Live data from Hacker News

Firefox 72.0

mozilla.org

291–300 of 543 posts

Re: Firefox 72.0

#291
post #176

Earlier quoted context omitted.

If you’re interested in having Chrome’s look in Firefox, check out MaterialFox. https://github.com/muckSponge/MaterialFox

Wow, I didn't even realize this was possible. It looks really nice. Thanks!

Unfortunately userchrome.css breaks frequently with firefox updates and is a pain to maintain since it is outside of the extension ecosystem.

There used to be a great addon called ClassicThemeRestorer that got killed during firefox's terrible web extension debacle. It lives on in https://github.com/Aris-t2/CustomCSSforFx but different parts break everytime the fickle designers at mozilla change their mind on what a tab should look like.

Re: Firefox 72.0

#293
I find the notification permission change (requiring it to be after a user gesture) really annoying. It doesn't do almost anything to combat permission spam (if you're visiting a clickbait site, you'll probably click or scroll at least once, especially on mobile where sites usually have a giant header so you need to scroll past the fold to even see that the article is clickbait), and just inconveniences developers who have to special-case Firefox's divergence from the spec when trying to legitimately use the gated features.

Re: Firefox 72.0

#294
post #100

Earlier quoted context omitted.

I use Firefox as my default browser. But the dev tools is too slow, apparently someone thought it was a good idea to rewrite it in React, causing it to lag even on a few thousand console log messages on a fairly powerful workstation. And there is no search functionality, only filter. So I've started to prefix all my debug messages with whatever I'm working on just so that the logs can be found in the Firefox dev tool…

DevTools member here. We have been working on performance over the past release – is there anything specific that you can point out as being slower for us to focus on?

I have not yet tested v72, I'm on v71. The Chrome dev tools Console feels (not measured) about 100x faster then the Console in Firefox. Even Chromium Console can be limiting though. Yeh, I probably should not use console.logs for debugging, but it's much faster forkflow then firing up the debugger. Even with the lag. With the Chromium Console the lag is not that bad, and it has a search function. Both console's have a bunch of cool features, like being able to inspect DOM nodes, explore objects, etc. Witch I think might be a reason for the lag. Most of the time I just want to search the logs to find out what happened. An idea is if you can just get the plain text from the console.logs somehow. In chromium all console.logs get sent to stdout, so I can pipe the logs to a file, then use any editor to search it. But I prefer using the dev tools.

Maybe make a perf switch in the console, which dumps the text without formatting or other features. Something that can handle one million lines of logs and search in it. Being able to search the logs is critical. Filtering doesn't work as I need to see the surrounding logs, what happened after. I don't need to see all one million lines at the same time though, it can be buffered to only show 100 lines above it and 100 lines below it. Chromium has such an optimization. But search is a bit slow.

Re: Firefox 72.0

#296
post #94
post #40

"Firefox replaces annoying notification request pop-ups with a more delightful experience, by default for all users. The pop-ups no longer interrupt your browsing, in its place, a speech bubble will appear in the address bar when you interact with the site." That should pretty much kill off a lot of the notification request crap, especially if Chrome follows suit. I can envision the conversion rate massively falling…

Slack has started disabling unrelated features if you disallow notifications. For instance, you can’t add channels to the “ignore @here” list if notifications are off. I’m sure this’ll get worse over time. FF needs a “disable notifications, but lie to the website and say they’re enabled” button.

I feel like if a specific browser, ie. Firefox went down this path, it wouldn't take long until websites stopped supporting that browser. It'd take browser wide action to take on big players such as Slack imo.

Re: Firefox 72.0

#297

I find the notification permission change (requiring it to be after a user gesture) really annoying. It doesn't do almost anything to combat permission spam (if you're visiting a clickbait site, you'll probably click or scroll at least once, especially on mobile where sites usually have a giant header so you need to scroll past the fold to even see that the article is clickbait), and just inconveniences developers wh…

If it's treated the same way as popup windows, then scrolling shouldn't count.

Re: Firefox 72.0

#299

I know the hacks blog tends to stick to "what web technologies changed" but 72 also introduces "Experimental support for using client certificates from the OS certificate store can be enabled by setting the preference security.osclientcerts.autoload to true (Windows only)." which has always been a huge PITA for me while testing and many interested in these technology changes are probably interested in this flag (thou…

I've always used the security.enterprise_roots.enabled flag for this, is this new one different?

Re: Firefox 72.0

#300
post #299

I know the hacks blog tends to stick to "what web technologies changed" but 72 also introduces "Experimental support for using client certificates from the OS certificate store can be enabled by setting the preference security.osclientcerts.autoload to true (Windows only)." which has always been a huge PITA for me while testing and many interested in these technology changes are probably interested in this flag (thou…

I've always used the security.enterprise_roots.enabled flag for this, is this new one different?

From the property name it sounds like your property is about trust roots (CAs) and parent's about client certificates (for mutual TLS authentication).
Post reply on HN