Hacker News Reader

Comment by skybrian | Hacker News Reader

Given how AMP eventually died, it seems unlikely that web developers would go along with it. What’s in it for them?

Also, I don’t see any sign that Google even wants to do it? This is not really evidence-based reasoning, it’s just “I can imagine something evil that Google might do.”

View on HN

Both are already happening.

Google markets Chrome relentlessly, with popups in search and YouTube if you're using other browsers, browser choice dialogs in Google iOS apps (despite iOS having a default browser setting for 5 years now), Chrome getting bundled into random Windows software installers, etc.

Many devs actively desire single-engine development and testing and many aren't shy about using Chrome only features already. If they had the capability to tell users to go install Chrome instead of targeting broadly supported features, they would do so in a heartbeat.

I have hit a few sites over the last year that threw up full page "This site only works in chrome" blocks, even though they usually work perfectly fine in Firefox if you set the user agent to chrome.

So in the enterprise world, it has been common for years for companies to "only support Chrome" even on iOS, where it's just skinned Safari. I have constantly had to call vendors mean names and point out how obviously iOS support means they are Webkit/standards-compliant. This is how I know, in fact, these websites will also work on Firefox. Apple's annoying iPhone monopoly is the last thing protecting the open web as needing to be standards-compliant.

The moment iPhones aren't allowed to force browsers to use Webkit (the EU is already pushing for this), the open web dies. There will no longer be any draw for web developers to develop for standards instead of developing for Chrome.

That's an interesting take that I hadn't heard before

And it's not just the WebKit monopoly in iOS, but also being slow on adopting new features pushed by Google. Often even being slower than Mozilla funnily enough. I don't care about what Apple's intentions could be for being a slowpoke on adopting the new features, as long as it allows independent browsers like Pale Moon to catch up with the mainstream.

> This is not really evidence-based reasoning, it’s just “I can imagine something evil that Google might do.”

Please read Mozilla's story on how Google sabotaged them: https://archive.is/tgIH9

Oh. And they very literally killed Internet Explorer: https://blog.chriszacharias.com/a-conspiracy-to-kill-ie6

Oh. And Google's mobile apps always conveniently forget the setting of "always use system browser and never ask me", and will keep asking you to open with "chrome", "google", or "system browser".

Oh and...

> Oh. And they very literally killed Internet Explorer

I disagree with this. Firstly, in this article they talk about how they "killed" IE6 in favour of later versions of IE, but MS ultimately killed IE with neglect until it was far too late.

Microsoft might have been neglecting IE, but Google was Definitly playing games with IE, constantly breaking video acceleration on YouTube in IE in any way they could. They were literally introducing invisible elements overlapping the video for no other reason than to break IE.

And web developers everywhere thanked them for killing IE6

How well do Gmail and Google Docs work on Firefox today?

Effortlessly, at least with uBO installed.

> What’s in it for them?

Never having to use polyfills or CanIUse tables, plus testing on the same environment they develop on.

There's no way to test on Safari without either buying Apple hardware or subscribing to services like Browserstack.

This is a problem of Apple's own making.

I wish Apple had some sort of "Geforce Now" style setup to run a Mac in a box. I know they'd never go for something like a legit image you could run in a VM, but surely they could come up with something.

My work sent over some old MacBook for when we need to test something unique to Safari, so it's not even the hardware aspect. It's the "I need to find another place to stash a machine, and then wire up KVM switches to use my highly opinionated I/O device choices, on a finite sized desk" factor.

Reply on news.ycombinator.com