It’s really a shame that all of these unofficial clients are like this. APIs are APIs, if they’re on the public internet why can’t anybody call them? If there’s potential for abuse then that should be secured server side anyway, since needing a unofficial client isn’t going to stop hackers
Eventually the unofficial client gets "fingerprinted" and its over. Its easy to remove the suspension, though. I vaguely remember it was through email.
Eh, there's no humans any longer IIRC. It's all down to whether their support AI blesses you or not.
I got banned for trying to move my chats from Android to iOS using the app rather than the special tool I didn't know you were supposed to use. I was lucky I eventually got unlocked.
I want a decent WhatsApp client that doesn't crash constantly and where you can search properly, but it's not worth the risk of getting banned again.
I actually wish they'd delete the warning so then anyone who gets banned from WhatsApp for using an unofficial client would have a more legit cause of action against WhatsApp. Meta wouldn't be able to defend themselves with "well you were warned" if you weren't warned.
I think this doesn't apply to European users, as WhatsApp announced around one year ago that under Digital Markets Act they were required to allow interoperability with third party providers [0].
Egui is pretty bad idea. It uses "immediate mode" rendering which means whole window is redrawn on every mouse move. I ran PhotoCraft (built in Rust + egui) in a VM wihout a GPU and the UI felt like it was running on 3 fps.
Egui is definitely not what you want to use on a 4K screen. It is only good for tiny screens of embedded devices.
Why there is no reactive UI that tracks all dependencies and redraws only what needs to be redrawn? (not React, it also re-calculates (but not redraws) everything on every change).
> Why there is no reactive UI that tracks all dependencies and redraws only what needs to be redrawn?
Nowadays the focus seems to be on video game-style interfaces that redraw every frame. The justification for this is typically that computers are so fast they can now come up with the entire interface every frame, so why not take advantage of that and forget all this retained-mode complexity?
They don't realize that this is a classic case of the developer's machine fallacy, where the developer has a machine so vastly more powerful than any of their users' that they produce a terrible experience for said users. This is also why games these days typically require upscaling even on 5090, because developers usually have that or something even more powerful and they think the bleeding edge is pushing even THAT machine to its knees.
Developers should be testing their software on the most entry-level, bottom of the barrel Chromebooks from 2015, running the heaviest and most bloated operating system they can think of. If people still optimized their software for that, then maybe computers would still feel almost as fast as the Apple II from 1977.
Wrong. First, even if every device has a GPU, not every GPU supports Vulkan and other modern standards, and not every GPU supports executing code. Second, drivers are often buggy/unsupported. And not every OS has support for graphic acceleration. Third, even if there is a GPU, it is often very slow to compile the shaders and the UI must be rendered on CPU during this time.
Finally, for app to have access to GPU, it must run on a host unrestricted which is just stupid. Why would you take somebody's code from Internet and run it with full privileges, violating the principle of minimal pivileges? That does not sound smart, especially if you have a crypto wallet on the same computer. Those who do not use VM or sandboxes will sooner or later learn this.
> Did you measure anything in a real setup?
VM is a real setup. Why would I run third-party apps without a sandbox? I cannot afford buying a dedicated laptop for every app.
Also, today we have an advanced AI that is capable of solving science problems. Should not be difficult to build a GUI that works fast without graphic acceleration, right?
> Why there is no reactive UI that tracks all dependencies and redraws only what needs to be redrawn?
There's plenty of them. Slint, Floem, iced, Vizia, fltk etc. Slint and fltk in particular also exhibit excellent performance and very little resource usage even without a GPU.
I vibed some code using ImGui and it had this problem. I made it so that it only redraws when data changes with a fixed refresh rate and it went away. So it's not a permanent state of affairs.
dont get me wrong, I like the extra performance, but every single alternative whatsapp client has the same issue where it's missing enough core features that you end up needing to go to the browser version often enough to be a nuisance and I end up just using the browser instead (which I hate, I really really would prefer a performant client instead!)
$ cat $(which whatsapp)
#!/bin/dash
# A kiosk is full-screen by design. You need to go to about:config and toggle
# full-screen-api.ignore-widgets to true to prevent our window manager from
# creating an actual full-screen window.
# Additionally, you need to set the name to "kiosk" which will set the app id
# under sway to "kiosk" so that in sway you can disable full-screen on
# startup.
firefox --name kiosk -P whatsapp --kiosk https://web.whatsapp.com/
Also, I looked at how dates are written in a screenshot and thought that "2 days ago" would be so much better than "Tuesday". Because you do not have to compute in your head how many days ago Tuesday was.
Sounds reasonable, but IME apps that use relative timestamps switch to absolute time past certain threshold, and then I end up with two adjacent entries "14:38” and "15 minutes ago" and scratching my head.
I hate relative timestamps, I find it much easier to remember specific week days, so I very much prefer this style so I don't have to compute in my head which day of the week I got the text.
Anyway, I tried this application and it seems to mix different styles depending on how long ago you got the text: "yesterday", "Friday", "18 Oct 2026".
For an easy-to-read chat list, I chose in a recent tui application to adapt to the age of the message and write only the hour for today's messages, "Yesterday" for yesterday's messages, the week day for 2 to 6 days messages (so this couldn't be ambiguous), month+day for previous messages, year+month+day for last year+ ones. Works pretty well!
That's the annoying thing with WhatsApp, Slack, Discord, etc. Let's say I've got a personal client (that I'm definitely not using...) which works so much better than the official one. I cannot distribute it or use it according to their rules. It's forced enshitification, even though we have better options.
On the good side though - if you're replicating the official behaviour close enough... maybe nobody will know. But then you have to make sure everything, including key sort order, and JA4 of requests matches. And that's tough...
> In our Linux test it opens in under a second and uses about 200 MB of idle RAM
Is it just me or does anyone else feel that 200 MB is a lot for a native chat app written in Rust, that's essentially a remote interface for the WhatsApp on your phone?
I remember back in the day, apps like Miranda IM (and early versions of Trillian) used only digits of RAM, even when you had multiple plugins talking several protocols at the same time, whilst having a ton of advanced features not found in WhatsApp. Yes I get that times have changed and software has become bloated etc, but 200 MB still feels a bit too much.
> Is it just me or does anyone else feel that 200 MB is a lot for a native chat app written in Rust, that's essentially a remote interface for the WhatsApp on your phone?
> I remember back in the day, apps like Miranda IM (and early versions of Trillian) used only digits of RAM, even when you had multiple plugins talking several protocols at the same time
I think what's happening in the second one is that because you're using the native OS toolkits a bunch of stuff is shared, whereas with these "self-contained" "native"-y apps you end up having to load up the whole world and aren't sharing things with everything else
I'd love to be proven wrong but I feel like if you use e.g. Qt you're gonna have to hold onto a bunch of stuff that you wouldn't if you were "just" using OS-native stuff (though maybe that gets shared if you're using the exact same version? Is there something that can be done in this space?)
As an additional data point, I recently vibed an egui wrapper around a library I wrote for mod management for a game, and testing it out now, it idles at 92 MB. This implies to me that at least half of the overhead is probably not from the gui framework, but I also don't know how much effort they spent trying to optimize it; I didn't try to optimize my vibed GUI app at all, but I did hand-write the library that it uses for file management, state tracking, etc., and I tend to go out of my way to try to avoid unnecessary allocation in my Rust code. I don't know how much of the memory usage is essentially required overhead or from other dependencies for non-GUI parts of the app, but my instinct is that there's probably a bit of optimization that can be done. I don't think it's crazy for them not to have spent the effort on trying to do that yet though (or potentially even at all, if it's efficient enough for their own personal use).
I just checked with the native WhatsApp client on Windows 11:
650 MB of RAM as it is running, across all its processes. Thus, taking ≈⅓ of RAM is nothing to brag about for a reimplementation that focuses on being lean and fast.
Also, "opens in under a second" – back in the days when SSDs where new, programs opened instantaneously (<60 ms)!
Lastly, I still have to be convinced that Rust is a good choice for GUI programs, considering their messy mutable state and ownership.
What purpose would that serve? You have the source code. In the old days we’d just share tarballs. No provenance. It’s fine. You have the unit of legible modifiability.
There have definitely been times when I've cloned repos that were vibed, pointed my own agent at it, and then hacked up whatever changes I've wanted. I don't think there's been a single time that a lack of prompts has caused any issues, but there have been at least a few times where stuff like AGENTS.md checked into the repo has caused hiccups from them encoding stuff that conflicts with how I want things done.
Matrix bridges for sup, signal and the rest are good enough to never use the web client anyway. You can also skip the part where you need to have it installed on the phone by running waydroid (if you can figure out how to scan qr codes with it).
70 comments
[ 0.54 ms ] story [ 10.5 ms ] thread> Using an unofficial client may be against WhatsApp's terms of service and could get an account suspended.
Same for all of these unofficial clients.
I got banned for trying to move my chats from Android to iOS using the app rather than the special tool I didn't know you were supposed to use. I was lucky I eventually got unlocked.
I want a decent WhatsApp client that doesn't crash constantly and where you can search properly, but it's not worth the risk of getting banned again.
0: https://www.whatsapp.com/legal/dma-notice-non-users
but you shouldn't trust these unofficial clients, easy way to get your whatsapp account banned
Egui is definitely not what you want to use on a 4K screen. It is only good for tiny screens of embedded devices.
Why there is no reactive UI that tracks all dependencies and redraws only what needs to be redrawn? (not React, it also re-calculates (but not redraws) everything on every change).
Nowadays the focus seems to be on video game-style interfaces that redraw every frame. The justification for this is typically that computers are so fast they can now come up with the entire interface every frame, so why not take advantage of that and forget all this retained-mode complexity?
They don't realize that this is a classic case of the developer's machine fallacy, where the developer has a machine so vastly more powerful than any of their users' that they produce a terrible experience for said users. This is also why games these days typically require upscaling even on 5090, because developers usually have that or something even more powerful and they think the bleeding edge is pushing even THAT machine to its knees.
Developers should be testing their software on the most entry-level, bottom of the barrel Chromebooks from 2015, running the heaviest and most bloated operating system they can think of. If people still optimized their software for that, then maybe computers would still feel almost as fast as the Apple II from 1977.
https://danluu.com/input-lag/
Basically every device has a GPU.
Why is it bad on 4K?
Did you measure anything in a real setup?
Finally, for app to have access to GPU, it must run on a host unrestricted which is just stupid. Why would you take somebody's code from Internet and run it with full privileges, violating the principle of minimal pivileges? That does not sound smart, especially if you have a crypto wallet on the same computer. Those who do not use VM or sandboxes will sooner or later learn this.
> Did you measure anything in a real setup?
VM is a real setup. Why would I run third-party apps without a sandbox? I cannot afford buying a dedicated laptop for every app.
Also, today we have an advanced AI that is capable of solving science problems. Should not be difficult to build a GUI that works fast without graphic acceleration, right?
There's plenty of them. Slint, Floem, iced, Vizia, fltk etc. Slint and fltk in particular also exhibit excellent performance and very little resource usage even without a GPU.
Signal does that and I hate it when it happens.
- Screenshots don't age well: you need to know the screenshot time and do math
- They need to be maintained, which means the screen keeps updating even if nothing else is happening
It is silly to mix the two. I usually display the relative time and offer a timestamp on hover or if there's space, show both.
Anyway, I tried this application and it seems to mix different styles depending on how long ago you got the text: "yesterday", "Friday", "18 Oct 2026".
On the good side though - if you're replicating the official behaviour close enough... maybe nobody will know. But then you have to make sure everything, including key sort order, and JA4 of requests matches. And that's tough...
Is it just me or does anyone else feel that 200 MB is a lot for a native chat app written in Rust, that's essentially a remote interface for the WhatsApp on your phone?
I remember back in the day, apps like Miranda IM (and early versions of Trillian) used only digits of RAM, even when you had multiple plugins talking several protocols at the same time, whilst having a ton of advanced features not found in WhatsApp. Yes I get that times have changed and software has become bloated etc, but 200 MB still feels a bit too much.
> I remember back in the day, apps like Miranda IM (and early versions of Trillian) used only digits of RAM, even when you had multiple plugins talking several protocols at the same time
I think what's happening in the second one is that because you're using the native OS toolkits a bunch of stuff is shared, whereas with these "self-contained" "native"-y apps you end up having to load up the whole world and aren't sharing things with everything else
I'd love to be proven wrong but I feel like if you use e.g. Qt you're gonna have to hold onto a bunch of stuff that you wouldn't if you were "just" using OS-native stuff (though maybe that gets shared if you're using the exact same version? Is there something that can be done in this space?)
$ ps -o pid,rss,command -p $(pgrep -f SimpleQtApp)
PID RSS COMMAND
1471929 33052 ./SimpleQtApp
// Following is the code.
#include <QApplication>
#include <QWidget>
#include <QLabel>
#include <QPushButton>
#include <QVBoxLayout>
int main(int argc, char argv[]) {
}That was my first thought. But considering it as a significant improvement from > 1 GB, I guess it is what it is.
> I remember back in the day, apps like Miranda IM (and early versions of Trillian) used only digits of RAM ...
Sadly this was true for many apps.
Turns out all those zero cost abstractions add up to a lot!
Is there actually a way to see a breakdown of memory usage for a program? If you want to see where all the memory is going?
Flame graphs are a pretty common tool for this (maybe slightly less common than using them for seeing the breakdown of the time spent when running).
Lastly, I still have to be convinced that Rust is a good choice for GUI programs, considering their messy mutable state and ownership.
In the early days of AI coding someone did publish the prompts and it was utterly useless to me https://github.com/i365dev/LetterDrop
The source code itself is a sufficient prompt.