Beyond this, it is clear to me when an iOS-minded developer creates a web app that ends up on Android. Each have their own specific styles, and this gets lost with web apps.
WebSockets changed wire formats multiple times before we settled on something. Last I checked, the API still isn't final, and changes in browser behavior forced a project I was on to abandon it for socket.io. Web Audio has been through at least one breaking rewrite since 2012, and still isn't finalized.
Sure, '<h1>' still gets me a big, bold heading (probably, subject to CSS), but playing a sound file without plugins is still bleeding edge, and subject to arbitrary breakage. But we've broken plugins (for the best, eventually), so the old solution to that problem no longer reliably works.
No, it rarely removes deprecated features.
The market has spoken. Doing those things is hard enough, people can charge lots for it.
That sucks. "Sucks" == suffering. Suffering == opportunity. (Deliberate Yodaism. Actually high market price == market opportunity.)
So there is a market, people are trying to fill those gaps, and yet none of them seem to be able to bridge the gap in such a way that devs prefer maintaining two or more native apps over one that compiles for multiple devices with a common framework.
So there is a market, but it doesn't look like it's an easy problem to solve.
Really? More unaffordable and unavailable than at least one native developer for each platform you're targeting? I find that hard to believe.
The entire article is saying "I used to believe it was too expensive to do native apps so we did web, I now believe I was wrong, even though it may seem cheaper at first the costs of the web are far less predictable and it's now economically better to just do native apps"
I've been writing web based applications for about 20 years now. It's a lot better than it was in the late 90's and early 2000's, so stop bitching or get off the lawn. (sarcasm) Seriously though, I'd much rather deal with webpack, babel, react, redux and the like than angular 1&2 (just feels like a decade old solution now), and heaven forbid having to ever look at an <ILayer> again, or breaking up forms dynamically in the v4 browsers.
Yes, there are a few features not broadly supported, but there is also tooling that makes web dev as good, and in some ways better than app dev (though less than natural, and all out performance on a complex app is hard)... all of that said, it's a VERY valid option for many/most sites/applications.
That's the killer.
They're saying it shouldn't be done, or that it's possible and if you can't do it you're a lame developer. I didn't see it as condescending at all.