I think we're really starting to see the saturation point with native apps, particularly on phones. I am so sick of having to install yet another mobile app, for something that basically shows me a bunch of text, pics, and like buttons. 90% of these shouldn't exist. There is a time and place for it, but I think there is going to be a shift to the web app with more OS integration stuff.

This has been happening for a while. This 2012 article from Nielsen [1] wrote that while app downloads were up (in 2012), user time spend in all apps was flat. It also notes that the 4 of most popular apps were Facebook, YouTube, Google Search, and Gmail.

This 2014 recap [2] of a comScore report notes that more than half of all smartphone users download 0 new apps a month -- this started the catchphrase that "the average user downloads 0 new apps a month", which has both been supported and challenged by additional analyses on old and new data every year since. Whether it's 0 or 1.5 [3] isn't the point -- the growth is measly.

Aside from a select runaway hits, most of today's most used apps are the same as 4, 5 years ago. The app boom has been over for a long time. Occasionally, one can still strike it rich, but it's the exception, and not the norm.

[1] http://www.forbes.com/sites/parmyolson/2014/02/19/exclusive-...

[2] https://www.quora.com/How-was-WhatsApp-built

[3] https://www.tune.com/blog/no-the-average-american-does-not-d...

Agreed! Thanks for the sources.

We've got the worst of both worlds a lot of the time. Tiny little apps that would be just fine as web pages, and web/electron apps that should be real apps.

Aren't most of those "native" apps just webapps packaged standalone with some framework anyway?

Not really, not the majority of the ones that people really care about and don't drain their battery.

Even those that are, they can access the full spectrum of OS APIs, instead of a document model bended to be a VM.

Just furthers my point.

Hard to tell, Android Instant Apps and the like may solve those problems and keep native apps ahead of web apps.

> Feels like this has been 2-3 years out for almost as long as "native apps" have existed

Yup. I've been reading about the imminent demise of native apps forever, but even Steve Jobs couldn't sell that vision. Web apps on mobile are the way of the future, and always will be.

It's not really the demise of native apps. I think it's almost undeniable that a percentage of native apps should be websites, it's just a question of whether that's going to end up as 30% or 80%. Seems unlikely it will ever be 100%.

I agree but there really has been a lot of progress lately. Service Workers allowing offline, web notifications, and just today I read that Chrome webapp will appear in Android as if they are full apps.

The part I'm still concerned about is UI performance - it's still miles off native.

> The part I'm still concerned about is UI performance - it's still miles off native.

Many times needlessly low expectations of both developers and managers are the blame for this. Yes, there are certain elaborate UI effects that are genuinely difficult to do well on web. But when there's a full second lag after pressing a button, or a menu slides into view at 3fps, the platform is not the problem.

A native app that has incredibly choppy, laggy UI for simple interactions wouldn't make it past many engineering managers or PMs. But when there's needlessly a full second lag after pressing a button on a web app everyone shrugs, inaccurately blames it on the web platform (instead of piss poor engineering), and it's out to production furthering the stereotype that web apps are inherently jank.

As a web development community we need to stop this cycle. Web engineers need to stop building, and engineering managers need to stop approving, web apps with needlessly janky, laggy UI for even rudimentary interactions.

In addition, a portion of the cross-platform native development toolkits use bottled web app as their approach anyway.