Thread 106544071

Do you follow the 14kB rule?

>Free beginner resources to get started with HTML, CSS and JS
https://developer.mozilla.org/en-US/docs/Learn - MDN is your friend for web dev fundamentals
https://web.dev/learn/ - Guides by Google, you can also learn concepts like Accessibility, Responsive Design etc
https://eloquentjavascript.net/Eloquent_JavaScript.pdf - A modern introduction to JavaScript
https://javascript.info/ - Quite a good JS tutorial
https://flukeout.github.io/ - Learn CSS selectors in no time
https://flexboxfroggy.com/ and https://cssgridgarden.com/ - Learn flex and grid in CSS

>Resources for backend languages
https://nodejs.org/en/learn/getting-started/introduction-to-nodejs - An intro to Node.js
https://www.phptutorial.net - A PHP tutorial
https://dev.java/learn/ - A Java tutorial
https://rentry.org/htbby - Links for Python and Go
https://quii.gitbook.io/learn-go-with-tests - Learn Go with Tests

>Resources for miscellaneous areas
https://github.com/bradtraversy/design-resources-for-developers - List of design resources
https://www.digitalocean.com/community/tutorials - Usually the best guides for everything server related

>Need help? Create an example and post the link
https://jsfiddle.net - if you need help with HTML/CSS/JS
https://3v4l.org - if you need help with PHP/HackLang
https://codesandbox.io - if you need help with React/Angular/Vue

/wdg/ may or may not welcome app development discussion. You can post and see what the response is.
Some app technologies of course have overlap with web dev, like React Native, Electron, and Flutter.

We have our own website: https://wdg-one.github.io

Submit your project progress updates using this format in your posts, the scraper will pick it up:

:: my-project-title ::
dev:: anon
tools:: PHP, MySQL, etc.
link:: https://my.website.com
repo:: https://github.com/user/repo
progress:: Lorem ipsum dolor sit amet

Previous: >>106528949

i don't get how to use FF profiler, it seems complex interface, and i don't know what i really need to look for on the data it shows me.
last day i opened lighthouse, since it provides direct diagnostics

>>106544071 (OP)
This thread has a better OP than the previous thread. Good work.

>Do you follow the 14kB rule?
No of course I don't. Pic related.

>>106544364
second AAAAAND second
56kb modems are not to be considered

>>106544562
>56kb modems are not to be considered
Indeed. A lot of web pages that people visit these days will download a few megabytes of stuff - JS, fonts, images, whatever. As long as it loads fairly quickly then it's fine.

you faggots are so useless that i'm considering installing discord and learning from them instead

>>106545388
The purpose of /wdg/ is to shitpost about web development. The purpose isn't to teach you for free.

>>106545509
> NoT YoUr PeRsOnAl aRmY

Cynical take. Are you 12? You don't have to engage with posts you don't want to engage with; people come to this board with very different intentions.

Shit, anons, I'm really fucking lost for the past week now.

How does one implement resizable grouped columns with Tanstack table? In a group, they all resize proportionally and size values are ignored.

Any advice on how to dynamically generate said groups and append them depending on data coming in from API? Shit really gets slow with 100 rows and virtualization seems to not be enough.

Fuck, I hate fucking tables. My whole job for the past 7 months has been implementing basically every feature of this library and adding lots of business logic related custom features as well. Fuck me.

>>106545760
>You don't have to engage with posts you don't want to engage with
Yeah I usually don't. I just think it's dumb that some people have an expectation that they'll have everything spoon fed to them for free just because they asked about it.

>>106545928
>for free
What do you mean? Sharing information is the point of the internet. At least it used to be.

While you do have a fair point of criticism, I personally always expect someone to just tell me I'm a faggot. If I get lucky, I get useful info. I think there's many more people, who also hope for (but don't expect) this.

My point is that your attention to such faggots is pointless waste of energy.

>>106546027
>Sharing information is the point of the internet
If someone seems like they're making a proper effort to teach themselves, but they just have a quick question, then maybe I might answer it, but some people seem to think they're entitled to be spoon fed everything for free.

>>106545388
dilation fees, pay up
>>106545509
it is to share knowledge, aka learn, u need to lrn2 what old 4chins really meant, one word, /b/ros

>>106545928
this has happened to me and it is probably why I help others here, lurk moar

I'm going to ask again.
Is it possible to varnish something with CSS? I mean an element having a glow that moves in relation to something else like the position of the mouse eon the screen.
I assume javascript would have to be involved but that's the least of my issues.

did something happenen to /wdg/ why is it just one mentally ill samefag now

>>106546414
If I understand you correctly, this will definitely need JS. However, I'm not too sure what you mean by this mouse moving "varnishing" shit.

You mean smth like this? https://valiant-cron-0pnw.pagedrop.io/

>>106546027
>Sharing information is the point of the internet
Bro never heard of RTFM?
And today you can ask chatgpt and perplexity free and its even better. You dont need to bother internet strangers anymore.

>>106546624
4chins is diededing

>>106546785
AI is a meme that breaks down whenever you're doing non trivial and performance critical job.

Sometimes you read the manual and the manual is not enough or plainly wrong. Not all docs are like Arch wiki, negro

>>106546822
Kys nagger

>>106544071 (OP)
why 14kB specifically?

>>106546414
>css?
yeah probably, try and tweak to death with some js or just hover, if ur lucky there's a premade example
BUT, there's canvas and things like webgl

>>106546624
idk, troons are extremely degen, but it fixes itself, 41%
>>106546785
>le gatekeeping newfag
ironic

>>106546822
correct
>>106546894
faggot
>>106546822
to.be fair, the ascended take is reading source code, not about that life though, and if it is obscure then you can forget about it

>>106546903
>retard fell for retard clickbait
that's the main reason

It's your time to shine /wdg/, we need a logo: https://github.com/trueadm/ripple/issues/49

>>106546743
>You mean smth like this? https://valiant-cron-0pnw.pagedrop.io/
Oh yeah, this is exactly what I wanted.
Mind sharing the code?

>>106546624
You should see /wdg/ on the weekends, it's much more dead

nodebros... that package... I'm sorry
>>106545250

>>106546903
It's because of the TCP slow start algorithm.
https://endtimes.dev/why-your-website-should-be-under-14kb-in-size/

>>106548741
enterprise adoption status?
>a single retard that fell for yt clickbait sets a new global trend: REAL | FAKE

>>106549136
It's what you have to do if you want peak performance. The great thinker and ex-FAANG engineer ThePrimeagen has also talked about it.
https://www.youtube.com/watch?v=ciNXbR5wvhU

>npm did it again
https://www.youtube.com/watch?v=QVqIx-Y8s-s
I told you naive kids about snorting packages

>>106549493
Gochads win again. The standard library is enough for almost everything.

I'm back end developer thinking about launching a couple of my own web apps, but can someone tell me where to get nice tailwind templates. I'm not talking about your gay, everything is rounded and soft color templates, but some actual creative designs than I can pluck into my website block by block and then fill the data in and change some things if I need to.

Tailwind 4 plus and https://lexingtonthemes.com/ are kinda ok, but I'm trying to find as much other retro shit as I can also.

Python CGI is the easiest language for programmers that have learnt Python before to learn web development.

>>106549493
bro wtf, look, YESTERDAY!
>>106521141 (Me)

>>106550417
Don't like Python. Simple as.

>>106544071 (OP)
>dude I just saved 18ms
>*website takes 7 seconds to even render*

>>106550398
I used Daisyui back when I used Tailwind. They have a bunch of different templates and you can customize them and set different themes (dark/light etc.). You can also change the base colors, primary, accent, info color, warning color etc. It's pretty nice.

They also have a "random theme generator" you can play around with:
https://daisyui.com/theme-generator

>>106547341
why name your package after a crypto currency?

>>106547496
Open devtools, copy everything, done. Written in one HTML file inline.

>>106551509
second, taking away muh {}? no way
no but why use python when node exists, no reason that I can see

>>106554682
I also despise Python for the lack of curly braces.

Python is probably better outside of web apps, but almost any white man's lang will be better than Python in that case.

>>106555091
>I also despise Python for the lack of curly braces.
pic related
>
the saying goes that python is top notch for ez AI lib frankestaining, so I used it for that, still, ugly, and the sayings are often wrong, but it gave me an easy time I must say

Just saw a job advert for a "vibe coder". Never seen such a big red flag from a company

>>106555711
>be me
>6 yoe """"programmer""""
>use AI anyways
>correct it
you know VS Code autocomplete? now imagine it completes all the code and you can just tweak this and that and your work is done

>>106555767
if only it wasn't retarded

>>106555767
Yeah I think AI will play an increasing role in software development, but "vibe coder" just strikes me as some TikTok meme rather than a serious job title

>>106555790
yes, they are indeed "taking a risk" and if they're not careful, they'll pay for it

C# or Java, which should I learn?

>>106557407
They all do the same shit in the end

>>106546414
if you mean for a complex texture (the bumps) like your picrel. You could probably do it in a svg filter together with some js to change the "light" source direction via attributes

>>106546414>>106558349
why fake it (poorly), 3d plane, with texture/material and bump/displacement map. use webgl or similar

>>106558665
>why use rendering tricks when you can do raytracing
the absolute state of modern devs

>>106558704
webgl performs just alright specially with the settings I specified, kys, brainlet

>>106558725
That'll be +30 MB thanks
because there's no way he's doing it from scratch

>>106558817
take your 14 kb retardware and stick it up your ass, I'll take the low resources consumming webgl varnish, kys

>>106557407
Learn Go.

>>106558971
now u explain why

>>106559008
Go is easier to learn, more performant, easier to deploy, has a great standard library, and it's built for web dev (great concurrency model etc.). That's why tooling like docker and kubernetes are written in Go.

How do you approach form validation in Go with htmx? Here is what I do:
1 - client side validation using native HTML attributes for inputs like required, maxlength
2 - validate those attributes server side, and usually I return errors using http.Error because only bots fail it
3 - pass inputs to my service layer
4 - if i get a db violation error like a foreign key or a unique constraint fail I render the form and show the user a message

Basically the part I'm hung up on is the step 3 and 4, because the err from the repo is a code but doesn't tell me what field is wrong. Should I instead design my DB in a way that the only way for an INSERT or UPSERT to fail is if it's an internal server error or a single column being violated? For example, if currently account requires a unique email and username and when I get a UNIQUE violation I don't know which. Is it better to instead only require a unique email?

My question is vague because I'm learning Go and only still in the prototyping stage, but I am making a forum.

I got a few natural users, it's nice to see people using the service

>>106544071 (OP)
modern websites are struggling to follow the 1400kb rule

>>106558209
Probably

>>106558971
I've tried Go and I like it but I keep seeing job adverts that want Java or C# experience

>>106560480
Got some annoying error when I tried to post but look at the image for the answer.

>>106559767
sounds good enough, what about error handling, it's often critizised in /g/ in general

>>106561426
>the 1400kb rule
you mean that yt clickbait thing?

>>106562438
whatever pays the most, and second filter is whatever is less legacy, and third filter is .NET is comfier than Spring
I won't talk about GO cause I've never touched it

>>106562965
>>106560480
just make sure you do proper input sanitizion on insert or u get rekt one way or another

>>106563003
>what about error handling
Go forces you to be explicit about error handling. That's good. Some people find it to be annoying but you have to handle errors if you want to write good quality software.

Do you guys test your websites in Safari? I think I should, given that many people use Safari (iPhone users as well as Mac users)

Therefore I might do one of the following:
1. Buy a cheap second-hand iPhone so I can test my websites on that
2. Buy a new laptop (which I should do anyway) which can run MacOS in a VM, so I can run Safari within that (and I could even run the iPhone simulator)

Thoughts?

>>106563084
java/c# and even node seem to have more elegant and convenient error handling
and don't forget it comes from google, the ones that effectively killed the web

>>106563557
browserstack

>>106563701
>browserstack
Hm... minimum price seems to be $30 a month. And I dunno how well it's going to let me test all the particular quirks (and features) of an iPhone compared to other browsers and devices.

>>106563927
it's supposed to be "native", optimistically, you pay one month and you're done

>>106563557
only if i've used a relatively new browser feature, otherwise no

>>106563557
If workplace doesn't provide testing devices, that's their issue. I wouldn't buy into the Apple ecosystem meme, as that's exactly why they have these restrictions.

I either ask someone to allow me do a few tests on their Apple devices or the company should provide Browserstack access.

Should I put this on my resume?
https://nothesahofiran.neocities.org/

>>106565627
yes, be sure to send it straight to the CEO

>>106563039
Thanks anon, my only question is how do you check constraints like UNIQUE in the business logic. Do you use a SELECT? Everywhere I read I'm told to minimize the amount of trips to the DB so I thought relying on error codes from an INSERT/UPSERT would be more efficient.

Thanks for all the other advice, I'm currently refactoring the UX of forms and I'm relying on client side validation. But if the server side fails I simply have htmx append a generic error message at the bottom of the form. My experience with Go has been fun so far and I enjoy using it a lot, my stack right now is sqlite, go+templ, and htmx

>>106565860
I remember that I did this, I don't remember how lels, I guess yeah, the DB/BE returned an error, and that was it, but I assume you can merely just try the insert, and the DB will shit an error and you just handle it with grace, as in you identify the error (unique constraint violation on name column) and you tell the user error here, username taken

>>106565860
>my only question is how do you check constraints like UNIQUE in the business logic
You would have to go to the db to check for constraints like UNIQUE.

>relying on error codes from an INSERT/UPSERT would be more efficient.
That sounds like a good idea when it comes to checking for UNIQUE constraints. That's also how I do it. But you shouldn't return db errors directly to the client for security reasons.

For example:
Someone might want to check if a certain email address exists in the users table. And so they try to sign up with that email address. So you shouldn't return the db error that says that that email address exists directly to the client. Just return some generic error.

>my stack right now is sqlite, go+templ, and htmx
You don't have to worry as much about trips to the db if you use SQLite since it's a embedded/serverless db. So the trips take less time than if you used Postgres or MySQL. But it's still good to handle it in the business logic if you can. For example checking if the email address has the correct format.

>>106565860
>sqlite
Make sure that you use WAL mode if you use SQLite in production.
https://sqlite.org/wal.html

jQuery nigga

>been working with boomers for 6 years
>jquery and raw php at first
>move towards vanilla js, improve php a bit
>eventually build a web components library
>start using sass and tailwind
>then some react
>they want to build the whole app from scratch
>hire a consultant
>see code involving nextjs
>absolutely disgusted about the amount of boilerplate and complex shit
Why is websites getting to heavy and complex fought with more complexity?

>>106566589
I haven't used Next but isn't it basically rendering React components on the server? Maybe it's getting popular because there are lots of web devs who already know React, so why not let them use it for server-side rendering too

Yesterday I saw YouTube's alternative theater mode where the video takes the whole screen and you can place the comments on the side, but today I'm not getting that theater mode but the older one where the video simply expands horizontally.
How do I get the newer theater mode again?
Captcha: AY8GAY

>>106567331
New theater mode:
https://files.catbox.moe/eozyed.jpeg
https://files.catbox.moe/dqyqyn.jpeg
Old theater mode:
https://files.catbox.moe/dcu5fp.jpeg

>>106566113
Thanks
>if you use SQLite in production
I don't know yet, I'm using SQLite because it lets me run integration tests on the same environment as in the real world. I never expect my shitty XenForo clone to go anywhere so SQLite should be good enough lel

>>106544071 (OP)
Hello everyone, I'm happy to know that with the calc-size CSS function being added (soon in firefox too hopefully), you can now have floats that expand in the inline direction instead of block wise, assuming you know the width of the floating thing.

In this case it is just hardcoded values since I'm trying it out: max-width: calc-size(calc-size(max-content, size), size + 13px + 93px + 20px - 40px);
The pixel values being: size + left thumb margin + thumb width + right thumb margin - left post blockquote margin (since the left margin moves the non-floats over a bit)

Btw, shape-outside also allows you to customize the float shape, but that's unrelated

>>106567673
Maybe it should've been a fit-content? Whatever :P

page 10 Sirs

jQuery is the father of modern js, deal with it

>>106544071 (OP)
serious question why is it 14? why can't they just make it variable or something?

Why is webdev so hard without Vyvanse bros?
Updating fetchThreadPosts to match my actual backend is just too grim
I'd rather Watch Yu-Gi-Oh season 5

>>106570675
Because of TCP. But I guess the deeper reason is because you don't want to send too much data over the internet before checking if some TCP packets are lost? That would be wasteful.

Why the FUCK ist there still a Web dev general. None of you will be employed by the end of next year.

How are you supposed to keep the 14kb rule if HTMX alone is already 14kb?

>"Spring/.NET are enterprise dogshit bro, just tons of boilerplate bro, fuck Java/C# dogshit, slow as molasses garbage bro"

I hear some variant of this everywhere I go and yet they're the comfiest and most solid frameworks I've used, just great all around. I'm not saying other stuff is bad, everything has its use case, I just don't know where this mentality came from.

>>106573101
early nerds were all gay hippie retards so you can't separate that out from the scene people hate C# because it's microsoft which they were told to hate by 80s nerds calling it micro$haft etc

>>106573101
>>106573731
I've tried Java and C# and they seem alright, I wouldn't mind using them more. They can have a fair bit of boilerplate though and they can both take a while to start up on slower machines. With something like Node there's arguably a lower barrier to entry. Just write a few lines of code and you've got a server running.
>people hate C# because it's microsoft which they were told to hate by 80s nerds calling it micro$haft etc
I think Microsoft can make some good stuff. I think TypeScript is good. I still try to have a healthy scepticism about massive companies though.

>>106572275
I'm starting Tuesday morning actually, gonna be working with Nuxt and C#

>>106573819
>Next and C#

>>106573819
king shit, count your blessings

>>106573819
>Nuxt and C#
dream setup

This was far too much work before realizing, yet again, that I don't need to manually decode the pixels because DirectDraw Surface BC7 is part of the webgl2 standard extensions through gl.getExtension("EXT_texture_compression_bptc")

Have a scroll through this nightmare
https://registry.khronos.org/DataFormat/specs/1.1/dataformat.1.1.html

>>106575402
What are you making with WebGL?

>>106576613
Making a 0-dependencies tool for Space Engineers to generate interactive 3D & 2D planet maps from cubemap textures and terrain rules xml game data, so others can do what I did manually.
https://files.catbox.moe/us3e4n.png

There's also another project where I'm cloning the visual scripting (and with time, including lua) environment from Stormworks: Search & Rescue
so that you can view and debug vehicle controllers outside of the slow-loading game and its strict restrictions

MCP servers are actually very useful. This is going to be the next big thing in web dev. I used this repo to build one because I like Go:
https://github.com/mark3labs/mcp-go

General question: Have you tried out building MCP servers?

>>106576769
Sounds cool. Does that mean the end result would be a 3D planet map in the browser, and you can turn the planet around, stuff like that?

>>106570675
When the HTTP server receive a TCP request, it doesn't know the connection speed of the client, of itself and any hop in between.
So it starts with a brust of 10 packets and then it waits for the ACK response by the client.
The maximum size of a TCP/IP packet is usually 1500 bytes, remove the header and its flags which is 40 bytes and you have 1460 bytes of data.
1460 × 10 packets is 14600 bytes.
14 KiB is 14336 bytes, a very similar value for those 10 packets.
If the server received the acknowledgement for every single packet then the connection is closed.
On HTTP/1.1 and on HTTP/2, if even one of those is not acknowledged, then the server restart sending them but one at a time.
HTTP/3 is different because it uses UDP instead of TCP, but first connection to a server is never HTTP/3 anyway.

>>106577004
>HTTP/3 is different because it uses UDP instead of TCP, but first connection to a server is never HTTP/3 anyway.
I think HTTP/3 uses TCP for the first connection so the 14kB rule still holds for HTTP/3.

>>106576970
Nice, I like it

>>106576970
this is really cool

>>106577084
No. First connection is either HTTP/1.1 or HTTP/2.
The server nudge the browser into using HTTP/3 with the Alt-Svc header.

>another deadline hell week for new featurues
>everyone just vib-coded it to reach the deadline in time (they wouldn't like the other features)
>the remaining features still ain't fixed yet
Does Web Dev get better?

holy fugg I'm finding myself more and more relying on gpt slop to reach deadlines and calling it a day

we're cooked

>>106581227
It gonna get worse from here.

>>106579779
the junior i have keeps relying 100% on AI and just accepts what it gives without even thinking about it, driving me fucking insane.
he's spent two weeks scaffolding out the frontend of a new feature and it's a completely nonsensical shitshow that i have to clean up this week.
guy even thought the feature would be done this week, lol

>>106581567
Bet the product manager love him.
I have to deal with that crap too, but I am a junior as well. The worst part is that my company don't have a senior, so the guy just 'random bullshit go' on everything with no restriction.
I am quitting this job before I have to deal with the technical debt later on.

>>106581567
Not a problem if you write tests. AI + TDD is GOAT tier fr, no cap on the productivity gains.

>>106581614
small startup so i basically am the PM, devops, senior dev etc lol.. attempting to hire another junior/intermediate has been a nightmare too since no one has been able to complete my bottom of the barrel exercise without completely butchering it with AI

>>106581806
i agree but when the task is full stack 0-100, TDD doesn't really apply here. implementing designs from figma, designing backend control flow, designing client state control flow and so on. guy takes the established conventions i already have in place + documented and doesn't even point his AI tool to it all

>>106544071 (OP)
I was skeptic about PWAs at first. I tried adding one to my homescreen from firefox and it was basically the same minus the firefox ui. Acceptable but not exactly game changing. Looking for a way to remove the firefox badge from the PWA icon corner I read that PWAs can be installed as bookmarks or as actual apps on the system and apparently you can do the latter only through Chrome on Android. Once I did that I experienced an instant increase in responsiveness. Why is that? Is the chrome engine that better? Or is the fact that firefox is only capable of installing PWAs as pseudo bookmarks the reason behind the lack of speed? Regardless I experienced a renewed interest in PWAs and was wondering what you think of them

>>106582104
>PWA schizobabble
You just know...

>>106582104
I think PWAs are pretty cool, and yes I've also noticed that you get the best experience on Android if you install them through Chrome rather than through Firefox, at least on my phone.

>>106576970
this is really cool, but you have to polish it, perfect it, the displacement map is a bit out of control

>>106582180
this cannot be real

>>106582180
All the millinos mpozzila gets from google go to stuff like this and making new AI features in firefox.

>>106582196
I like them because you can bypass app stores but apparently corpos always find an excuse to shit on new cool stuff they have no direct control on https://techcrunch.com/2024/02/15/apple-confirms-its-breaking-iphone-web-apps-in-the-eu-on-purpose/

>>106582431
It's a wonder Apple lets their users choose which websites they visit

>>106582431
Yeah Apple did break PWAs for a short while but then I think they fixed them again. I suppose there's the risk that Apple might break them again at some point in the future. But to be fair to them they have added more PWA features to iOS. I think iOS 16.4 added some new PWA features such as push notifications.

>>106582657
works on my machine

>>106581846
>small startup so i basically am the PM, devops, senior dev etc lol..
kek same except I only have like 1 month of experience in the company stack + tech support on top of that shit.
How do you manage with deadline stress and shit pay?
I have been working over time on Saturday and sunday to match the deadline because le AI makes you build faster, but it felt like this shit has been killing any motivation I have left in coding.

>>106583046
founder is somewhat non-technical so in terms of timelines and stack, what i say generally goes.
i'm not in the US so my pay is terrible compared, but it's pretty good for the flexibility i have and for my country, got a 55% payrise leaving my last company so i'm happy. they pay for my claude max sub too so i get to use that in my free time on personal stuff.

>>106583091
>payed for my own github copilot
>minimum wage
>constant deadline
Fuck, if I am gonna suffer, I should have driven a hard bargain like you did

>>106583110
>>106583091
>pay for nothing
>lmao @ deadlines
>company pays for copilot which includes chatgpt 5 and all the other exotic AI solutions
get a better job

All this fucking whitespace on my widescreen display,
browsing one of the foremost sources of web development

>>106583808
not optimal, but I think I see what the issue is, they didn't want to break the lines of long text, send them a better design q;^)

>>106584185
Just let me close the fucking sidebars without having to make the window smaller

>>106583808
I would rather have a web page like the one you're showing, than pic related. Pages like pic related spread across the entire width of the screen. If you have a wide screen then you find yourself turning your head all the way to the left to see some parts of the page, and all the way to the right for other parts. Also the distances you have to move your cursor are greater. When I'm making websites I set a maximum width for the content and then I centre that, so the whole page stays within the middle of the screen, and doesn't stretch across the whole screen width.

How do I detect if for example the "4" key is pressed on the keyboard together with shift?

Some browsers report the key 4 regardless of shift is held or not, others report they key typed. Meaning it depends on keyboard layout (US layout has Shift+4 turn into $, nordic layout has Shift+4 turn into ¤, German layout has... well I don't know, I don't want to type out special cases for all language keyboard layouts in the world.

Is there any solution for this "key typed is reported" problem?

>>106584341
https://developer.mozilla.org/en-US/docs/Web/API/KeyboardEvent/code

>>106584251
agreed, send them your design, go further, "toggle full table view"
>>106584341
there's probably "key down" event, if that is so you want to track that event
yah
https://developer.mozilla.org/en-US/docs/Web/API/Element/keydown_event

>>106584393
MS gets it right

>>106584435
yes

>>106584375
>>106584393
thanks for trying to help but if it was that simple i would have solved it already

pic related is how it looks when I press 12345 without shift and then with shift (using a privacy preserving fork of Firefox)

>>106583494
>>lmao @ deadlines
what sector do you work at?

>>106584600
multinational company, as much as I'm going to say

>>106584594
https://developer.mozilla.org/en-US/play?uuid=c8d66b7c-0800-428a-b1af-25f781ac19c6&state=TVCxUoQwEP2VnR0LbmQ4CyuFK7y5wtHONsUxyXpEwgZJUJHh393AFVZ5eXnv7b7M2MTO4QOWlvsxQu9qTY13hoZK4dFZ3UJDA%2BUQG2LoBwoBjP9mqKGlqVAIwf6SaO%2FvBO8PisserBHC%2BYvCQ7nvhcMcdQgyRsBHOrXnEGGbWUmgHjviWHyONExv5EhHP2QKV4HC3aPizSGh%2F%2FUXiidHCT5Nzya7Dk1yxau3qI05fcn7qw2RmFKo7J0aKMxT3AtNm%2F59ZB2t5yuZ0Q5mxZCuRaSfePQsCRFuKzjDzUyF9oaWs3iXtaB8UEdSzdRDi8sf&srcPrefix=%2Fen-US%2Fdocs%2FWeb%2FAPI%2FElement%2Fkeydown_event%2F
and you also get called a retard and a faggot for making me work for (You)

>>106584624
>>106584594
https://developer.mozilla.org/en-US/play?uuid=c8d66b7c-0800-428a-b1af-25f781ac19c6&state=TVCxUoQwEP2VnR0LbmQ4CyuFK7y5wtHONsUxyXpEwgZJUJHh393AFVZ5eXnv7b7M2MTO4QOWlvsxQu9qTY13hoZK4dFZ3UJDA%2BUQG2LoBwoBjP9mqKGlqVAIwf6SaO%2FvBO8PisserBHC%2BYvCQ7nvhcMcdQgyRsBHOrXnEGGbWUmgHjviWHyONExv5EhHP2QKV4HC3aPizSGh%2F%2FUXiidHCT5Nzya7Dk1yxau3qI05fcn7qw2RmFKo7J0aKMxT3AtNm%2F59ZB2t5yuZ0Q5mxZCuRaSfePQsCRFuKzjDzUyF9oaWs3iXtaB8UEdSzdRDi8sf&srcPrefix=%2Fen-US%2Fdocs%2FWeb%2FAPI%2FElement%2Fkeydown_event%2F

>>106584594
https://developer.mozilla.org/en-US/docs/Web/API/Element/keydown_event#addeventlistener_keydown_example

>>106584624
>>106584636
Why so hostile? And you didn't solve the problem as is demonstrated in pic related. Notice that Digit2 becomes Quote and Digit3 becomes a blank string.

>>106584654
same thing here

>>106584662
pfft, just map it to keys, come the fuck on
>Why so hostile?
>if it was that simple i would have solved it already
I just solved it for you, just you're too dense to realize
sorry for "hostility" but we helped and you tried to downplay and discard our solution, even when we fucking solved it for you

>>106584680
You think you solved it but you really didn't. I'm 45 years old btw, you're probably a lot younger with this attitude. Like, you thought I didn't try e.code before seeking help with my issue? Come on.

>>106584702
excellent, brilliant, good job

>>106584702
you want me to write the full solution, or are you going to take the solution I gave you, do you want me to push it to your git repo?

>>106584721
It was indeed a mistake coming here with a rare problem. For future reference you should have said "huh that's odd, must be something with your browser fork that doesn't follow the javascript standard" instead of calling me a retarded faggot.

>>106584731
it doesn't work you dumb scriptkiddie.

>>106584702
Regardless of whether you're young or old, you should RTFM

>>106584750
last chance, tell my why "it doesn't work"

>>106584594
Works on my sv-se machine
quit using a fag browser

Webdev is impossible without vyvanse

>>106585704
Personally I hate Big Pharma and I don't want to take their mind-altering substances

>>106548741
>That 14kB includes compression — so it could actually be more like ~50kB of uncompressed data
there is literally no reason a regular website should be over 50 KB

>>106585794
Good goy

>WebGPU has been available in Google Chrome since 2023, and is expected to be available in Safari 26 this fall.
>
>Although Firefox 141 enables WebGPU only on Windows, we plan to ship WebGPU on Mac and Linux in the coming months, and finally on Android.

Welp, time to forget everything and start over again so your browser can mine crypto faster

>>106586089
Surely I would be a "good goy" if I did take Big Pharma substances. I'm saying I don't want to take those substances.

>>106585794
>Personally I hate Big Pharma
as you should, only retards get hooked on that poison

>>106586320
I wouldn't have been able to come this far without THREMBO COMBO
I'm lucky I got to the fun part as everything prior has been a fucking grind.

>>106586680
Cause they're cutting me off I think. I only have like 20 pills left to get me started on gamedev so I'm saving them for after the webdev project is finished and I can move onto the next.

>>106586235
What's the difference between this and WebGL

>>106587411
Support for DX12 , Vulkan, Metal, and future expansion

Great talk about Go:
https://www.youtube.com/watch?v=4hxKEbWO5u0

Go is the best language for web dev.

i need to take an ios development course for my job but i only have and old 2012 macbook pro. i tried using the openwhatever legacy patcher but it just doesnt want to work so im stuck with catalina and xcode12/swift 5. am i losing a lot of features? it's just for a course for now, probably some basic ui prototyping and whatnot, considering the last course i had to take was about flutter and it was pretty much just making a small app. they want me to come to the office to lend me a newer macbook but im too lazy to go so id rather use the one i have now

>>106588437
>too lazy to go so id rather use the one i have now
This is a (You) problem, I'd take the extra time to get the company laptop so I don't have to deal with constant bullshit.

>>106544071 (OP)
shit need to learn spring in speedrun mode but every course I see is like 30 hours long.. is this shit really that complicated?

>>106588437
mid ass

Canvas based UIs?

>>106590981
The way things used to be, before "everyone" became blind

>>106590981
webgl based uis

>>106590981
the elefant in the room is SEO, otherwise it'd be "the way"

>>106587468
I see. Personally I have no experience with either. I tried Three.js briefly and I'd be interested to try it again but I don't think I can justify spending time on it

>>106592258
It really just means that more CPU code can be moved to the GPU.
...and that programmers will have to learn the new api and shader language

>>106578091
>No. First connection is either HTTP/1.1 or HTTP/2.
That means that the 14kB rule still holds for HTTP/3 though.

>>106592686
I'll bite, what kind of thing would you liad in that weight, can it even have images or semi complex css gradients?

What have you done to get a substantial raise? I replaced everyone’s need for mailing services like mailchimp and whatever else that costs 100x more than its worth but looking for more ideas.

>>106576970
Made me laugh harder than I should have

>>106577004
>first connection to a server is never HTTP/3 anyway.
>>106578091
Orange reddit had a thread up where it said you can configure DNS to say youre server supports HTTP3 so the client can start with HTTP3
https://news.ycombinator.com/item?id=45236444

>>106585794
>>106586394
That sounds very noble and all, but some of us weren't lucky enough to have a suitable education when growing up and cannot reliably activate the reward circuit needed to keep doing this type of work, despite being intellectually engaged in it. Without Big Pharma, I'd either be compensating for this particular inefficiency with illicit substances, or have to abandon my career altogether.

>>106593567
Fair enough, if you find them helpful that's fine. I would rather avoid mind altering drugs if I can, but yes sometimes they might be genuinely helpful.

I will NOT use webp.

>>106593567
sounds like a shitty excuse, if I can do it so can you, these people are beyond evil

>>106594011
>sounds like a shitty excuse
I know, it's what I've been told all my life.
>if I can do it so can you
Just like if I can stay under 10% body fat by consuming 4000kcal a day for 6 months, so can you, right? Our brains and bodies are all the same anyway, we just need to do the exact same things as everyone else, right?
>these people are beyond evil
I buy Indian brand pills off darkweb. They benefit my life. That's the extent of my caring about some "evil".

>>106578091
Really? I thought one of the goals of HTTP/3 was to get rid of TCP altogether. It must be a stop-gap measure while HTTP/3/QUIC support isn't near-universal.

Can I code with one arm?
I'm about to have surgery for the staged accident where gangstalkers blew a stop sign and totaled my car

>>106594173
idk man, as the old meme saying goes, bite more than you can chew then chew it, if I told you something else I'd be lying, I had to grind too, like crazy

>>106594308
Answer your own question

>>106577084
>>106592686
>>106593519
>>106594173
I'm a tard. Just checked with Google and some Cloudflare protected sites.
HTTP/3 works for first connection.
It's my Caddy2 servers who answer in HTTP/2 and then upgrade to HTTP/3.

>>106593378
Google homepage is 52 KiB after compression, those 14 KiB should be for FCP, then you load the rest dynamically.

>>106594441
I was gonna say. Making HTTP/3 default to an older version for the initial connection would go directly against one of its main selling points.

>>106593846
Whta will you use instead?

>>106594441
google page and results are extremely minimalist, unlike pretty much any other website

>>106594534
Macromedia Flash Player

>>106594441
>HTTP/3 works for first connection.
Yes. It seems to be the case for Google at least.

>It's my Caddy2 servers who answer in HTTP/2 and then upgrade to HTTP/3.
Do you know how you get it to answer in HTTP/3 for the first connection?

But it seems like QUIC, which is used in HTTP/3, still recommends the 14kB rule.
https://datatracker.ietf.org/doc/id/draft-ietf-quic-recovery-26.html#section-b.1-2.2

>>106594608
top kek

>>106590981
draggable elements too

>>106594765
You gotta fix the dampening brah

>>106594765
impressive, how did u get there? that's a path most webdevs will never touch

>>106544071 (OP)
what are callbacks used for in the real world, just starting to learn about them and struggling

>>106594871
It's treating functions as variables.

>>106594880
when is that useful/used?

>>106594837
yeah, I switched from a geometric solution to pure 'physics' which has made it worse, but quicker resolve for overlap. I'll see about some kind of sort of zone re-centering or force amortization.
>>106594841
I started with canvas games. One interesting thing about this bubble thing is that the check for click is pixel color based which means it's something more like O(1) for any number of buttons and resolves Z-fighting visually with the user.
Here's my games:
https://pricklypeargames.itch.io/

>>106594885
Whenever you want to customize what happens without editing some else's code

>>106594894
that makes sense

>>106594919
The most common being Buttons.
The browser will call the function you supplied as a callback when one is clicked. Imagine if websites couldn't decide what happens when a button is clicked

>>106594919 >>106594941
someButtonElement.addEventListener( // I "call" the inbuilt addEventListener function
"click", // I supply two arguments to its parameters
jumpscare // I supply a function that addEventListener can call when something happens.
);

// Nothing happens until the browser calls me back.
function jumpscare() {
window.alert("Boo!"); // I scare the user with a popup
}

>>106595046
I've used this a million times, funny how that goes
chatgpt was giving me an example for "events" which click is, I'll try to dig more into this, sounds interesting

>>106584313
its because if they did what youre asking. thats harder on the cache and network.
they need to generate basically 2x as many images and thumbs as you see right now.
from your point of view it is a bad experience.
from their point of view its saving fractions of compute.

>>106595097
>they need to generate basically 2x as many images and thumbs as you see right now.
also that'd be dumb because consider that you just click the first result that you see

>>106595080
There's a lot of needlessly complex jargon for simple concepts. Fuckin' design patterns

>>106594628
>Do you know how you get it to answer in HTTP/3 for the first connection?
No, but I'll investigate once I'll have the time. I bet that it's because I haven't upgraded those Caddy binaries in 3 years.

>But it seems like QUIC, which is used in HTTP/3, still recommends the 14kB rule.
>https://datatracker.ietf.org/doc/id/draft-ietf-quic-recovery-26.html#section-b.1-2.2
I still don't understand why they use the 10 packets rule for QUIC.
It was used in TCP since HTTP/1.1 for a reasonable middleground between initial speed and reliability.
Since QUIC uses UDP, it could be sending all the packets in one go and then requesting the missing ones.

>>106594887
THIS NIGGA USING A NON MONO-SPACE FONT IN HIS IDE

>>106594837
I think I came up with an even better solution, I put a toggle in a pointermove event and it sets up main to run based on how much you move the mouse. Now you can cleanly move in bursts which makes the processing way less too because main just waits for the boolean and toggles it back.
>>106596099
Oh yes

>>106596572
>an even better solution
This is /agdg/ levels of delusion

>>106596630
it is interesting you would say that, but it doesn't address any kind of problem with the design and since you can't put any reasoning for any kind of argument forward I know that I'm right.
To a more direct refutation; I can still do a damping fix, but this also provides an enormous performance boost when idle.

>>106596572
https://dtinth.github.io/comic-mono-font/

>>106596721
decreased informational density and harder to read, but thank you for the suggestion.

>comic sans
>tacit confirmation
>>106596099
that's no nigga, it's a lost boy

>>106566589
same but as an older fart
PHP, AJAX, Vanilla JS, jQuery, then the horrors of Angular 1 then 2 and now diving into React I'm just numb to boilerplate hell, 1000 modules, and the fact that JS eats away all your foundations of good design

>>106595119
>also that'd be dumb because consider that you just click the first result that you see
also also, its loading up in the background, the video preview (with audio) of each of these videos. so that when u mouse over them all one by one they can quickly show the vid

>be first dev hire
>get friend hired year later
>he got hired at more than me for some reason
>find out everyone is paid more
>after a few years of percent based raises they all make like 8k more than me yearly.

I have an interview for a second job on wednesday.