span of a few years. Ecosystem is small, but at least there's FFI interop. It also has some baggage from it's early days when it was meant as JavaScript replac…

> It's not a bad language.

I agree. The team behind the language is the real problem.

> It just lacks some needed features. They'll probably get implemented over a span of a few years.

For years they have dismissed calls to add multi-threading in the language, despite it being sorely needed for anything computationally non-trivial (isolates have ridiculous overhead).

Post reply on HN