WhatsApp Kills Off the Electron-Based Desktop App (androidpolice.com) 37
WhatsApp has announced it is retiring its Electron-based desktop app, forcing users to switch to the native app for their OS to continue using WhatsApp. Android Police reports: Back when WhatsApp was in the early stages of development, the developers created an app for desktop, based on the Electron JavaScript framework. This allowed them to share a code base between WhatsApp Web and the new, platform-agnostic desktop app that worked on both Windows and macOS. Around four weeks ago, a countdown timer showed up on the main screen of this desktop app, announcing its shutdown.
Doomsday is now here and WABetaInfo reports anyone visiting the Electron-based app just sees a screen saying "App expired." The deprecated app helpfully links to the native WhatsApp Desktop app available on the Microsoft Store or the Mac App Store. The new native app has been stable for around a year now, but is still relatively new. Some users may lament the transition period was too short, or the native app still doesn't have all the functionality for business users, like catalog management and quick replies, and they would be right.
Doomsday is now here and WABetaInfo reports anyone visiting the Electron-based app just sees a screen saying "App expired." The deprecated app helpfully links to the native WhatsApp Desktop app available on the Microsoft Store or the Mac App Store. The new native app has been stable for around a year now, but is still relatively new. Some users may lament the transition period was too short, or the native app still doesn't have all the functionality for business users, like catalog management and quick replies, and they would be right.
Downsides.. (Score:5, Interesting)
Re: (Score:2)
That would be an incredibly dumb assumption. Electron is quite capable in hooking into the OS just like any other app. Just ask the MS Teams developers whose shitty electron app can't even be removed from the damn OS.
Re: (Score:2)
..or so they say.
Remember IE6 and how it was also "impossible" to remove it from the OS?
Re: (Score:2)
I didn't say MS can't remove it. End users can't.
Re: (Score:3)
Not sure about Windows, but on the Mac side, being in the App Store means they have to abide by sandboxing provisions and possibly some other things of which I’m unaware. If there’s a noteworthy difference, I’d expect the Mac version, at least, to be less invasive.
Re: (Score:3)
There doesn't appear to be a native Linux client.
Re: (Score:2)
I've been using the native app for almost a year n (Score:2)
I've been using the native app in beta since last year, it's fine, don't really think it's missing something I'd use in its current form. It was a bit rough when I started using it, but it caught up quickly.
I had to switch to it. The electron app was a hog before, but when I switched to an M1 Pro and everything else went super fast, that app under Rosetta emulation slowed to a crawl. It took over 30 seconds for that crappy messanger app to load!! This, on a machine where office, Photoshop etc take a couple
Electron is crap (Score:3, Interesting)
I don't know if their native app is any good, but electron-based apps are fundamentally inefficient. They always have run-away memory problems and use more processing power than needed, due to using a browser engine and web technologies to render it, it has the same problems that web browsers have. It should only be used as a stop-gap to bring an app to an OS faster, not as a permanent solution. Unfortunately developers are often too apathetic to wasting users computing resources.
Re: (Score:2)
the extra overhead gives you cross platform, which is actually a huge deal, and in the vast majority of use cases and environments electron performs more than well enough for purpose.
by the same logic we should have never used java (oh wait), nor lua, nor any windows library but straight win32 api calls from c ...
honestly, unless their app was unmaintainable spaghetti or there is some other good reason they're not telling us, developing these 2 native clients will have been a waste of effort and resources a
Re: (Score:1)
I know that's the reason it's used, a quick way to bring cross platform support. But the sacrifice is too great. Every electron app I've used have been resource hogs and runs like crap. They all eventually have to be terminated and restarted after prolong use. They basically perform the same as web browsers which have the same problems.
Sure any language and framework can be misused to create poorly developed programs. But unlike those that you mention, it's fundamental to the underlying web technologies tha
Re: (Score:2)
i disagree, and my experience is very different. e.g. visual studio code and even atom run perfectly on my desktop. i still use atom instead of notepad for quick edits. it's very fast. before you ask, my desktop is old enough for windows 11 to refuse to install (which is perfectly fine for me). visual studio code runs without a hitch on my laptop, which runs linux on an i3.
also, i think you have an outdated view on browsers in general. i've personally developed several clients that run inside browsers, even
Re: (Score:1)
I think you misunderstood, I didn't say java or visual studio or whichever tech you mention has the same flaw as Electron does.
And I'm not saying all web tech are bad and inefficient. But the modern web has accumulated too much cruft, and a widely used subset of that is very inefficient and causes websites and apps to be resource hogs. I'm sure if we carefully select and cull web tech and only use the parts of it that are well-designed, we could cut it down into standards that won't have these flaws. But th
Re: (Score:2)
i think i should clarify, maybe my disregard of capitalization might be to blame, but i'm unrepentant in that regard: i just cited "visual studio code" as a form of "quod erat demonstrandum". it is a hugely popular source code editor from microsoft, an excellent tool and also ... an electron app.
now, this shows that the tech is perfectly viable to say the least. and that alone just flat out contradicts in practice everything you just said from ... some unprecise theoretical context. much like a regular unin
Re: (Score:2)
Unfortunately developers are often too apathetic to wasting users computing resources
Yeah count me as one of them.
Every single kind of cross-platform development system is less efficient than a true native app. This is because, in all cases, they require a translation layer between the native APIs and a set of common APIs exposed by the platform. Electron is no different.
If you don't have the money or time to develop separate native apps for every platform, then a cross-platform toolkit can be a good option.
Electron has an advantage over many of these systems, because it also supports web (
Re: (Score:3)
Really, because I've written QT apps that will run on a 1Gb ram Raspberry PI that do more than most big company apps will in a lifetime.
Absolutely nothing about "Cross platform" requires innefficiency and the fact we've had 50+ years of efficient cross platform apps is a testament to that.
Just don't expect wrapped web page apps to be in that category.
Re: (Score:2)
Just because it's possible to write relatively efficient QT apps, doesn't make them as efficient as a native app. 50 years ago (as you pointed out) an "efficient" app had to fit within 64K or less of RAM, including the OS. In that context, your 1 GB QT app isn't so amazing.
Re: (Score:3)
Not that the app was 1GB.
You can't run a 1GB anything on a 1GB Pi.
Re: (Score:2)
And I don't think it's a solvable problem, all modern browsers are resource hogs and no one has been able to solve it. And it's not due to the lack of competency of all browser developers or electron-based app developers. They can't fix problems that are fundamental to the tools they use.
I know ease of cross-platform is the selling point, and I acknowledged that. There's a spectrum of languages and toolkits and frameworks with various trade-offs, and my argument is that Electron and those like it sacrifice
Re: (Score:2)
Browsers basically have to be game engines and virtual machines packed into one thing.
Re: (Score:2)
Re: (Score:1)
True in a way, but browsers and the modern web is a 2-way street, it's a feedback loop. Browsers adapt to what the web needs, and the web implements what is possible and provided by browsers. And that feedback loop has brought us to a very inefficient and wasteful web.
IMO good software tools is as much about what it provides as what it restricts. Giving you the tools to make poor software too easily is a legitimate criticism. For example buffer overflows and memory exploits being too easily created is a leg
Re: (Score:2)
The fact that browsers are all resource hogs, doesn't mean the problem isn't solvable. It means the problem isn't a high enough priority to the browser makers, to invest the time and money to solve it. Performance tuning is a science, it is definitely solvable, but it's not cheap. On the other hand, memory *is* cheap.
Re: (Score:1)
While I agree that technically it doesn't mean it's impossible, we do have to let the results speak for itself. It's like a lot of other ideas and concepts, even if theoretically it's not impossible, if all the cases and every time we tried it has the same flaws, then we have to acknowledge that. Science is based on observations, and if our observations keeps showing us the same results, then it means something.
Re: (Score:2)
Of course. This lack of "fixing" browser memory usage indicates to me that the problem isn't severe enough to justify the expense. It says nothing about the feasibility.
It's kind of like how we all howl about lack of respect for privacy on the web, but we all keep using Google and Slashdot and all the other "free" web sites anyway, because if we're honest, most of us would rather give up our privacy, than pay money for privacy.
We can't track you well enough on the desktop (Score:5, Insightful)
Re: (Score:2)
I can't imagine why a desktop app would be restricted in any way, from tracking you as well as WhatsApp wants to track you. There is no inherent tracking limitation in desktop apps. In fact, in general, they leave you *more* open to tracking because they have access to operating system functions.
Re: (Score:2)
Re: (Score:3)
Interesting take. The entire browser is a kind of sandbox that prevents you from accessing most OS functions. Maybe it's different on Mac, but on Windows, once an application is installed, it can pretty much do what it wants to.
Re: (Score:2)
The entire browser is a kind of sandbox that prevents you from accessing most OS functions.
The browser may have been but Electron was not. It used the browser to render but had the functions of any other app. This is why MS Teams is such a god awful hooked into the OS unremovable piece of shit despite being written in Electron.
There are no new capabilities being provided here. But thank Christ there's one less Electron POS running on my system. It shouldn't take hundreds of MB of RAM just to display a chat window.
Re: (Score:2)
It used the browser to render but had the functions of any other app
This isn't quite true. Electron uses the browser to render, but it exposes only specific functionality to access the underlying OS. By contrast, a native app has access to ALL native OS functionality.
Re: (Score:3)
So please install our app instead
At least on the Mac, a desktop app in the app store will have many restrictions a downloaded app won't. It will be sandboxed for security and privacy, and it will need to divulge which data it collects and for what purpose by submitting "Privacy Nutrition Labels".
Re: We can't track you well enough on the desktop (Score:2)
I've written a Proton Mail client using Electron. (Score:3)
Using WhatsApp can get you killed. (Score:3)
Everyone is better off using Signal [signal.org] instead. Signal has a nice desktop app that also works on Linux.
TIL There's an Electron Client (Score:2)
Windows app (Score:1)