Hacker News Reader

Comment by shadowgovt | Hacker News Reader

Web browsers make more sense when you think of them as emulators for a novel architecture / OS built atop an existing OS.

It is "impossible" to build a new browser in the sense that it's "impossible" to build a new Windows 10. Of course, a key difference is that the web is specified atop an open standard so someone can at least try.

View on HN

I don't think anyone will try. It's 'impossible' in the sense that both Microsoft and Opera threw in the towel and turned their browsers into Chromium skins. If even Microsoft can't pull it off, it seems unlikely anyone else is going to seriously try starting afresh.

Mozilla have been successful in swapping out large chunks of their browser for superior replacements (for instance replacing the JIT, several times iirc), but that's not quite the same thing.

Another point: why would a company bother? Amazon's 'Silk' is just another Chromium-based browser, but I imagine that works just fine for Amazon. What's the downside of taking Google's hard work and putting your branding on it?

Exactly. It's much like asking "Why would a company bother writing a new OS from scratch when they could instead become a new Linux distro?"

Like Google's Fuchsia, oh wait...

Fuchsia seems to be an exception that proves the rule.

And it doesn't find the rule wanting. Three years since it stealth-launched, and forward momentum on Fuchsia seems... Questionable?

Fuchsia looks pretty active as far as commits go:

https://fuchsia.googlesource.com/fuchsia/+log

It also seems I regularly see new posts about it on HN, with a generally positive reception and optimism among its users.

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

The only piece of Fuchsia that's excited me so far is the Xi Editor, which seems to be developing rather slowly if not stalled out completely. https://github.com/xi-editor/xi-editor

(Yes, it's not part of Fuchsia anymore, but so far I haven't heard anything about the remaining parts of Fuchsia that makes me inclined to care about it until Google forces me to care.)

Looks ambitious, and happily there's some recent GitHub activity, but I already have Vim and Notepad++. What does Xi have that those two don't?

It's just intended to be way faster than existing editors: "Incredibly high performance. All editing operations should commit and paint in under 16ms. The editor should never make you wait for anything."

If it succeeded in that, I think it would pave the way for faster, less-bloated software in general, so I'm rooting for the project to succeed.

I like lightweight software, but performance isn't a problem in any modern text editor. Notepad++ and Vim are already light enough.

Vim is 'real-time' responsive even over SSH. I don't see much room for improvement.

Google has their search, ads, and pending browser monopoly to fund their endeavors. Most companies don't.

Why is it impossible to build a new Windows 10? Shouldn't that be an area of great interest to Microsoft? What other endeavor would give Microsoft as great a leap forward as a ground up rebuild of the operating system?

Microsoft doesn't need to build Windows 10 from the ground up; they have a Windows 10.

It's not impossible, but it will take an enormous amount of effort to replicate. Consider the ReactOS project, which strives for compatibility with Windows Server 2003. I use it occasionally in a VM and it is roughly at par with Windows 2000/XP, but this project has been going on for over two decades (granted, the target has moved over the years; originally it targeted Windows NT 4.0 if I remember correctly). It will take years of work for ReactOS to reach compatibility with Windows 10.

They tried for 9 years and eventually (rightly or wrongly) gave up: http://joeduffyblog.com/2015/11/03/blogging-about-midori/

Reply on news.ycombinator.com