Since I started writing web applications in around 1999, there’s been an ever-present boundary around what you can do in a browser. As browsers have improved, we have a new line in the sand. They’re more enabling now, but equally annoying.

Applications running on servers in a datacenter (Ruby, Python, PHP) can’t:

Browser applications can’t:

It may seem like native applications are the way to go. They can do all the things browsers cannot. But!

In practice, this means there’s a step before building an application where I figure out where some functionality lives. “This needs to think real hard, it goes on a very specific kind of server. This needs to store some data so it has to go on that other server. That needs to present the data to the user, so I have to bridge the server and browser. But we’d really like to put this in app stores soooo, how are we going to get something resembling a native app without putting a lot of extra effort into it?”

There’s entirely good reasons that this dichotomy has emerged, but it’s kinda dumb too. In other words, paraphrasing Churchill: