The native app was by no means perfect, but it felt like a real productivity tool that was trying to be respectful of it's environment.

I've come to the conclusion that native desktop apps are just not viable from large companies, even if there is headcount. The problem is coordination cost.

If you want to launch new features and experiments here, there and everywhere, then the coordination complexity increases nonlinearly with the number of platforms.

If you can sustain a more deliberate, low churn pace of development then it's workable. Features can be well defined and then implemented by the platform team as they see fit. But if you want a more fast-paced, "just in time" style of development, you need to coordinate with every team for every change... wouldn't it be nice to just write web code and be done?

Even Microsoft are building this way these days.

This is why ironically small companies seem more able to support native apps than large ones. The more "stuff" that's being worked on concurrently, the harder it is to support multiple platforms.

You may not like that from a 'native look and feel' point of view, but the question 'what is a native Windows app these days anyway' is very much unanswerable, and you can actually implement stuff like this in a performant and offline-sensitive way.

But, yeah, by the time the resulting GPU worker process balloons up to 400MB, that pretty much goes out of the window. I'm actually sort-of impressed, in that I have no idea how I would even make that happen! But that's why I don't work at a powerhouse like Meta, I guess...

The new electron app does take more resources, but at the very least it works.

We often hear stories about the speed of development and the issues of maintaining native apps, and then there are these rewrites every few years. Don't they waste more resources vs. creating / fixing the gaps in the native app? And this isn't somes quick startup prototype app that can flop and the effort would be wasted

Take Microsoft for instance, they have been pushing heavily what they called the New Outlook, which is basically a web-based client mirroring Outlook on the web (OWA) , packed into an EXE file (not sure if it is Electronized or not)! Then, they renamed the real native Outlook app as Classic Outlook to feel old-fashioned and outdated and as result, we ended up losing some core features that made Outlook, Outlook. We lose: COM Add-ins and VBA Macros, MAPI Support, Word as Email Editor, .PST File Support.. to name a few..

This would be probably one of the reasons contributing to the Collapse of Civilization (https://news.ycombinator.com/item?id=25788317)

Meta makes more money than god and there's over a billion WhatsApp users. It's not like this thing is Blender or a AAA game, it's a chat frontend. Maintaining it has to be a rounding error in the budget.

Yet, I really don't understand why WhatsApp would need app especially with the state mentioned here (which is a basic wrapper)

There are no calls in the web app, but modern web stack is more than enough to provide all the real functionality needed for it.

There is essentially no way to tell if a JS app is using a lot of memory just by looking at what the process has reserved. There's loads of things that end up in that space - cached pages, cached compiled code, cached bitmaps of rendered pages, etc.

The task monitor tells you what Chrome or Chromium (e.g. Electron) is doing, not what the web app is doing.

There is a good argument to suggest Chromium is hogging more than it should. That's not really WhatsApp's fault though.

Because just parsing html/xml/some declarative ui description to create a scenegraph and pushing it to the GPU is something I'm sure even native frameworks like Qt do.

Yet I can feel the latency in vs code vs. say sublime text.

It's tempting to blame the huge js blobs trying to create an application abstraction over a document object model but i feel like there should be more to it than just that...

(disclaimer, I help out a little with that plugin, amongst others for Pidgin-

I wonder if they avoided that so they could use Electron and target MacOS / Linux too

I’m forced to be in parent group chats in this accursed ecosystem and give Zuck another DAU.

- slim down to a more efficient, non-virtual-DOM web stack (e. g hypermedia)

- move to a wrapper around native webview (Tauri)

- use one of the excellent cross-platform frameworks (Flutter, or that new one just open sourced by Snap)

For WhatsApp specifically, I don't understand why a company with the size and resources of Meta can't support native versions for Windows, Mac _and_ Linux. I think many people would accept non-feature-parity with the web/mobile versions, in exchange for a tight, reliable desktop messaging app...

There are far too few people who truly advocate for the user, and it is this dereliction that has fuelled the race to the bottom.

It's not a matter of native libraries versus x-platform solutions versus W3, it's about valuing UX over DX at EVERY TURN. It's about educating yourself as to the resource and performance consequence of the technologies you are advocating for.

It used to bundle MSN, ICQ, IRC and everything in one messenger - super resource efficient.

Unfortulately I cannot reach many contacts that way these days but it showed me how inefficient applications became by using more abstraction layers.

Same on my Mac - it uses almost 16GB of RAM in idle mode with some Tabs, VSCode and Figma open - How did we get here?

edit: Just saw another comment mentioning there is a WhatsApp plugin for Pidgin. Awesome!

Well sounds like a lot of useless work was being done then, how does it gobble 100MB when idle? Are the protocols that complex?

Just do as I do and open web.whatsapp.com in your favorite browser

meanwhile, the web-based version in the browser was nearly instant.

I should try the new web-based app to perform benchmarks, but I think I'll just keep using the browser.

I'll also mention, the "whatsapp web" website is extremely resource intensive, and in my laptop it makes the CPU usage ramp up to 100% very often, and using WhatsApp while having any other programs or even browser tabs open becomes unbelievably annoying. the desktop takes 700MB, the web browser takes 400MB, and WhatsApp web takes at least 1GB alone. this is on Arch Linux, with 4GB of DDR4 RAM.

And now it seems that there is not even a benefit to installing the native vbersion, they will probably do the same for Mac also now.

I installed omarchy and realized that yes, all the apps I use except the terminal, are browser apps, so I don't need a desktop environment at all.

I was also running a mIRC client in rooms with hundreds to thousands of people at once.

And I was able to do both at the same time.

Now that computer can't even run 1v1 chat.

No 1 GB or installation needed

Why is the desktop app even a thing?

All this “engineering” from Meta and algorithm acrobatics with hundreds of optimization puzzles and they accept a solution that performs completely far worse in both runtime and space complexity.

Whoever believes that this is the acceptable solution from a trillion dollar company would have failed their own interview.

No company would accept a huge regression like that and push it to billions of users but Meta.

Total amateurs.

I was very happy with that situation - it was the last and good enough reason to uninstall all FB-related apps from phone, and never looked back. That company (and not only that) is a cancer to whole society, as per design.

Whatsapp screams antitrust. If you look in the dictionary for antitrust, you see Whatsapp

https://news.ycombinator.com/item?id=1692122

At least look back at the color of the other grass!

Telegram has its own faults and issues, but the native Windows app is incredibly good and fast.

You know I heard something funny the other day, someone saying that now thanks to Claude Code, companies might go back to doing actual native support because the specialized knowledge and effort have both been basically automated now.

Well, give it a few years!

A company's goal is to make money by optimizing its resources. What benefits would Meta gain by maintaining native apps for WhatsApp across the three major operating systems? I can tell you: absolutely none, only negatives. Nobody except a negligible fraction of users would care about native performance or idle memory consumption. No one is going to switch to Signal or whatever the flavor-of-the-year messaging app is because of this.

It would be a different story if WhatsApp were to lose a significant portion of its user base due to the app becoming unusable or extremely slow. But for the vast majority, this change will go unnoticed or frankly won't matter at all. So, expect most companies to continue adopting Electron-like apps (for the few that still have native apps anyway) for exactly the same reasons.

Sorry to be blunt, but it's really tiresome to see these discussions going around in circles here. It’s pointless to keep debating this, it's not going to change. If one day a framework emerges that's comparable to Electron (or something similar) but requires fewer resources to develop against, I could see Meta and other companies considering it... provided the migration costs aren’t too high. But again, no end-user truly cares about this.

Another example of WinUI anemic state.

Must be a tiny percentage, which is why this version is now a basic web wrapper now.

Anyway, I’d remind everyone that “using” RAM doesn’t mean “would not function with less RAM.”

Many applications just use a lot if it’s available.

RAM is not really something you explicitly ration.