It's so sad that it became "acceptable" to not test in Firefox (as estimated by the number of sites I randomly encounter that don't work in FF but do in Chrome) right around the time that Firefox Quantum happened and Firefox became good again :(
Agreed -- but FWIW a fair number of times I've encountered problems using Chrome which disappeared in FF. As a veteran of the original browser wars (circa 1998-2000) I'm grateful standards have come as far as they have -- and to support Moz at the heart of the open web.
> because Firefox on Linux was never actually that bad.
On Linux, I use Firefox primarily for browsing, but for development I use Chromium. The reason for that is because the JS debugger in Firefox is pretty damn buggy. Some things I've encountered (though they don't happen every time):
- On a breakpoint, type expression in console, hit Enter and it just hangs there without giving you the result. The console will be unresponsive until you unpause.
- On a breakpoint in some part of the callstack, type expression in console and see that variables that should be in scope at that point in the callstack are not in scope for the console.
- Go to a different spot in the callstack and see that the place that Firefox tells you you're at is not correct. It might be off by a few lines.
This might be a very good reason why developers prefer to develop for Chromium/Chrome. Not because they prefer it for browsing or for its performance, but because its development tools actually work.
I made the switch recently, like a month ago, and I've been discovering little things that are just more pleasant when working in Chromium. For example "Copy as cURL", is formatted neater. Firefox puts all the curl options in a single line, but Chromium separates the options in multiple lines.
I left Firefox in 2011 because it was unusably slow on my macbook. My recent experience has been that it is much faster and I am considering switching back.
At work we use CentOS, which only houses the ESR builds of FireFox as it fits Red Hat's intentions. I started here a few months ago, and found out that the studio was using Chrome as it's default browser. Turns out they had made the switch during the 57 ESR branch and never bothered to try the browser after, completely unaware of the Quantum project. Since we were having Chrome issues (which were admittedly our fault) I suggested trying FF again (CentOS now shipping with 68 ESR) as I couldn't replicate the issue. A colleague tried FF 77 on a personal laptop and was caught completely off-guard by the sheer performance difference. I'm excited with 78 ESR shipping on the 30th this month, it'll be a fun time getting all of the improvements over the past year!
It's the snappiest browser at the moment and I love its render as text function and automatic ad blocking on mobile. I've made a permanent switch on mobile, but I tend to miss the Chrome dev tools whenever I use it on a computer.
It's also wonderful that the browser is truly privacy conscious.
Tried it again last week. Still slow, I'm afraid, despite Mozilla literally inventing a new kind of programming language to speed it up!
A lot of it has to do with design, though. ie. when opening a new window, Chrome draws the window instantly and then fills in the UI, while Firefox waits until the window is completely built to display it. Even though they become usable at roughly the same time, Chrome responds instantly to the command while Firefox exhibits zero sign of life for hundreds of milliseconds.
Definitely not true on my system. There is a flash of some reorganization of the windows content that Chromium does not do, but both windows open instantly.
And also: Come on, that wouldn't make a browser slow. Users open new windows once when starting the browser. The rest is about how fast they render pages, react to JS workloads, and maybe how fast they open tabs. FF is more than on par in all of this.
I open new windows constantly. Opening the actual browser though is a whole other story, where the difference is even more striking.
Curious what you said about FF reorganizing the window contents, in my experience Chrome does that and not FF.
It's true JS performance is quite similar. But I've noticed (measured) differences in page load speed as well -- including latency again, the time until it actually responds to the enter key and initiates a network request. (In the network panel this is reported as Stalled.)
Double-check that your initial privacy excitement during setup isn't the culprit. The couple of times that happened to me (Paypal being one) lowering my "Browser Privacy" setting from "strict" or "custom" back to "standard" fixed the issue. I would be curious to see examples of broken sites if you have em.
I feel like I’m using IE in 1995 hearing this comment, but it’s valid. I feel like time is repeating itself. This is one of those tips that never really stop being relevant.
The modern equivalent is tuning uBlock Origin, uMatrix, Privacy Badger, or similar products, on top of correctly configuring your browser itself.
Some businesses are willing to (and sometimes should) take a 30% cut in traffic in order to ship a product more quickly. It's not a company/developer's fault that there are subtle differences between browsers.
Certainly not their fault. What's wrong is that it's possible to get the majority of traffic by developing for a single browser. They should not be sustainable doing that. It should be the case that in order to have sustainable levels of traffic they should develop for the standard. It should be the case that the cut is not 30% but rather 70% or more when compared to developing according to the standard.
