The problem is cross-platform support. Depending on the OS is obvious if every OS supported the required features.
When I say e.g. skia should be part of the OS, I don't neccessarily mean MS should ship it and update it yearly. I mean Google should still ship and auto update it, but also go though the effort of documenting it, maintaining strict backwards compatible APIs, and letting other programs consume it. I don't care who the actual vendor is. I know that's a lot to ask for, but OTOH it is insane to statically link that kind of code. Especially if you have multiple electron apps, that would work fine with a shared runtime.
Ideally, I'd want critical code (encryption, code signing, bootloaders, kernels, runtimes) to be from a trusted vendor, and preferably simple and open source. I trust the MS, Apple, Google of 2017 not to completely fuck it up. (We already trust them as browser vendors.)
I don't care if the keep calc.exe stable for 10 years, but I expect them to patch crypto.dll immediately. You could do that stealthily, outside of major updates, as it has no user facing changes.
The benefit of this model is that it allows third party apps from small vendors to profit from the up to date security that only the tech Giants can provide.
The downside is of course that it is quite hard to maintain perfect backwards compatibility while pushing updates, but if the components and APIs are small enough I think it is possible.
I really like elinks, it's a shame I can't use it to view blackboard and other sites I have to use...
There is no point in speeding up the raw compile time by a couple of minutes if you are increasing the development and testing time by a couple of weeks.
I think large parts of chrome actually belong into the OS. The network parts, the drawing library (skia), the crypto implementation, the window and tab management, and so on.
The javascript engine could be factored out, too, so more apps could benefit from it (without bundling a whole frickin Chromium).
Video and audio would be deferred to DirectShow, Quartz, VLC, Mplayer, ...
Ideally, what remains is just a layout engine and some glue code for the UI. It's the monolithic kernel vs microkernel debate all over again.
Plugins have a bad rep in the context of browsers, but I think this "microkernel browser" where everything is a plugin or OS library can be potentially more secure than the current state, since we can wall off the components between interfaces much better.
I also think it would be much "freer". Browsers like Firefox and Chrome are open-source, but they are free in license only. I can't realistically go ahead and make my own browser. The whole thing is so complex that you have to be Google or Apple or Microsoft to do that. The best I could achive is a reskin of WebKit. I think that would be different with a more modular browser.