[robenkleene]

An internet comment is a great example of a use case the browser is good for because it's low effort, low in specialized skills, and disposable.

Now we're talking about writing applications with web technologies for high specialized skill and high effort tasks, e.g., code editors, photoshop clones, etc... Which is what a lot of us are skeptical of. These "professional" applications are designed to reward a huge investment in muscle memory, if you are putting so much effort into learning an application, you don't want the return on that investment to be stunted by limitations in the technology the tool is built with.

[oceanswave]

Not following that last statement. There's no guarantee if I build my desktop application with XYZ framework that some technological limitation will crop up with it.

How much emphasis is being placed on JavaFX/Swing/AWT these days. Not so much. WPF? Microsoft has said that "...we are not investing in any major changes to WPF"

There is so much effort being placed in HTML/JS optimizations by WK, Blink, Edge folks that it's hard to find another framework with as much emphasis placed on it.

[robenkleene]

I'm talking about limitations in web apps that exist right now. For example with browser-based apps:

* Lack of data ownership, how do you export/backup?

* Poor keyboard shortcut support, e.g., constant conflicts between the browser itself and the app being run in the browser

With Electron based apps:

* Performance obviously, this might get better in the future, but that's a big bet of your time investment in learning and building on the tool. The fact that Atom is slow on a top-of-the-line machine is concerning. Sure that'll improve in time, but what if I also want to run it on more constrained device, tablets, etc...

* Electron apps still have a long ways to go to support basic platform expectations, e.g., on OS X Atom doesn't Services, AppleScript, or piping to the command line utility.