Well, they *are* browsers after all. Of course they do.

How's that different than if you install Firefox on your Android? (Or Google Chrome on iOS).

You are trusting any browser you install with your full browsing data, which is why Firefox, Safari, Brave, etc. use privacy arguments to try to convert Chrome's users. How many users realize that their browser decision isn't keeping them safe when they click a link in an app?

It's not like they really have a choice.

They don't and that's not OK. But more to the point, giving a choice will not fix the problem in practice. This is a serious design flaw that users shouldn't need to know or care about.

It would be great if more in-app browsers (or even all of them) could be the good kind. Tough problem to take away or restrict existing APIs, but worth it to work on it. In any case, the in-app browsers that are implemented the less good way are still browsers.

I've been defining browsers as "apps that accept all navigation intents"; that's a bit Android specific (as you can actually replace the default browser there), but it captures the essential difference in responsibility to the user, IMO. These WebView things are franken-browsers.

Arriving at a sort of agreed-upon definition would be interesting and would help inform conversations. Maybe a @w3ctag finding? How can we make progress when there's no agreement between browser engineers as towards what we're working?

I’m not sure a TAG finding on “what is a browser” would be of practical use, and it could easily fly off into the abstraction stratosphere. (My quaint view is: if it does web browsing, it’s a web browser. But not sure this has anu practical relevance.)

Does browsing imply being able to type your own URLs?