Is Chrome the New IE? (2023)

> If it works on Chrome, no one cares or even tests for other things.

We've been through an extensive standardization pass for this to not happen. Anything not matching the specification whether in Chrome or any other browser should be considered a bug.

This is not at all the same as IE, where it just went its own way.

> should be considered a bug

Should be, but isn't. At least not in a practical sense.

Yes it should be, but it isn't, that's the problem

The unstated major premise of this assertion is that the standard is a spec that every browser must comply with exactly. It's not; there's not a single browser that has ever implemented 100% of whatever was the latest standard at the time, and major browsers typically also include many of their own additions that go beyond the standard.

This latter bit isn't in conflict with the standard; it's an essential part of the standardization process. The typical route for something making it into the standard is for a browser to release their own browser-specific extension and use that as a basis for advocating that it be added to the standard. XMLHttpRequest, for example, started as an IE-only feature and didn't make it into all the other major browsers for several years. It got a published W3C spec a little bit after that, which meant that browsers needed another couple years to also get synced up on their behavior.

In this respect, Chrome has definitely now taken IE's old position: new Web standards have a tendency to start as Chrome-specific extensions, and then the other browsers have to implement their own versions and get them ratified into the W3C specs in an effort to try and keep up. Which in turn suggests that a compatibility-minded Web developer might want to choose a similar strategy from what was done in the past: test on the most popular browser last.

> The unstated major premise of this assertion is that the standard is a spec that every browser must comply with exactly. It's not; there's not a single browser that has ever implemented 100% of whatever was the latest standard at the time, and major browsers typically also include many of their own additions that go beyond the standard.

Sure, but there is a big difference between implenting 99% of the standard and only implementing like 10%

As I wrote in a similar thread a year ago: Whenever I point out that some bug which happens in Firefox my colleagues usually responds with some variant of "we tested in Chrome, and that is the standard", or "can you ask the customer to use Chrome instead". Even if Firefox or some other browser may be using a proper standards implementation and the Chrome one being the one with some quirk.

It makes a difference in how easy it is to get it working in multiple browsers.

But if developers don't check, then either one could break the site for all the users.

ChromeOS and Project Fungu.

What about half-assed specifications that Chrome throws over the wall, ships them, and advertises them as standard everywhere, including web.dev? Even though their status is "not on any standards track"?