File: EhyRoXlWAAAEsvC.jpg (37 KB, 511x671)

37 KB JPG

Fuck frameworks
Fuck soidevs
Fuck pajeets

>>102415930
3 (web backend) languages that aren't complete trash (my top 3) but could be better: Go, C#, Kotlin

>>102416233
I should try Kotlin, I've used Java for backend but not Kotlin. I get the sense that Java is about a billion times more common in web dev jobs than Kotlin is though.

>>102416453
Companies don't want to switch if they are already using java. I think kotlin does oop much better than java though

>>102415930
>Go
>Error Handling: Difficult
skill issue

>>102415930
>Error Handling: Difficult

is this bait?

>>102416163
this + fuck J*vaScript

>>102416941
Errors aren't just the "error" values returned from functions, they are also deadlocks that can happen either because of channels or mutex, panics that when uncaught will crash your whole program, even from a secondary go routine.
JavaScript for better or for worse don't even have those categories of errors.

>>102416233
I love Kotlin!
t.Android Dev

>>102416944
>this
>JavaScript
you cannot use this in JS, not in a reliable way at least.
Therefore your statement is undefined. No fucks have been dispatched.

What is a reasonable freelance rate to charge for a simple website for a business? Basically, just a few pages describing what they do and then an email.

>>102417361
Ever since ES6 I've never had any problems with this.
I don't even have to think about it, just declare your methods as proper functions and the callbacks as arrow functions and it just works.

>>102416233
elixir/pheonix

File: file.png (1.14 MB, 1925x957)

1.14 MB PNG

Watch Theos latest video regarding the new o1 model. Hes been an AI skeptic and thought that AI had peaked for the time-being but he completely capitulated in this video and admitted that its basically over. o1 was able to solve all of his programming challenges (advent of code)

>>102417516
anyone who watches this stupid fucking person should lose any credibility as a programmer, I hate this guy so much

>>102417546
>I hate him so his points are invalid

>inb4 smug webshitter says "Yes." with chad meme

>>102415930
The nice thing about Go is that it just works on OpenBSD and you can easyly crosscompile
GOOS=openbsd

>>102417546
He's React Andy, the dev.
Just a few weeks ago he attempted something different by trying to make a benchmark for backend languages and went full retarded with the tests, creating a program that according to him reads data faster than his own SSD reading speed, all because he doesn't understand caching.

const [myVal, setMyVal] = useState<MyValType>(myDefaultValue)

I want the value of myVal to be equal to myDefaultValue when the user first loads the page, but for some reason, it's null. What am I doing wrong?
myDefaultValue is definitely not null.

>>102417546
All of the high profile coding faggots are so full of themselves. Theo's had some real fucking out there opinions stated as facts at some point

>>102417566
I 100% agree with his points, o1 is incredible, I just fucking hate him

>>102417610
There's nothing wrong with that line of code, the problem is somewhere else.

File: file.png (924 KB, 840x969)

924 KB PNG

>>102417610

myDefaultValue

is null, verify by adding a breakpoint and inspecting it

>>102417361
>>102417428
Its been such a long time since I thought of or seen "this" used in javascript (because of my current job not because it isnt actually used). I have gone two years without ever needing it, what is the point of it again? just sugar syntax for referring to an object? why not just refer to the object so its clear what "this" actually is?

>>102417745
this is used to clarify scope

>>102417745
If you've only got a handful of objects or never refer to sibling members in object functions, sure.

>>102417745
To refer to itself, dumb dumb.
That's a pretty common concept in any programming language, except C. I'd expect anyone who has a job to know at least that much.

>>102417702
shit, you're totally right. how could it be null? it's being passed in as a prop, I'm using it elsewhere, and it isn't null there.

File: file.png (679 KB, 850x1243)

679 KB PNG

>>102417813
it must be null there, set a breakpoint

>>102417794
>>102417798
>>102417799
I understand what it is and what it does, I just dont think its necessary. Like in a small personal pure JS app I can see it being used but outside of that no. In a real application you would want to just refer to the object so that you know what object "this" is. Even with frameworks they usually have something that under the hood technically is "this" but in practice is a prop or something. It seems like bad practice to use "this".

>>102417863
You're retarded. You can't always deanonymize every object reference.

>>102417927
Im reading the following in an example from MDN. They are using "this" to refer to the test object yet instead of "this.prop" you could just use "test.prop".

const test = {
  prop: 42,
  func: function () {
    return this.prop;
  },
};

console.log(test.func());
// Expected output: 42

>verification not required

>>102417827
I think the problem might be that it takes a second for the data to load, but that function is being called immediately upon opening the page. Is there a way to make it wait?
I'm a total webdev noob btw.

>>102417991
Ok buddy, what do you do when you've got 10,000 different instances of that object?

>>102418010
setTimeout

>>102418026
>test[0].prop
>test[1].prop
>test[2].prop

>>102418120
दोस्त...

>>102417991
Now let's try this with objects of different types

const square = {
    side: 15,
    area() {
        return this.side ** 2;
    }
}

const circle = {
    radius: 10,
    area() {
        return Math.PI * (this.radius ** 2);
    }
}

const shapes = [square, circle];

Please provide me a way to get the area of each item in "shapes".

File: noob.png (14 KB, 437x296)

14 KB PNG

>>102418301
You're also an idiot.

>>102418328
Ok, now think about it.
You're using a method that makes use of "this".
Do you understand why you would want to use "this"?

>>102418348
>You're using a method that makes use of "this".
No I'm not.

why do i have to learn react? it seems so bloated

>>102418362
Guess I was outed by a retard that at this point may as well be doing it on purpose

class Square {
    area() {
        return this.side ** 2;
    }

    construct(side) {
        this.side = side;
    }
}

class Circle {
    area() {
        return Math.PI * (this.radius ** 2);
    }

    construct(radius) {
        this.radius = radius;
    }
}

const forms = [new Square(5), new Square(10), new Circle(15), new Circle(20)];

>>102418410
Syntax error, construct is undefined. Nice try.

>>102418328
kek my point exactly. thank you for blowing >>102418301 out of the water. using "this" is totally unnecessary when you can just refer to the object name. I am sure there are some niche use cases where you would have to use "this" but in every example I see it isnt necessary

def __init__(self, *args, **kwargs):

>>102418463
It's necessary when you start using object constructORs and factories.

Either way, all you really need to remember when using 'this' in event handlers is that it's the window directly calling a function that happens to be referenced in some object's member list. The function itself has no allegiances or knowledge of belonging to something unless you generate a new function via .bind(newThis)

It's a holdover from early JavaScript that lets you use "generic" functions via .call(...)
See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Function/call

>>102416689
yeah it's not difficult but rather horribly ugly.

I learned how to use Turbo today and it's so fucking easy
I love Rails :D

File: apu_apustaja1711366175256.png (39 KB, 400x400)

39 KB PNG

>>102417702
Whats the value of typescript if its not catching this with type checking

>>102418375
It's mostly a business decision. React shitters are a lot cheaper to hire because of market rates.
You can always use preact or ember for less bloat.

Svelte 4 is the current SOTA of web frameworks, but id still pick react or react imitators for economics reasons

>>102420709
you've fucked up your types by asserting that myDefaultValue cannot be null in the first place.

>>102421756
Whats the use of types if everything have to be Any? type
Thats just dynamic programming with extra steps

>>102422087
post your code, you shouldn't have anything that's "any"

>>102417132
With JS you can just try/catch, easy peasy

>>102417572
Indeed. I tried cross-compiling a C project and it takes a bit of effort. With Go it's stupidly easy.

>>102418375
It is bloat but like the other guy said, everyone knows it, so it's easy to hire devs, and therefore companies choose to use it

>>102420362
>Rails
It isn't 2006 anymore

>>102420743
Is Svelte worth learning?

>>102423711
Don't be jelly

>>102424672
Rails is okay but I just prefer other languages to Ruby to be honest

>>102422087
Typescript types dont persist at runtime; they only exist to verify the assertions that you have made a priori. Typescript improves safety but does not guarantee it

Shopify and Github.
Both are implemented using Rails.

File: go.png (108 KB, 1400x791)

108 KB PNG

Any Go devs here? I'm trying to learn Go by doing a small side project. I have a problem when it comes to using one interface for testing (storing items in a map) and also using that interface for prod (using a Postgres db). The return types of the methods will be different... How do you solve this? Do you create one test interface and one for prod, using e.g. a Postgres db. Or is there some way of using the same interface because there has to be (right?).

Do you return interfaces instead of types? Or do you use generics. I guess both of those solutions would work. What do you do (and why)?

>>102425705
Ohh husbant you chose rails and now we are homeress
https://news.ycombinator.com/item?id=41458441

>>102425705
Yes but weren't both of those started in the mid 2000s?

Hey anyone mind giving feedback on this: https://captchify.com ? I just launched it and I’m trying to get out of the big tech rat race

>>102425820
I am a Go beginner and I have never used generics in Go

Help me choose a UI library for a local repairman's webpage. I feel Tailwind makes it too sexy.

>>102415930
I paid for Bootstrap Studio. You can't even open or edit HTML files, it's all barely usable visual editor crap and a shitty .bsdesign file. I was literally scammed... How else do people use Bootstrap then? I need the templates as I have literally negative 1000000000000 artistic talent, but the editor is legit ass and a scam.

File: alpine.png (13 KB, 1352x386)

13 KB PNG

lemme guess, you need "more"?

Okay bros got any good resources for java spring boot?

>>102425820
I'm more of a Java developer.
You'll want to use the same interface but different implementations:
One mock implementation for testing; then your Postgres DB for production.

Why should the return types be different?

>>102429527
https://www.baeldung.com/
https://stackoverflow.com/
And nothing else.
t. been doing Spring Boot for the past four years.

>>102429630
>doing Spring Boot for the past four years.
What kind of design patterns, package style, folder structure, strategies do you use

should i use cookies or tokens for authenticating sessions

>>102429896
Tokens in http only cookies? You compromise security if you store tokens any other way. But then cookies or tokens is a weird binary cause cookies are storage medium. You should make a choice between whether you want session based auth or token based auth for auth type, opaque tokens or JWTs for token type, localstorage or cookie for storage type.

>>102430016
makes sense thanks
>You should make a choice between whether you want session based auth or token based auth for auth type
this is what i meant

>>102430199
Session based auth involves session id and session store instead of tokens.

>>102417516
>advent of code
lol, lmao even

>>102424590
Svelte kind of went retard with v5, Solid looks better

>>102429812
Damn. Good question. It's been higgledy-piggledy over the years, but it's converged on a few things:

Design patterns:
- Controllers + DTOs + DTO factories (to and from entities)
Contains all programming logic until you can refactor it elsewhere.
Specifications encapsulate request parameters.
- Entities annotated with Lombok's @Data (typically without JPA stuff: That get's used closer to the JPA repositories)
- Repositories + DTOs + DTO factories
I'm often dealing with external systems, so I rarely have the luxury of applying JPA stuff beyond the system's own MySQL database.
Specifications are also translated/reinterpreted here.
Don't be scared of JPA criteria API.
And depending on IDE, you might need to write the static metamodels for JPA criteria by hand.
https://docs.jboss.org/hibernate/orm/5.0/topical/html/metamodelgen/MetamodelGenerator.html
- Services are RARELY used
Read Domain-Driven Design by Eric Evans for a good run-down of where and when each pattern should be used.

Package style:
controller/service/repository/etc. for small projects.
General domain concept names (masterdata/order/user/etc.) for large projects.
You'll get a feel for when the project's grown too large and you need to do the switch.
I generally try to follow Google's naming convention here and use singular over plural; but sometimes I can't help myself and go for plural.
https://google.github.io/styleguide/javaguide.html#s5.2.1-package-names

Folder structure:
Maven's standard (src/main, src/test).
My last project, however, grew to the point, where I split up the whole thing into Maven modules with the following convention:
groupid/artifactid/domainconcept/technology
and the technology was either repositories for a specific ERP system or the common implementation with repository interfaces.
Wrangling that thing included getting each module to use common test/demo data and inheriting version numbers through POM files.

(1/2)

>>102429812
(2/2)

Strategies:
Unit test tricky logic.
Integration test repositories; mock them elsewhere.
Spring Boot Test the MockMvc for each endpoint.
If testing times get you bothered, find a way to convince Spring Boot to load less unnecessary components during testing.

Don't load dependent entities until you're in the DTO factories heading to the outside world; leave them as references within the system itself.

REST endpoints are of pattern:

POST /api/v1/entities (Add new entity, HTTP 201)
GET /api/v1/entities (Get all entities, HTTP 200, make sensible use of request parameters, MAKE USE OF PAGEABLE! NO EXCUSES!)
GET /api/v1/entities/{id} (Get entity with ID, HTTP 200; avoid having to deal with compound IDs)
PUT /api/v1/entities/{id} (Add new or replace entity with ID, HTTP 200)
PATCH /api/v1/entities/{id} (Partial update entity à la https://datatracker.ietf.org/doc/html/rfc7386 , HTTP 200)
DELETE /api/v1/entities/{id} (Delete entity with ID, HTTP 204)

There's bound to be more, but I can't think of anything else at the moment.
Domain-Driven Design is a must-read.

>coming back to /wdg/ after years of being a product designer
what did i miss

>learn this language
> no that language
Me a chad practicing my fundamentals so I can learn any language

>>102415930
I'd argue golang's error handling is easier than js, but after seing zig's error handling, it feels like shit now.

>>102434161
Yes but what language are you using to practice your fundamentals? I'd say learn at least one compiled language, one interpreted, JS because web, one scripting "glue" language (bash).

Which ones doesn't really matter unless you are looking for a specific job.

>>102429195
People no longer uses bootstrap. I use tailwind and if I just want widgets like bootstrap I use daisycss on top.

>>102434275
>daisycss
daisyUI*

Imagine using java in 2024

>>102435767
Redeem the maximum performance for all your business needs. With the power of jooby and vertX will be the instruments with which JAVA affirms its position as the icon of web server performance
https://www.techempower.com/benchmarks/#hw=ph&test=fortune&section=data-r22

A GC lang shits on your manual memory memelang

JS is king.

>>102436232
JS = Jesus chriSt

>>102436403
not that kike

Got a pull request in one of my public repos today. It 100% looks like GPT automated slop.

Last week during hiring I also got GPT cover letters + resumes. Is no one aware that the word bloat these GPTs are throwing up is absolute slop?

File: Screenshot_20240918_07064(...).jpg (185 KB, 720x1544)

185 KB JPG

>>102427843
This element needs centering

>>102434275
Thanks for letting me know. I'm not really a web dev but the server applications I make need a web frontend UI. I shall now research this.

The t in lgbt stands for typescript

javascripters made a better concurrency model, than those garbage collector fans

youll see how Go and other fiber-like models fail, though i see it now already

>>102437427
>cannot convert undefined value to boolean

>>102437719
>Boolean(undefined)
>false

>>102427843
Just reading the name initially gave me the impression that it was some sort of service to do with CAPTCHAs (i.e. adding a captcha to a site).

First impressions of the front page: A whole lot of text cards that mean little. I'd like to see previews of the product itself in action.

Style-wise: A bit too similar to the default Bootstrap

Code-wise: you should consider switching from Vite to something like Astro or Next.js for the front marketing pages to at least serve the HTML content without needing JS.

Content-wise: you need some sort of previews (images or videos) of the user area e.g. if you have graphs show them closest to the first fold. You're there to sell the utility of the product and it's hard to filter out from all the marketing noise and blurb.

https://www.swift.org/blog/announcing-swift-6/
Swift is already at 6 while kotlin just launched 2
Its so fucking over jeetbrains bros

3 YOE with php and looking to make a switch because my job sucks, what framework would provide me the best job opportunities?

>>102425705
Github had really fucked security issues because of it.

>>102439615
The C++ interop seems like the killer feature. Hopefully it's good, it looks like a nice language.

>>102439829
Do you want to switch languages? Most jobs are Js/Ts
Do you want to switch php frameworks? I'd say most are Wordpress (ugh) or Laravel.

Starting a new SSR web app and with the announcement of swift 6 I'm wondering if I should do it with Vapor or stick with Go that I've used previously.

I guess another option could be C# but razor is pretty vomit inducing desu

>>102440319
I actually would prefer JS/TS, it's what I knew before getting this job

>>102439615
is swift just apple flavored c# at this point

>>102442564
Yeah but apparently its reference counting GC is better than C#'s stop the world GC.

>>102439829
>php
I'm so sorry

The only benefit of node is that pretty much every programmer is already somewhat familiar with js

>>102442564
>>102443020
Swift compiles to machine code right? While C# requires the CLR runtime

>>102437427
Typescript is Microsoft’s software equivalent of the wage cage. It’s the speed bumps and safety belts they needed to hire a bunch of gay retards and call it DEI.

>>102443826
Yes, but it requires a ton of libraries even if you don't use them.

>>102445381
I could not disagree more.
TS is not a speed bump. TS is not a safety belt.
TS does not make shit type safe.
TS is a time wasting linter. And that is the reason why people push it, because it hides the fact that they are not productive behind a massive wall of additional crap.
>look, we totally types now
>we just had to add these 500 lines of interface code to make our input component work

>>102445969
What you just described is a speed bump.

So why don't people use managed VPS services? why is the discussion vercel / aws managed cloud which can bill you 400000 out of nowhere VS set up VPS which gets ddos's and goes down. I never see any discussion about managed vps solution - is the markup not good enough to hire shills or something?

>>102446045
A speed bump (potentially) saves pedestrians. It serves a purpose. You are free to debate how useful it is in doing so.
TS on the other hand does not "slow you down" in a reasonable we in order to save a third party. TS only slows you down in writing code, but it does not make that code better or safer at all.

Tricking TS is easy and people do insane mental gymnastics to make certain structures work - just in order to comply with the type behemoths that they created.

Yes, that is a "speed bump", but for dev time only. Not for safety, therefore I disagree with the wording.

>>102416944
This, all my homies use ECMAscript

>>102443020
Reference counting is not inherently better than generational garbage collection, it's in fact less efficient, both at freeing chunks of memory and avoiding fragmentation.
Reference counting is however deterministic, which may be desirable in some use cases.

>>102417680
Prime, Tsoding, sphaerophoria and AdamLearns are the only good ones on purple site

>>102425820
In go you want to accept interfaces and return structs.
Have a Storage interface with your storage layer implementation defined. Your 'app' accepts this interface. Then in testing you can pass in your concrete implementations of that interface backed by the map, and in prod, the one backed by postgres.

I can give some examples if needed..

Realistically how many visitors could you expect a simple php/apache site handle on a 512mb/1vcpu vps?

>>102448464
at least 0

its impossible to tell, the code running and the complexity of the app will make a difference, is it doing DB calls? Is it calling other services? They will be the bottlenecks over just returning some static HTML.

>>102448663
A super small shopping cart/e-commerce store, the products will be stored in a flat file, either php array or json, stripe will process payments and the orders will be saved into a SQLite database, I’m not worried about concurrency issues with SQLite because a high conversion rate will justify switching to mysql and a bigger vps. I’m just curious how far something simple like this can stretch.

>>102448782
cunt i doubt your items are going to be the next yeezys overnight this is something you're years away from worrying about

>>102448782
make sure sql is in WAL mode

depends on how you code it, that small of a machine and stack could easily support up to thousands of rps

>>102448889
I wouldn’t be interested in squeezing a $5 vps if I thought that you big stupid dummy

>>102448982
thanks, I’ll look into that

>>102415930
I was looking for a stupid questions general but it's not there so sorry for the dumb question. I have a little neocities site and a domain name. I was kinda hoping to use it but I don't wanna pay their supporter fee on top of what I'm already paying for the domain. What's stopping me from just like pointing the domain to the site anyway? I don't really care if it still shows neocities when they get there I just want people to be able to use my domain to find it.

>>102448782
It is able to serve plenty for a $5 a month project. It can also handle hundreds of percent more with caching solutions.
Though I wouldn't have the balls to run SQLite for concurrency, especially for things which are very important like orders and payment info. I'd rather run MySQL on the server to have a more reliable setup (which is also vertically scalable).

>>102449075
See if your registrar supports domain forwarding (basically a redirect or loads the site in an iframe) and try it. Neocities might try to detect and prevent it but maybe not.

>>102447886
Prime memes more than he codes. Taoding and sphaero are good though.
>>102417516
I've never seen this guy say something interesting that someone else didn't already say or that I didn't know already.

>>102449075
You could copy it to a github repo for static web and point your domain there.

>>102445969
>TS is a time wasting linter
>we just had to add these 500 lines of interface code to make our input component work

this is false and i can't be bothered to explain why. just by merely having TS in your project gives you inference and autocompletion which is already a huge speed boost.

>>102445969
retarded post

File: 57b08a3754a87938f69b9096a(...).jpg (42 KB, 705x540)

42 KB JPG

https://lobste.rs/s/bvgpwg/zine_static_site_generator_written_zig#c_ygmg6j
Troon territorial fighting can be brootal. Ziggoons are just straight up stealing rustroon project names and rewrite in zig.

>>102415930
React question, useCallback is only useful if used with React.memo? does it have ANY use by itself?

>>102451063
>inference and autocompletion
you don't have to write code in ".ts" files for that.
>>102451095
show me your TS code then.

>>102451874

You could also use "useCallback" if you need to save a function reference (with it's own dependencies) for use in other hooks (most notably effects). But yeah, if you want to avoid renders of memoized child components, you still need to memoize some references (like functions) you pass to them.

I've gotten far with django and tailwind but I fucking suck at javascript. its over

Are frontend devs even actual devs? Making a website or a react app feels like glorified markup

File: 12.gif (42 KB, 300x100)

42 KB GIF

>>102415930
I would not call error handling in go difficult.
It's tedious but not difficult.

>>102452716
IT IS glorified markup. Yet people will create the most insane transpiling house of jsx cards imaginable just to avoid writing actual markup.

Still, I do not see how this clashes with what a dev is.

>>102452716
>>102453029
you know youre good police when you write a lot of import statements

File: Swift_logo.svg.png (26 KB, 1024x319)

26 KB PNG

>Write a hello world program in Go
>It doesn't compile that quickly
>It requires boilerplate like 'import "fmt"' and 'func main'
>The binary is 2.1 MB

>Write a hello world program in Swift
>It compiles much quicker
>It requires no boilerplate, it's literally just 'print("Hello world")'
>The binary is 73 KB

Maybe I should learn Swift. There is a back-end web framework for Swift called Vapor.

why is go so fucking comfy?

>>102452786
i agree, its not difficult, its just annoying af and it bloats functions
i started returning errors from my handlers just to make the functions shorter

>the current state of the job market
India and Pakistan have won havent they lads?

>>102453871
Better than just not handling half of the code (error path). You could if you wanted to (just put down a _ or nothing at all) but this will lead to code that isn't as strong.

Is there a point where using plain JS/HTML/CSS becomes inefficient? For example, I was trying to create advanced filtering with simultaneous sorting by name, color, and size, and unselecting options without re-rendering or making new HTTP requests. While I know how to do this in plain JavaScript, it feels inefficient to write 300+ lines when a framework/library like React or Angular could handle it in 50 lines.

My friend apparently had to build an e-commerce store with a stack of his personal preferrence, only to be lured into a false sense of security as they told him he was overcomplicating things with react+typescript without any good reason, and that he should have sticked to the basics. That was his tech assignment for a junior role.

>>102454098
we use shit like react because it makes the dev experience way better compared to writing vanilla stack
if your friend had to build an e-commerce site then he probably did the right thing to use react, just imagine reading vainlla js code for that cancer

>>102454098
My current project is a nightmare of complex workflows, and I'm not about to reinvent the wheel by writing it all from scratch in HTML, CSS, and JS. Shitting on react really show you never handle a real project with overly complex business flow

File: file.png (60 KB, 1059x439)

60 KB PNG

>>102453529
i don't know much about swift but I wouldn't call go compilation slow nor would I call go compiled binaries slow in general.
make no mistake go is a garbage collected language with far different goals in mind.
Now i have heard that with newer versions swift lang is aiming to be less apple oriented, which is good.
Haven't tried it though.
Andreas kling seems to like it enough he wants to use it in ladybird development.

>>102454600
I wasn't saying Go binaries are slow, Go is pretty fast from my understanding, but it's just interesting that Go binaries are a lot larger than Swift binaries

>>102454098
>it feels inefficient to write 300+ lines when a framework/library like React or Angular could handle it in 50 lines.
I have done excessive research on this specific topic and I have yet to see a single example where the most readable, most efficient vanilla solution required more LOC than a framework example.

>only to be lured into a false sense of security as they told him he was overcomplicating things
Who is they?
What exactly was he building?
What was his actual fuck up?

Frontend frameworks fundamentally focus on the way we bind actions and state to the view. They are not solving the issue of not complicating things, they are not solving the issue of filtering lists particularly well, they are not making API requests in a special way.

>>102454196
>nightmare of complex workflows
>Shitting on react really show you never handle a real project with overly complex business flow
How exactly is that related? React does UI, not business flow. If you think that the latter got less of a complex mess because of react, you dun goof'd

>>102454098
>>102455598
>Is there a point where using plain JS/HTML/CSS becomes inefficient?
>it feels inefficient
>the most readable, most efficient vanilla solution
Right, I'll pipe up. I wanted to avoid being anal-retentive, but it's bugging me.
Efficiency with respect to what, exactly?
Memory usage, CPU usage, network traffic, size on disk, developer time and effort, etc.?

>>102455598
>Who is they?
>What exactly was he building?
>What was his actual fuck up?
Reading comprehension suggests, that the assignment-giver made his friend insecure about his chosen tech stack.
Anon might want to elaborate on whether his friend followed their advice and went plain JS/HTML/CSS or not. It's implied but not explicitly stated.

>>102455903
>Memory usage, CPU usage, network traffic, size on disk, developer time and effort, etc.?
All of the above. Post your code (or an example which resembles it) and I will show you.

The thing is: DOM "manipulation" has become *very* efficient. Both with respect to dev time (thanks to modern APIs) and of course in term of performance. The DOM is as a matter of fact - per definition - always faster than whatever framework you are throwing at it, but it is also a very, very efficient data structure. For example: Having "items" in a JS array vs list items in html - the same speed *except* if you are doing things which actually change the appearance/require a re-render. But at that point react (or similar) is also not doing better.

>Reading comprehension suggests, that the assignment-giver made his friend insecure about his chosen tech stack.
Exactly, which is why the statement as is stands is null. If I wrote that "they told me that writing shit *with* a framework was slow on dev time* you'd be asking questions, too.

Is 16 gigs of RAM enough for web dev? The laptop I'm looking at says it accepts up to 32 gigs in its manual, but you can only configure up to 16 on the website.

I guess I can just make a large swap partition for memory hungry stuff.

>>102455548
>but it's just interesting that Go binaries are a lot larger than Swift binaries
Static linking vs dynamic linking.
Go is self contained, not to mention your build most likely wasn't stripped from debugging data.
Swift can shit itself if your OS doesn't have all the dependencies with the correct version.
But this topic may be too complex for a retard who thinks a main function is boilerplate.

>>102457072
16 is plenty, I use memory hungry IDEs like IntelliJ and have Docker running locally and have plenty spare.

File: file.png (116 KB, 700x493)

116 KB PNG

>>102455548
Go is statically compiled although if you really for some reason want dynamically compiled binaries(if for some reason you care about binary size)
There is also go from GCC
Although sadly that only has compatibility with go 1.18 as of now.

>>102457824
Boilerplate is boilerplate but you're too stupid to understand that

>>102453529
boilerplate?
dude java has boilerplate, especilally if you want to set up a project with a build system.
go's "boilerplate" i would say is the recommended folder structure that is technically not required.

Is freecodecamp.org dead or still a good resource? Should I just stick to the links in OP?

>should i use some random shit I found or the stuff you specifically arranged for me

Bow you shits https://github.com/varabyte/kobweb

>adding unit tests to small project
>want to test that objects are getting written to the db
>create user A
>save to db
>create an identical user(?)
>fetch user A from db(??)
>compare if user A's data is equal to the identical user's data(???)
I think I might be retarded. I'm just testing my database library here aren't I? I was planning on switching from jdbc to jpa so I can learn the latter, and I thought adding the unit tests would make it clear of what shit I broke in the process, but I have no clue what I'd even test.

>>102459321
Yeah that’s a bit redundant if your db library and db have their own tests. You should be testing the functionality of your code, so if you’re transforming or validating data independent of the db then that’s what you should be testing.

>>102460847
(Cont) I’m speaking about unit testing there^. If you want to test that everything works then do end-to-end testing but those should be written in the context of the user and what the user expects your software to do. My philosophy is start with end-to-end tests and good error logging and then if something happens to break fix it and write a unit test to prevent it from breaking again but there’s no need to go overboard with unit tests unless you’re building rocket ships, missiles, pacemakers etc. but that’s just my retard opinion.

File: wat_2.gif (958 KB, 500x305)

958 KB GIF

>>102459321
>Unit testing
>Interfacing with a database
Surely you mean integration testing in this case.

>>102459321
>>102460847
>>102461069
>>102461145
I have never had to use a testing framework like jester or selenium, I just manually test things myself. I want to learn more about using testing frameworks but I feel like its redundant. What am I not seeing?

>>102457072
You don't need a separate partition for swap in 2k24. Get more RAM if you use virtual machines it you find it too little but 16 gigs should be good for a start.

>>102462244
You can automate it so that you don’t have to manually test each thing each time you make a change. For simple applications it’s overkill but as it grows it will become tedious to go through and test each feature and you are likely to forget to test certain things.

>>102459321
you wouldnt really create another identical user.
you would grab the data from user A and then test that it has the values you expect.
for example users should have a unique field, like email or username, that you can test

its true you are just testing your database library, but thats only if the only thing in your application code is a sort of ORM create! function

but thats obviously simplistic, normally youre probably going to test something a little more complicated than that.
the benefit also is that if youre testing data returned after creation you can change the implementation of your application code.
maybe you started with a create! call and you want to switch it over to regular sql - then you can do that and the test will still work.
remember, test the interface, not the implementation its more flexible.

thats one of the reasons why putting methods on ORM models is so clunky. to have access to the ORM method calls you NEED to instantiate ORM model instances, which are expensive and limit you to a certain interface. needing to use that interface is limiting, while just looking at data on generic objects is extremely flexible

>>102415930
What operating system do you use?

Is it Windows or Linux?
I'm trying to follow an online course, and the material insist that I use Ubuntu because that's supposedly that majority of webdevs use.

I still heavily prefer Windows, mainly for having a mainstream OS instead of dualbooting and I have managed to recreate the setup the course provides with Linux on Windows (ex: setting up git).
I was wondering if I really do actually have to use Linux because they're right, and it actually is the industry standard? I wouldn't want to be the special snowflake using Windows when everyone else is working on Linux. Or is Windows actually more widely used than the course says it is?

Thanks

>>102464738
It's entirely up to you, senpai. I'd say if you're going to stay on windows, use WSL. I will say that I tried this back when I didn't know shit about shit, and the tech that I wanted to use (Ruby/Rails) was slow as fuck and I hated it. I eventually switched and I've been using Linux Mint for around half a year now. Feels better than Windows, now. Shit loads faster, there's less jank, and I never tinker with anything. When I game, I do it through steam and everything I've tried has worked well thanks to Proton - which takes no configuration.

Overall, I like Linux a lot more for web dev because there's just a lot of stuff you may eventually need from other people's code that was very likely developed on Linux, for Linux, with directions only for Linux. You're also very likely going to be deploying to a Linux based server, which means using it as your OS get's you a bit more grounded. You'll get comfortable using the terminal, not because you need to, but because it makes you faster, which will also make you more comfortable when deploying to a remote VPS via SSH...

Do you -need- to? No. It can only help, though, and there's really no barrier after installing it. There's no "oh gee, how do I do X in Linux??". I'm assuming you're doing TOP, and if I remember they try to force you to learn git first too, but this is retarded because you have no need for it. Meanwhile using a Linux distro isn't some hurdle/heap of shit before you can continue progressing and makes installing languages/libraries/frameworks/etc a simple terminal command.

>>102415930
in react, does useTransition trigger a re-render of the current component?
https://jsfiddle.net/s84a9get/
why is it that whenever I click a button a console log of the current components gets printed and then the new one?

File: The_Wolf_of_Wall_Street__(...).jpg (84 KB, 1200x630)

84 KB JPG

>>102465291
Dont worry about it
Dont listen to peak performance autists they dont ship anything. Your here to make money or be right?

>>102465291

Yes it does. It's so that you can indicate to the user that the component is updating during the transition. Transitions are meant to be used if the render is going to be costly (i.e. hogging the main JS thread causing fps freezes) and can be canceled with further setState() calls.

React is already smart enough with batching setState calls together so don't feel like you need to use transitions to batch up an update, it's only if the render is going to block the UI

what's the best way to scrape twitter using python?

>>102458001
I'm not saying Go has tons of boilerplate, I like Go, I'm just saying it's more boilerplate than Swift

I’m playing with Cursor for the first time. We’re cooked bros.

>>102467672
>not using MS Paint to code
ngmi

>>102457960
Fair enough, you are right, my bad

I just built a hello world program with GCC Go and it's only 89 KB

Meanwhile you can build statically linked Swift binaries if you download a special SDK. So I did that and compiled a hello world program, and it's 41 MB, way bigger than Go's which is only 2.1 MB.

>>102417516
Now get it to do something it hasn't seen before.

>>102468459
Like what?

>>102457072
>Is 16 gigs of RAM enough for web dev?
Yes. If you do solo work/students/single person projects, then it is more than enough.
If you work on a larger codebase with lots of libraries and stuff, then get more, either 32 or 64, and the company should pay for it.

At my previous company I had to run multiple projects with multiple databases, memory hungry libraries, processes, etc. so having 64 GB memory and lots of cores were nice. A weaker machine would slow me down so much, but I don't need that or want to pay that much for a laptop for my personal projects.

File: dumbIdea.png (143 KB, 1323x817)

143 KB PNG

Given pic related in vanilla JS/ES modules and html: Why is it a dumb idea that does not scale? Roast me.

>>102469391
>OOP brainrot
everything is static, yet you still push it into a class

>>102464738
>What operating system do you use?
Linux, but you can use Windows for normal web development.
Lots of developers use Windows, not that it is great but it works.
Note that some packages requires you to install some system dependencies, for example FFMPEG to handle processing and modifying movie files, that might not work on windows since then your Node process will actually startup another FFMPEG process that will run in the background. This is one example, and sometimes it works on windows since the program has a windows version, other times it won't work.
But these are outliers, and only relevant to backend tasks. For frontend development, it should not matter at all which OS you use at all.

Just use Windows, it is fine.

>>102464738
you could always just use WSL on windows

>>102469391
at the very minimum if you intend to go full vanilla, I'd consider using web components and then subscribing to a map/data struct when the component has been created instead of using slower query selectors.

>>102469391
>3 files instead of 1
And how are you planning on doing conditionals and loop renders?

>>102469414
>everything static but classes in ES6
allows for a nice syntax especially on getters / setters. Not completely sold in this though, but that does not change the overall structure.
>subscribe to map ... webcomponents ... because querySelector is slow
Really bro? You need way more code for this and I kinda doubt that querySelector is slow, especially compared to your alternative which would be running more JS code instead of whatever internal solution the browser is using.
>>102470006
>And how are you planning on doing conditionals and loop renders?
Well, I do not to be honest. With what I am describing I can simply add remove elements at any point in time doing whatever pattern/function I want for that. If an $element is in the DOM, the associated "render" (I hate this term, but lets keep it this way) function makes sure that the element looks correct.
BTW: What do you mean with "loops"? People always make it seem like doing lists in html/js was a difficult that only a framework can handle - when in reality it is actually one of the easiest to get done.

Bump bastard

>>102470627
where the fuck did you pull out all that from me saying that OOP is brainrot?

Why is it that whenever I am working on a textbook to help me understand and learn how to make my own projects better, I am met with errors or resistance that should not happen and extend the expected learning session by what feels like 20 times?
Not angry, especially since it helps me become a better system administrator.
I just want to know why these things happen.

I feel like that because Ii heard from colleagues in the company that my apprenticeship amongst them might not end up with a full time position at the end.
So I am trying to teach myself some azure stuff and Full-Stack development with c#, angular and SQL server 2022 so I can apply to the other IT teams as a new dev ops or software dev with a solid foundation in system administration.

This is probably the only decent place to ask, so sorry if it's a little off topic.
Does anyone know of a decent Guestbook offhand that I can setup on my site? The only hard requirement is that it has some kind of anti-spam capability without making external calls to any websites, I want it to be entirely self contained. No reCaptcha, etc.

The old one I had got someone spamming it pretty consistently, which I asked for to be fair to them. I just miss having the option to see who stopped by my stupid shit-ass site, and figuring out how to setup anti-spam measures without making calls to external sites is hurting my peabrain.

>>102415930
I need to make a webpage for my professional profile / portfolio.
What should I use? Js? Django?
I want it to be seamless for mobile browsers.

>>102472718
html.

>>102472729
A bold opinion

>>102472729
Thank you anon this is a good idea

>>102471967
I obviously missed the reference to >>102469544
You could have noticed that if you did not intend to only bitch and instead focused on the actual topic.

>>102472718
if you're planning on having your website be part of your portfolio then use whatever stack you want to demonstrate some surface level of proficiency in
if you aren't then it doesn't really matter, use whatever makes it the easiest for you to build and maintain because you sure as fuck aren't going to touch it again for another 2-3 years

>>102472718
Use what you want to use

>>102472718
Typescript with React. Make the import React very obvious in your code. That way when employers click View Source on your page (which they will btw), they will immediately see that you are an employable candidate.

How do I expose my website without some fa/g/g/ot kicking me off and forcing me to reset my IP address?

>>102475187
>make sure that you look just as dumb as every other person, so people will absolutely have a reason to hire you and not any other person looking exactly as dumb.
right

>>102475418
Does it require a server? If it doesn't use a static host like vercel, cloudflare, or netlify

>>102475675
It's a ML demo, so I'll need to expose my GPU server. I think the cloud servers are too expensive.

>>102475526
Real apps use React with Typescript, nobody wants to employ Mr. Hobby Tinker Guy who "knows best" and can't integrate into a team. Radarr uses React, that was good enough evidence for me to pick it as my frontend language. Good enough for Servarr = good enough for (You).

Maintainability and scalability is VERY important and downplayed by crappy AIs who will make it seem like the only reason to not write a huge enterprise app in Assembly is "difficulty". It makes allusions to scalability without any real understanding because, obviously, AI is not sentient and never will be.

People can pretend to themselves all they like, yet we see the endless wasteland of Rust failures, and companies fucked by Rust now they realize years later "wait, fuck, we actually need to adapt our code to this new thing we're interfacing with".

>>102475418
>without some fa/g/g/ot kicking me off
Ask yourself to not kick yourself off, easy

When is React actually useful? It seems like overkill to me even for stuff like e-commerce.

>>102475778
>muh important maintainability
hence the seasonal rewrite I guess
>muh scalability
you do not plan on having a billion different buttons visible at once, do ya?
>AI
>Rust
what has that shit to do with any of this?

>>102476660
All of these things are equally as important as performance and resource use. The balance is never 100% in one direction or the other. Rust demonstrates the abject failure in neglecting one of those important elements, and Electron demonstrates the other.

>>102476763
OK, going back to the topic then.
I was to hire somebody, I'd have basically two options: Pick the fool who does the same idiotic shit as everybody else. The fool talks the same old crap, write the same shitty code and argues using the same dumb arguments which do not apply to 99% of all problems they are facing. I therefore know, that I can replace that person with ease and that there are hundreds of people waiting just to step it - probably while accepting less money even.

But there is the other option: Pick the person who can think for themselves and reason about the stuff that they are doing. The person who is willing to invest time in finding better solutions and to actually grow and get to the next level.

And you, as an employee basically have the same options: Work for the company that can and will replace you once they find someone even cheaper. Work for the company that wants you to stay stupid and do the same shit over and over again until some day in the near future you'll get completely replaced by AI, because that shit has gotten good enough to "build" the "quality" product or service that they are offering.

Or work for a company that is actually on the edge, where you are actually building unique solutions if they are needed.

The choice is yours. And mine.

Why would I start learn php or laravel right now?

>>102478441
to get a lambo?

>>102475526
it's not dumb it's called bare minimum
they'll look other qualities on top of that to find a reason to hire you, this is not too hard to understand
sorry but this is a real world

>>102479114
the orange lambo it's a meme for the creator of laravel owning one.. right?

I was asking for job opportunities or for freelance

File: image12.png (51 KB, 1080x1122)

51 KB PNG

Unburdend by what has been

File: 1724491109862457.jpg (185 KB, 1024x768)

185 KB JPG

>>102479114
Correct answer

>>102415930
What do you use to hash API-keys in your database? I want to make them as safe as possible but I don't want to take a big performance hit since they will be used often. The API-key won't be able to change anything critical, only access and modify some resources.

I use bcrypt with a work factor of 16 for passwords and the owasp cheatsheet says
>The bcrypt password hashing function should be the best choice for password storage in legacy systems or if PBKDF2 is required to achieve FIPS-140 compliance.
>The work factor should be as large as verification server performance will allow, with a minimum of 10.
So I'm thinking about using bcrypt with a work factor of 10 for the API-keys. What do you think? Is that a reasonable trade-off when it comes to performance and security?

File: 1531927242878.jpg (17 KB, 443x332)

17 KB JPG

Why isn't my react app rendering?
https://codepen.io/AyDotEfDot/pen/poXBpab
The terminal on my browser shows me "Uncaught ReferenceError: exports is not defined" and "Uncaught ReferenceError: require is not defined" coming from react and react-dom itself so I assume I just didn't write my code right.

Why is web hosting and web development still a clusterfuck in 2024?

>>102483728
you got a thousand problems with your code and that's the very least

I'll transfer the project over to stackblitz and try to fix it for you, but I don't even know where to start lmao

>>102483728
>>102484084
here you go, it's building, but you'll need to actually implement it

https://stackblitz.com/edit/vitejs-vite-1mrxn1?file=src%2FApp.jsx&terminal=dev

>>102483685
SHA-256 is enough since it's random and presumably long enough. There's no need for salting since it's random.

Is cloudflare pages like vercel? If I use the free tier to host a simple site using their kv and d1 is it possible that they surprise me with a bill?

File: grug.png (657 KB, 745x663)

657 KB PNG

>>102483831
We made it far more complicated than it should be. Most applications could be a stupid CRUD with maybe some JS on top, most of it would be fine with HTMX or even jQuery.
Develop a monolithic application in RoR, PHP, Node, Python, Java... with a simple MySQL or Postgres DB behind with some replicas and cronjobs to make backup, all of that hosted on a simple Linux machine, you can already serve thousands of users with no issue and minimal optimization work.
But no, we need layers of abstraction, and we need that because it make us feel smart to know 50 different things, and it also makes your career move faster.

>>102482498
explain the benefit, because I have never understood shadowroot in the first place.

What substances can I ingest to make me more productive at web dev?

Caffeine is one, what others are there?

>>102484971
This doesn't even get into all the crazy shit you can do with load balancing, DNS fuckery, anycast, proxies, etc.

>>102485466
Inside shadowroot ouside css rules do not apply
Its a local reset

>>102484722
Just dont give them your card details lole

>>102486346
So how is that a benefit then? Why would I want that? I want shit to look consistent after all, but having to define CSS in shadowroot a second time does not seem to help.

>>102486485
To unburden from what has been and focus on what can be

>page 9
Fuck's sake

>>102485527
Start something you actually want to make, even if it's something you think is impossible. Get it past the scaffold/boilerplate phase, and if motivation doesn't take care of itself then you probably just don't like webdev.

>>102488079
>t. man under 30

>>102483831
Because people refuse to accept that 99.99% of any application is just CRUD so everything get's fucky wucky because people can't handle the fact that it's actually really simple (and could be easy), so they create solutions to problems that don't exist which just further obfuscates and complicates it.

For instance, people think Rails sucks because it's old, but I just made a monolith that functions like a SPA for the user with no JS, has 1 source of truth (the server, as God intended), that has 10-20ms loads for large index pages with a lot of data, and 5ms for normal loads - and that's in dev mode with all of the logging and other shit wasting time. It comes already wrapped in docker. I can drop this fucker on a VPS and be done when it's time. It has likely saved me months of work. There have been days I've added 2 very large features in the span of a few hours.

There are other techs which do the same: Laravel, probably .Net Core (unsure how they handle the front end), and probably more. Once you get into the backend API, and frontend JS, webdev becomes a retarded clusterfuck, and that tradeoff provides basically zero benefits.

We are wrapping back around to "SSR is actually good", but it's being done in a backwards way via meta-frameworks. We are soon going to be in an era where SSR means "prerendering JS components" and people won't even understand what SSR actually means. Just like kids learning React but having no basic understanding of DOM manipulation in the first place. They're also being conditioned that "server functions" are a thing and not just code that runs on the server... Image paying a company each time they run a function you wrote... Schizo tin-foil is that this is all just so AWS wrapper companies can make shitloads of money.

>>102417516
are you mentally retarded? have you ever actually used it? I tried to make it do something trivial in go and it failed miserably even after I tried tuning the prompts a bunch of times, it just does what it wants because it's so biased towards its training data that it produces overcomplicated vaguely related slop
I've been using these since gpt2, while it's alright for generic boilerplate it completely shits the bed as soon as your application becomes specific
in multiple cases I would literally enunciate what the inputs of the function are, what it does with them and what it outputs and o1 still just slops everything up because of training data bias

>>102488179
I'm over 30, I'm just not a sack of fat and actually take care of myself so I'm not tired all the time. I bet you're one of those
>haha my back hurts
>being 30 am I right, LOL
fags. No, being tired and sore all of the time isn't normal at 30, or even 50. Take better care of yourself and your brain won't be stuck in 1st.

>>102488281
preach, I fucking hate normies
they never take responsibility for anything, I've seen your lifestyle nigga, you smoke and you drink and then you joke about being old at 25
I always just say that I can't relate and that I still feel 16

>>102485603
none of that is nearly as difficult as some people want to make it

>>102417516
>machine can replicate $thing that can be found in the webs
>person whose only contribution to mankind has been to create <ButtonComponent class="$unreadableTailWindGibberisch" /> is not creative enough to trigger the edge of what machine is capable
>anon on 4chan wonders about dumb person
really?

>>102485603
Like 99% of apps could just have apache, nginx, Caddy, etc sitting in front of it and be fine. You do not need 30 servers across the globe. You do not need a load balancer for less than 10k MAUs. You're worried about shit you either will never need to worry about, or will have enough time to figure it out well before you actually need it.

>>102484971
>>102488235
Good informative posts.

>>102483831
it's piss easy wtf are you talking about hosting a front end is free on netlify or whatever and for db/backend there is a ton of basically precofnigured stuff for whatever set up you are doing

What was the thought process behind the decision to use JavaScript as a backend language?

>>102489946
besides all the shit that javascript gets, the language is actually not that bad. It's got some quirks. All programming languages have their rough edges.
But people are simply doing insane stuff in the web in general and that is why the language got a bad rep, especially back in the days when browsers (in general) were not as capable and people still pushed stuff over the edge.
one of these groups thought that JS would make a good general purpose language and thanks to the web being the absolute dominating runtime - they were kinda right.

>>102489946
It was probably "it's better than PHP", which is true.

File: file.png (17 KB, 1271x92)

17 KB PNG

reeeeeeeeeeee fuckin async testing is so fuckin annoying

I created and own a 6 figure per year website about 7 years ago, without ever learning javascript or HTML etc. I did pick up HTML and CSS by happenstance though, as I had to make little edits in WordPress, but I didn't know even one line of JavaScript.

Well JavaScript is still horrific, but I picked up Typescript and this is excellent! It's like a real programming language. I made a login page and it was so insanely easy. This is awesome, I can actually make my pages DO THINGS without needing plugins, damn.

>>102488281
Being 30+ is a massive monumental shift in psychology. That's why people become fat etc. I lost sex drive basically the second I turned 30, and I only ever went to the gym at all to look good for women. I feel like 90+ mentally. Physically I could obviously work out and do cardio and whatever else, I just don't care anymore. No motivation to do anything but code.

>>102489946
dude le full stack js app lmao

>>102490672
>still preaching his bad habits
My brother in Christ I'm over 30 and I'm telling you if you feel like shit it's your fault. You don't care because your body is meant to move daily, and you're not doing that, so your T and mood are trashed which trashes your motivation. Be a fucking man and do what you need to. You aren't 12, mama isn't coming to make you take care of yourself anymore.

>>102490677
Yes.

>>102484722
>>102486363
They let you do that? Either way can someone clarify for me, if I host a full stack like remix or nextjs app on cloudflare pages and I go over the free tier, is it pay as you go?

>>102490799
Yeah I could, however I don't want to.

>>102491662
eat shit

>>102492861
Eat shit yourself

Does anyone know if Exercism is going offline and if so when? I read on here via a link at HN that they’re not making any money and that it’s not a profitable venture for them

>>102447495
ECMA has LIGMA

>>102416607
>oop
>better

>>102493896
>reading
>comprehension

>>102493976
>comprehension list

>>102493878
what's ligma?

File: file.png (26 KB, 540x540)

26 KB PNG

>>102493896
uhm, akshually, he made a comparisson between the syntax of Java and Kotlin

>>102494297
Ligma balls lmao

>>102489946
back end language/performance doesn't really matter for 95% of cases so just using the same one as the frontend is just convenient

>>102417361
> ()=>{} blocks your path

>>102489946
>how do we get the webpage codemonkeys to parse Excel files and set up REST endpoints for the webpages to feed them the Excel info

>>102417361
>uses OOP
ngmi

>>102495091
>back end language/performance doesn't really matter
You may not hit a hardware limit while using JavaScript, but you will be paying more for the servers because of the memory usage. It's not even a trivial difference.

>>102418375
you don't have to learn react, you need to learn web fundamentals so that you can prompt chatGPT to write good react for you

>>102476517
>When is React actually useful? It seems like overkill to me even for stuff like e-commerce.
No one?

File: 1726471543789.jpg (535 KB, 1841x1382)

535 KB JPG

>>102489946
>mogs your memelang is techempower benchmarks

>>102493363
All rails shitters is going bankrupt
Just waiting for that smug DHH ass to file chapter 11 so i can shitpost more furiously for a day

>>102497995
you dont need it for ecommerce, there are already a million premade solutions for that

>>102420709
Typesvript is for autocomplete not safety lmao

>>102422098
>using any
Breh

File: 1568937588-93453-full.png (23 KB, 136x102)

23 KB PNG

>>102498282
Kek

>>102498198

That 100% has to be a fucking dude. And what the hell is Elysia?

>>102498343
C++ wrapper https://elysiajs.com/

>>102498386

> Code of couduct
> Has discord

I'll see myself out.

>>102498486
>added by bullies
https://github.com/elysiajs/elysia/commits?author=hisamafahri
Doesnt count
True CoC is in the power of our friendship. Its when we laugh and cry with our friends.

>>102498585

Please, take your meds like the rest of on this website. I am not your friend and never will be.

>>102498386
>C++ wrapper
But no C++?

>Uses tabs instead of spaces
Erm, what the sigma?

Still, pretty neat, if it's claims of being faster are true.
But if you are going to write backend, might as well do it in a good lang, like Rust.

best tech stack for lazy AI wrappers?
I wanted to use Python or TypeScript due all the support they get (the official LLM SDKs, ten million RAG libraries, scraping libraries, etc.) but I just don't like Django and I despise Next.js.

>>102499338
>Uses tabs instead of spaces
Dropped. Sooner break bread with trannies than tabbers

File: 1725098352139355.png (52 KB, 891x718)

52 KB PNG

What are the best 4chan.usercsses?

>>102501605
For me, it's the default

>>102502382
Me too, son. This shit is so addictive. Dunno how moot pull this off.

Any of you use vim/neovim for anything more than quick edits?

I'm adding some plugins to vim. So far I have language servers for JS/TS and Go, and I have a file tree sidebar.

I'm not using neovim because Debian (my OS) provides an old version which doesn't support some plugins.

>>102476517
>>102497995
The only time things like react makes sense is if you are actually building something with many dynamic views which YOU do not control. e.g. stuff like facebook.
If you are building a web page or an app that has defined UIs, almost every other solution is better suited.

I uhh, haven’t coded since June. How fucked am I?

>does ur job

how long until people catch on

>>102503749
Why haven't you?

>>102445969
Embarassing. Closer to real code = happy times.

>>102452039
Not him but TS is easier to write than JSDoc, at least I think so

>>102452039
Let's just make this very clear. Go on any Javascript Udemy course, reviews by: Sangdeep R., Rajesh Ganesha, Singh P. Shiva, Pajeet K.

Typescript Udemy course reviews by: Max S., Petar Breivik, Michael B., Anthony M. Brown

That's all you need to know. Never promote javascript again. If you can write one you can easily write the other so just use the one white people use as white users proves superiority.

>create code to resize image according to the window resolution
>works on my PC
>works on my tablet
>doesn't work on my phone
I hate the internet.

File: img-2024-09-22-17-28-11.png (211 KB, 1818x1878)

211 KB PNG

>>102505791
Typical Typescript students.

File: img-2024-09-22-17-30-55.png (269 KB, 1718x2126)

269 KB PNG

>>102505841
And now Javascript. Same instructor for both.

File: the-west-has-fallen-v0-2o(...).png (50 KB, 659x692)

50 KB PNG

>>102505846
The J stood for jeet

File: 1567860108316.jpg (54 KB, 954x960)

54 KB JPG

I'm at that point where I can create virtually anything I want but I'm just too cynical and nihilistic to make more stuff, everything is just boring.

File: Herrscher_of_Human_-_Ego.png (752 KB, 960x660)

752 KB PNG

>>102498343
Elysia is Miss Pink Elf.
https://honkaiimpact3.fandom.com/wiki/Elysia/Story
https://www.youtube.com/watch?v=_BUh64YHE30

New thread
>>102508368
>>102508368
>>102508368
>>102508368

>>102505791
>>102505841
>>102505846
thank you anon. pivoting to typescript now. fuck pajeets (if they are women, literally)