It's not even "faster software" so much as eliminating the culture of Developer Convenience at the expense of User Experience. That's what got us Electron. I've actually seen comments on HN unironically describing the web as "the perfect app platform".
Unfortunately, the majority of users seem conditioned to accept software with awful performance, so there's no impetus for developers to upgrade their skills.
It's not about "developer convenience". It's about "developers are expensive".
I can build an application in electron 4-10x faster (at least) versus building the same application in C. If I'm costing a company $100-200 per hour, would they rather pay me for 4 months (500 hours and $50,000-100,000) or would they rather pay me for 1-2 years (2-4k hours) at a cost of $200,000 - $800,000?
What about when we multiply that by a team of 5-10 people? Don't forget that time to market is often incredibly important. Tell them 2 years and 8 million or 4 months at 1 million and what will they say?
You might build an application faster in electron...
OTOH, C isn't a GUI development environment. If you want to compare a C based environment you compare it with GTK/QT/winforms/etc.
In the end, as someone who has written GUIs in a wide range of tooling i'm not sure there really is that much difference.
I've yet to see an electron application with 1/2 the functionality of similar native applications. Electron maybe gets you bootstrapped faster but then you bog down in basic data manipulation, and functional behavior because it turns out HTML/CSS/Javascript are absolutely terrible for building rich GUIs. Even now 20+ years after people first tried to do it. There are so many things people took for granted in the past (ex: grids with arbitrary sort, editing, and a scrollbar that represents where in the data you are) that are far more difficult in HTML than they are in more native solutions. Plus, the scalability is miserable, take your favorite framework and have it load 10k rows of data into a table. That was something you could do in VB/delphi in the mid 1990's on a 486 in a matter of seconds. This is why pagination is so popular. Half a meg of actual data bloats up into half a gig when you try rendering it in chrome/etc so your forced to leave it on the server and round trip for tiny bits.
It is because you want to hire inexpensive web developers to develop for desktop and get the check box ticked. Your average bootcamp webshit doesn't even know Big O notation. I don't expect them to be as productive as good developers either.
The web ecosystem is a big mess where trends change every month and working in web ecosystem requires looking up things a lot because no one bothers to master the thing. It doesn't help that many web developers don't have solid foundations.
This is a trap managers generally fall into. Cheap developers aren't equivalent to competent developers, and their incompetence will cost you more than what you save by hiring them instead of a competent developer.
Electron is Chrome and Chrome itself is pretty tightly optimised for what it does. The issue isn't actually Electron so much as using a platform designed for typesetting for making complex GUI apps, a task for which it was never designed and isn't particularly good at.
But. That said. Whilst I'm no big fan of web apps, there are good reasons they're so prevalent. It's not merely about developer convenience. Native GUI toolkits can appear artificially performant because they're required by the OS and thus almost always resident, vs cross platform toolkits that may be used by only one app at a time. When you open the lid though the gap between an engine like Blink and something like Qt, JavaFX or Cocoa isn't that big. They're mostly doing similar things in similar ways. The big cost on the web is the DOM+CSS but CSS has proven popular with devs, so native toolkits increasingly implement it or something like it.
Unfortunately, the majority of users seem conditioned to accept software with awful performance, so there's no impetus for developers to upgrade their skills.