Trouble loading Firefox, Thunderbird and Basilisk this morning. Everything was fine yesterday, but today it can take up to a minute for any of these applications to finally load. No problem with Vivaldi.
The error for Firefox was,

Here is mine:
take a look in:
settings / appearance / applications style
and on the bottom right : style for GTK apps

But, who knows how this kde4 deals with gtk now

1 members found this post helpful.

For the longest time, I was not aware that the "block deceptive and dangerous content" function in Firefox basically redirected all your browsing through to Google for them to filter (or censor) your browsing content.

This has 2 main implications. First all your browsing can be surveillanced by Google, like everywhere else, even though you try to avoid it and not use any products from Google. So, if you're the type who don't want Google to surveillance you, this function MUST be turned off.
Secondly, this function "filters" your content. Sure, it might filter out some dangerous content online, but more likely it will "censor" the information you see online and block content that Google (or government bureau of information) doesn't like, like for example opinions they don't like, or views and expressions they don't want you to see. I don't want to open a can of worms here, but Google is by now well known for their censorship efforts online, by blocking free expression and "undesirable" opinions on platforms like youtube and google search.

So, if you are concerned about Google surveillancing what you do online (like they do when you use their search engine or products) and filtering (/censoring) the information you get to view, then you should look into the function in your Firefox settings under "privacy and security", and disable
Personally I would like some blocking/filtering functions in my browser, but not when it is provided by Google and the price is surveillance. It can be noted in all this that Mozilla receive alot of their funding from Google. But it would have been nice with a choice of some other filter functions. Perhaps from some open collaboration who doesn't surveillance and abuse users.

This function and any other creepy related and non-related Google thing, can also be found and looked into under about:config by searching "google". The actual name of the function is "Google Safe Browsing" and it's not run through a file or a database file, no, it's run through their servers.

3 members found this post helpful.

Wow! I had no idea! I've just turned that function off.

I noticed during the pandemic that when I DDG'd "Great Barrington declaration", because I wanted to find out what this document actually said, I couldn't find it anywhere. I only saw shedloads of newspaper articles explaining why it was wrong.

Don't know where you're getting your info from, but here's what Mozilla themselves say about this functionality:

How does Phishing and Malware Protection work in Firefox?
What information is sent to Mozilla or its partners when Phishing and Malware Protection is enabled?
Seems reasonable to me.

2 members found this post helpful.

By looking into the details of these functions. Perhaps those are subfuntions of some sort, but the main function is Google Safe Browsing. It seems pretty clear to me how it works, but like anyone I could be wrong, but the information I looked into was very clear about it, and the information about safe search you can find under "about:config" is also CRISPY clear.

Furthermore, if you read "learn more" under the function, it is also clear, as it links to "Google Privacy Policies" (and Mozilla Privacy Policies) how they treat the information.

I don't think it could be more clear than that, which is why I have disabled the function for many years (and I've not been victim of malware or dangerous content). And like I said, I would like a filter function like that, but from a reliable source, and certainly not from a company like Google. I'd like to be able to filter out content I consider harmful (based on filters reliable experts have put together as "pick and chose"), rather than what Google considers so. In addition to that, most search engines provide a moderation filter as an option, and mostly it is enabled by default.

What you quote is just a different variation of saying this:
and this

It should be noted that I did not get my information from such direct sources in the first place, but rather from passionate privacy advocates and the likes, who themselves looked into it extensively and wrote about it.. So, I'm not saying I'm the expert, but I did listen to what I considered experts on the topics, and they were in general agreement with each other. My own verification of their information made it very clear to me that I shouldn't use these function, for the reasons described in the post I made.

Even if it was just a list file provided by Google, which it isn't (it's client to Google server service), I'd still not want to use a Google "one-size-fits all" lookup table for this kind of function, but rather "pick and chose" tableS from reliable sources. And if this information is incorrect that Firefox uses the API which connects to Google servers realtime, why would there be a Google Privacy Policy attached to the function in the first place?

It's probably possible to test this yourself, you can activate the function and add the google safebrowsing address to a firewall and block it, and get logs of blocked attempts to reach that address.

2 members found this post helpful.

I read the Mozilla explanation and what I've got from it is that there's no surveillance by Google, mostly URL check against locally stored lists, also a quick online check when there's a match, to see if it's still on the list (because the lists are being updated regularly). Also an online check for an executable (not sure what falls under this description) file's name, size and hash against a malware list. Which I understand that not every downloaded file is even being checked.

I admit that I haven't looked into the code to see how exactly all this works (too little knowledge of C++ and/or Rust to understand it), but somehow I doubt we would find anything surprising there.

But if you still don't trust Mozilla and Google, an alternative would be using uBlock Origin extension and enabling Online Malicious URL Blocklist. Or any other content filtering extension with a similar blocklist.

Good on you. That's for everyone to consider on their own. I'm not going to tell them I'm right, I'm just presenting the information I have and I based my decision on. People are free to look the information up themselves if they want. There is alot of information on the topic available to the public. Also in much greater detail.

Like I said, I didn't want to open a can of worms here, and this is kind of a separate topic, so from my end it's done, and I'm not discussing it any further. Only the thread starter can convince me otherwise. If I have anything further I forgot to add, I'll add it in my clarification post above. End of story.

Logged out and deleted the ~/.config gtk-x configuration directories and signed back on. Didn't solve the problem. Setup a new user, still didn't solve the problem. It takes 37 to 40 seconds for Thunderbird, Firefox and Basilisk to load. Sending and retrieving e-mail via Thunderbird has slowed to a crawl.
Tried Xfce-4.18, but the problem is the same.
Everything else works just fine.
I'm just about due for a fresh install. Maybe that will help.

I am afraid that there is more work, if you want to get rid/ turn off safebrowsing:
enter in address line
about:config
browser.safebrowsing.allowOverride false
browser.safebrowsing.provider. delete all web references.
browser.safebrowsing.blockedURIs.enabled false
browser.safebrowsing.downloads.remote.enabled false
browser.safebrowsing.features. all false
browser.safebrowsing.malware.enabled false
browser.safebrowding.pishing.enabled false

This is just bare minimum

@cwizardone
Maybe try to reinstall kde-gtk-config. I think it provides colorreload-gtk-module

Thanks for the thought. Tried that, but no joy.

I'd sure be happy if someone had some good tips on reducing Firefox memory usage (in particular), and CPU use. I did look into it years ago, and I did find some things that could reduce memory and CPU usage, but in the end the problem was not fully solved, which caused me to eventually lock Firefox in a virtual machine. Yup, the biggest possible hammer, I know, but I had tried for years to tame the Firefox in various ways, and I'm still trying. But for now the VM prison of shame works quite well.. Imagine a situation where a whole virtual machine running Firefox is less resource problematic than running Firefox in the host OS.

Anyways, I do want to also be able to run Firefox outside a virtual machine, so I'm looking into these memory reduction adjustments again. I have a few worth mentioning as well, in about:config...
browser.cache.memory shows 3 options "capacity", "enable" and "max size".. By going to about:cache, you can verify that this cache is actually in RAM, so if you want to reduce memory usage, this function can be disabled and/or set to a low number.
browser.sessionhistory.max_total_viewers which is how many back and forward pages Firefox stores (in memory? and/or cache) it is set to -1 which is unlimited I suppose, or letting Firefox decide, but I recon it doesn't hurt to set it to 2 o 3 for most normal use or even 0 if you don't want cached back/forward browsing. There is also a similar entry "max_history" which can be set lower.

Furthermore, there is ofcourse the "simple" option to use a 32-bit version of Firefox.. I use LSM MAC (mandatory access control), and I don't like the way Firefox fumbles around in the filesystem. This isn't really related to performance much, but it's possible to use a "hidden" proc mount where processes only see "self". But be weary of using that function as it can break programs from running properly, among other Wayland will not start in that mode. It's just a step short of using the PID namespace basically, but from what I remember it does work with X. Which reminds me there should be some ways to limit (max) Firefox resources with cgroups.. But, didn't get that far yet..
Another thing is perhaps that you may want to disable the user namespace for Firefox.

Well, anyways, I'm a beggar for ways to reduce Firefox memory use if anyone have any good tips (preferably about:config stuff).

Thanks to the Kernel

Imagine a situation where Firefox on a 2cpu/2Gb Virtual Machine consume exactly the same amount of resources as on a 8cpu/16Gb host ...

1) install uBlock Origin, edit serrings and enable all filters. With the exception of international, import Actually Legitimate URL Shortener Tool
above will reduce memory usage.
browser memory limit is set to -1 change to whatever you think is valid.
2) limit session history (I have 3: good enough for me), disable crash recovery, disable safebrowsing.
3) get rid of most of add-ons

4) in VM setup limit memory use in VM settings and CPU number. Linux, whichever distro you installed will adjust during installation.
5) setup nice sandbox if you want limit firefox disk/user data access. In addition you can configure tempfs
6) 64-bit has better memoty handling than 32-bit
7) in VM if you are adventurous play around with kernel.
8) fix your WM/DE (remove stuff that is not needed for functionality) both main OS and VM.

You can easily get less than 100MB at boot so even if FF will take 1GB with all the tabs, 2GB is crazy generous.

And this morning things are back to normal. The problem has disappeared as mysteriously as it appeared.