Firefox 67 Arrives With New Performance and Privacy Features, Voice Search Widget on Android (venturebeat.com) 121
Mozilla today launched Firefox 67 for Windows, Mac, Linux, and Android. From a report: The 10th release since Mozilla's big Firefox Quantum launch in November 2017 doubles down on performance and privacy. Firefox 67 includes deprioritizing least commonly used features, suspending unused tabs, faster startup, blocking of cryptomining and fingerprinting, Private Browsing improvements, voice input in the Android search widget, and more. [...] Firefox 67 is better at performing tasks at the optimal time, resulting in faster "painting" of the page. Specifically, the browser deprioritizes least commonly used features and delays set Timeout to prioritize scripts for things you need. Mozilla says Instagram, Amazon, and Google searches now execute between 40% and 80% faster. Firefox also now scans for alternative style sheets after page load and doesn't load the auto-fill module unless there is a form to complete. Next, Firefox 67 detects if your computer's memory is running low (under 400MB) and suspends unused tabs. If you do click on a tab that you haven't used or looked at in a while, it will reload where you left off. Finally, Firefox 67 promises faster startup for users that customized their browser with an add-on.
They FINALLY started supporting... (Score:1)
They FINALLY started supporting the ability to open Firefox with a given URL with a given profile. I kid you not; until *today*, it has always ignored the profile parameter, instead opening the URL with the default profile. This has caused me endless grief, until today.
Now, they only have 48 trillion more bugs and "features" to fix/remove until Firefox becomes usable...
Re:They FINALLY started supporting... (Score:4, Insightful)
Wow, 48 trillion, that's a lot of bugs. Why don't you point toward that bugtracker so I can verify the number.
And while you're at it, compare it to the other browsers' bugtrackers so we can get a baseline of what's normal in such a gigantic project such as a mainstream web browser.
Re: (Score:2)
You can't compare bug numbers across systems because the number of bugs is more cultural than technical. Today I loaded a mechanical device with something to process - it was both slow, and it failed to correctly detect the contents of what I inserted. I that 2 bugs or one? One company might say add both bugs. Another might say check with the developers to see if they are already working on something closely related. Another might say log one bug and let the developers see if the root cause was the sam
Re: (Score:2)
Re: (Score:3)
This "new" way (which isn't actually new, I use an add-on (Auto Tab Discard) that does the same thing based on a few more parameters than just RAM avail) will suspend tabs that are already open -- i.e., you've been looking at the tab(s) but your RAM runs low, so FF suspends one of them. Obviously, not the same mechanism as having the tab not load after you first open FF unless you've clicked on it.
Re: (Score:2)
That's correct. It should effectively unload existing tabs dynamically and when you click on them, they'd act like those never-opened ones when you click on them.
You know what slows down browsers most? (Score:5, Interesting)
Ads. I didn't realize how much until I installed an ad blocker about a month ago.
Don't get me wrong - I understand that sites that are free (read: you don't pay money to use) need to make revenue in some way, and that's almost always going to be ads. However, there has to be some sort of reasonable tradeoff between getting these sites revenue and making my browser so damn slow in everyday use that it's practically unusable.
Re: (Score:1)
I'm not sure what you're on about, with uBlock I still don't have any ads on Youtube, trying various old and new videos.
Re: (Score:2)
The one thing though is that Google has been able to get around the blocks to play their ads on YouTube. Wow, are the ads awful on YouTube! Last week an eight minute video I played had five, count 'em: five ads, one at each end and three in the middle. 'And YouTube has now started doubling up ads where one plays and then another before the video starts. If there were another platform I'd switch quicker than a bunny, but DailyMotion and the like are sad, Vimeo is so niche and so on .. oh well.
I don't have much to say on that. I rarely use YouTube on my computers proper or on my phone. The majority of my YT viewing is done on my AppleTV. I don't so much mind the ads on YT, I'm sure they incur a lot of costs for storage and pushing out as many streams as they do, and if I'm viewing it without paying $ for it, I understand that the ads are a fair trade for that. At least on the ATV, most ads are 15 seconds long, or give me the option to skip after 5 seconds. Every once in a blue moon I get on
Re: (Score:2)
That's why I'm becoming more of a fan of Brave's alternative payment system. I honestly don't know how well it works out for websites, but it's the right kind of idea IMO.
You basically buy a browser currency called BAT (Basic Attention Token), and every site you visit gets a chunk of that. Sort of a "hey, I know you guys need money to run, and I appreciate that, here's what I think my attention is worth" kind of thing. The site has to sign up with Brave and verify their ownership of that site, and they can
Re:JavaScript is to blame, I believe (Score:5, Insightful)
I don't believe that JavaScript itself is the cause, it's the usage of it.
Professionally, I'm a developer, and I write and use JavaScript every single day of work. Used properly, JavaScript allows me to make web-based apps that work just as well or better than traditional applications. There are certainly cases to be made for web-based systems over traditionally compiled and installed applications (especially for multi-user systems), and for that, there's really no alternative to JavaScript.
Does JS allow performance-trashing code? Absolutely. Does responsibly-written JS necessarily mean performance-trashing? Absolutely not.
Re: (Score:2)
Used properly, JavaScript allows me to make web-based apps that work just as well or better than traditional applications.
You sir, are deluding yourself.
I guess that the companies that have employed me and their users, who have all praised my work, even when it is replacing a traditional application with a web-based one, are all deluding themselves also.
Bad developers write bad code, and good developers tend to write good code, regardless of what language they're writing it in.
You, sir, are deluding yourself if you think that JS can't be used properly and responsibly.
Re: JavaScript is to blame, I believe (Score:5, Insightful)
LOL. You're pretty funny. In the same way I have a responsibility as part of my job(s) to write good, efficient code, I also have a responsibility to make sure that the company/companies and their employees aren't flooded and harassed by internet users who disagree with me. Nice try, though.
Re: (Score:1)
I guess that the companies that have employed me and their users, who have all praised my work, even when it is replacing a traditional application with a web-based one, are all deluding themselves also.
Please give us the names of the companies, as well as the contact information for both management and users at each company. We need to verify these claims that you have made.
You shouldn't object to this. You have nothing to hide, correct? Besides, you were the one who brought them into this discussion.
We await the requested information. Please provide it swiftly.
LOL that's funny coming from a Anonymous Coward.
Re: (Score:3)
I'd like to evaluate the usability of your web applications compared to "traditionally compiled and installed applications". Among web applications that you've worked on that are open to the public, which do you consider your best work?
Re: (Score:2)
Unfortunately, none. In my career, (30+ years, for the record) I have specialized in internal systems. I hope this won't result in whomever dismissing my experience, although I suspect it will for many here.
I'm not claiming that every web app is better than any traditional app. Far from it. There are tons of terrible web apps, and tons of great traditional apps. I'm only saying that a well-designed, well-coded web app can, in certain circumstances, be better than a badly-made (or even an averagely-made)
CPU, RAM, and privacy (Score:3)
I too have developed internal web applications, most of them using little or no client-side script. Instead, they rely on navigation and form submission, with most of the logic running on the server. Even prevalidation before form submission needs less script than it used to since HTML5 added a whole bunch of values for <input type>.
I'm only saying that a well-designed, well-coded web app can, in certain circumstances, be better than a badly-made (or even an averagely-made) installed app. For instance, when multi-platform support is needed.
The traditionalist's answer to multi-platform support is Qt, GTK+, wxWidgets, or even Java Swing.
My issue in this thread is with the assertion that it is not possible for a web app to be good, and that even a badly-made traditional app is inherently better than a well-made web app, in every circumstance. That kind of absolute thinking isn't what real-world experience shows.
I think part of the complaint is that traditional applications distributed
Re: (Score:3)
I agree with everything you said in the above post.
I, too, put as much of my logic as possible server-side, and try to keep client-side code to what client-side code in a web app was always intended to do - update the UI based on changes in the data.
As far as privacy goes, I will readily admit that I have been fortunate to be developing internal applications, and as such, have never even been asked to integrate with a 3rd-party analytics or advertising provider.
For performance, yes, there is extra overhead
Re: (Score:2)
What is "traditionally compiled and installed" mean in terms of usability? Realplayer? ICQ? The latest and greatest fluid design framework from Windows?
The fact you discuss usability and conflate it with type of code clearly shows you know little of what you're talking about.
Re: (Score:2)
What is "traditionally compiled and installed" mean in terms of usability?
I would compare media players to VLC. I would compare chat applications to HexChat and Pidgin.
The fact you discuss usability and conflate it with type of code
Web applications and traditional native applications differ in usability in a couple ways. One is that native applications can achieve a tighter integration with the native platform, such as ability to drag things into or out of the native file manager, or ability to search for text in all the files in a folder. Another is resource use: web applications and Electron applications tend to take more CPU and RAM than t
Re: (Score:3)
I don't think we can solely blame JavaScript (or any language, for that matter) for bloat, bad performance, and producing incompetent programmers.
When I'm starting a new project, I ask the people that want it "There's good, fast, and cheap. You can't have all three. Which two do you want?" (for the record, fast in this context means the project is delivered fast, not that the application runs fast). Incompetent programmers are accepted because people are willing to accept "fast and cheap
Re: (Score:1)
From an end-user perspective, though, the options for dealing with bad Javascript code in the wild are all suboptimal to outright bad. You could allow Javascript globally and deal with all the problems that entails, like slow-loading sites and occasionally malicious Javascript code. This is the option most users take because (a) it's the default and (b) it's the only option that ensures someone without experience will have all sites working as designed (however badly that may be).
You could go with a blackli
Re: (Score:1)
Re:You know what slows down browsers most? (Score:4, Informative)
An ad blocker or a no-script plugin would strip all of those out and the performance is vastly better. Not to mention less intrusive.
This isn't a new phenomena though. Flash plugin used to take a lot of shit for being slow but it was mainly due to so many sites plastering flash ad-serving content all over the place. These days HTML+JS+Canvas does the same and it sucks just as badly.
Re: (Score:2)
Not just ads but social media and tracking scripts. A case in point. I was looking at a outdoor sporting store today called Cabelas.com. It had THIRTY-TWO external scripts, not a single one relevant to the site's function.
An ad blocker or a no-script plugin would strip all of those out and the performance is vastly better. Not to mention less intrusive.
This isn't a new phenomena though. Flash plugin used to take a lot of shit for being slow but it was mainly due to so many sites plastering flash ad-serving content all over the place. These days HTML+JS+Canvas does the same and it sucks just as badly.
Amen to that! I can't tell you how many times I've gotten annoyed at a site loading slow, not because that site's servers were slow, but because the page required external scripts, and the servers serving those external scripts was what my browser was waiting for.
When developing a site, if I want to use a 3rd-party library or something like that, I'm far more likely to grab a copy and place that copy on my server, rather than relying on the 3rd party's server. Yes, it means that I have to be vigilant abou
Re: (Score:2)
Why not write a script to cache the external resources every day (after the script verifies they're not returning an error or blank)?
Re: (Score:2)
until I installed an ad blocker about a month ago.
I really hope you only discovered this internet thing a month ago. Otherwise I would question your mental state.
"Doubles down on privacy" (Score:2)
The browser that literally had a critical bug that effectively stopped older out of support versions working, which forced people to submit to mozilla literally remotely running software on your browser.
And they have the gall to claim they're here to improve privacy? That action was on the level if not more invasive than data mining giant google itself in its entirety.
But hey, at least it's fast. Because new add-ons simply can't do the things they could before. Mozilla has become that car company that think
Re: "Doubles down on privacy" (Score:5, Informative)
I don't see such options in Chrome.
How can these features be done privately? (Score:2)
How would you recommend to do the following without sending data to anyone?
A. Determine which features are worthy of continued maintenance labor
B. Determine the parts of the code that most often cause an application to crash, especially on operating environments that differ the most from those of the developers
C. Protect users from inadvertently violating their own privacy by keying personal data into a phishing site
Re: (Score:3)
A. First, have a strong consistent vision. For example, you do not need user feedback if you want to build a secure browser which supports standards. A feature deserves maintenance labor if it it is truly core to the vision. Second, allow a flexible plug-in architecture so that most features can be pushed into "userland". For example, an addon can intercept non-standard HTML (thinking of IE6 days here) and translate it into standard compliant code which then can go the browser. Track importance of such adde
Re: (Score:2)
you do not need user feedback if you want to build a secure browser which supports standards.
Feedback would consist in part of which standards are and are not worthy of support. File submission through HTML forms is a standard, but the first five versions of Safari for iOS did not support it. RSS is a standard, but Firefox recently dropped support for it because its telemetry showed low use.
[Anti-phishing and anti-malware] is not core browser functionality. Users are responsible for their own choices. Stop handholding them.
Many users desire safety. Some stop using one browser in favor of another in part because of this sort of handholding. Even if Compuser believes that anti-phishing and anti-malware ought not to be core web brows
Re: (Score:2)
Users are responsible for their own choices. Stop handholding them.
The fact that ZDNet is running a story about web browsers that failed to show adequate phishing warnings [slashdot.org] shows that a sufficient number of users consider this sort of handholding to be a core browser feature.
Re: (Score:2)
Firefox has multiple settings that must work in tandem, so the GUI features often don't work. I explicitly disabled local storage on my Firefox installation, but the browser ignored my setting and continued to save data anyway. I found out later there were multiple settings in about:config that had to be disabled all at once, or else the checkboxes in the GUI do literally nothing.
Also, Mozilla has a tendency to rename options regularly, such as adding a random underscore. This ensures that after an updat
Re: (Score:2)
Why do people say that Firefox offers good privacy?
I recently read Firefox's privacy policy [mozilla.org]. If you're a Firefox user, I think you should read it, too.
None of that sound all that ominous, unless you believe that telemetry = spying, rather than as a way of improving software based on how users actually use said software. And I'm fine with sharing some data that collectively helps to improve browsing safety and block malicious content. Sharing data with third-parties? Not thrilled about that, but Mozilla needs to pay its workforce.
But the key is that all of this can be turned off if you're not comfortable with it. I'm pretty sure this is why Mozilla gen
Re: (Score:2)
None of that sound all that ominous, unless you believe that telemetry = spying
At least for me, the issue is that trust has been completely eroded. Companies like Google or Facebook ignore user choices and work hard to bypass any countermeasure people try to use to salvage what's left of their privacy. They come up with hundreds of techniques to de-anonymize your data, to identify, locate and track you between browsers, devices, even off-line. They lie about data collection, misuse your info or sell it to all and sundry, with limited or no consequences.
It's reached the point where I d
Re:"Doubles down on privacy" (Score:4, Informative)
It's worse than that.
After you update to Firefox 67, they demand you log in to your "Firefox Account" and offer to create one if you don't have one yet.
Because clearly, to care about your privacy, they need you to create an account that can be tracked against every Firefox install.
False. I updated on my non-primary computer (I always like to try new things out on a non-mission-critical machine first), and it neither demanded that I log into a "Firefox Account" nor offered to create one for me. In fact, it didn't even ask either of those things.
Sync and Lockbox messages on new tab page (Score:3)
it neither demanded that I log into a "Firefox Account" nor offered to create one for me. In fact, it didn't even ask either of those things.
I've seen "Messages from Firefox" at the bottom of the new tab page for several versions, back at least to 57 (first Quantum version). One I saw today was "Our brains are not meant to hold so many passwords. That's science. Securely store your passwords with Firefox Lockbox." followed by a link to create a Firefox Account to get started. Messages about Firefox Sync also link to the account page. Might this be what AC #58630952 was referring to?
Re: (Score:2)
Re: (Score:3)
>"False. I updated on my non-primary computer (I always like to try new things out on a non-mission-critical machine first), and it neither demanded that I log into a "Firefox Account" nor offered to create one for me. In fact, it didn't even ask either of those things."
I just installed 67. It did, in fact, ignore ALL my old settings in my profile, create a new profile that is completely unusable, and gave me only ONE WAY to get my "settings" back (and it would only be some of them) and that is to creat
Re:"Doubles down on privacy" (Score:4, Informative)
>"For now, I was forced to abandon 67 completely and use 66 until I can figure out what I am supposed to do. "
I also manually wiped that new/blank profile it created and edited it out of the profiles.ini file before I reverted back to using 66.
A few hours later I just tried it a second time to use 67. And this time, unlike the first time, it offered me a small "select which profile to use" with my old one being the only one there. I did, and it worked fine. And continues to work fine across several exits and launches. I have no idea what happened.
Re: (Score:2)
Yeah, who the hell are they to have a critical bug and at the same time are trying to improve their product! /s
Excuse me, Msmash... (Score:1)
Re:Excuse me, Msmash... (Score:5, Informative)
I'm afraid you are completely and utterly incorrect about the English language.
Here's an in-depth explanation of what is going on here:
https://learningenglish.voanew... [voanews.com]
The gist of it is, nouns can be used as adjectives to modify other nouns. It's not just a rare, sometimes-nouns-do-this kind of thing either. Nouns are often used this way, in all sorts of situations. A perfect example (pulled from the link above) is race car - race is a noun describing what kind of car this car is. It's a car used in races.
"Performance" in the context of the OP's post is a noun used as an adjective to modify "features". New is not modifying "performance" here, it's modifying "features". So there are "new features" and these features are "performance features" - features that modify the browser's performance in some way. All together it's "new performance features" - new features that modify the browsers performance in some way.
This is perfectly standard English grammar. English is known to be difficult, however, so I don't blame you for being confused about a noun acting as an adjective.
Re: (Score:2)
and that's probably a good thing, the web loads "slowly" for all, some just disguise it with an unusable page jerkily loading all the bits that you have to wait to finish before use, others load them all that you have to wait to finish before use.
All this talk of a few milliseconds difference means they spend all their effort on tiny optimisations and crap instead of fixing the things that matter - like replacing javascript with something not shit.
Re: (Score:1)
Yes, FOUCing [webkit.org] is the price we pay for rendering speed.
We can all thank Microsoft for their "embrace and extend" mindset that helped make today's Internet the pileo'shit it is today (Netscape, Sun, Adobe, Macromedia, etc probably share some guilt too).
HTML didn't need to be this complicated, but we're pretty much stuck with maintaining backwards compatibility with every ill-conceived "feature" ever implemented in a web browser for decades. Otherwise, we'd have to cut off a huge chunk of the existing Internet
Re: (Score:2)
FOUCing is the price we pay for rendering speed.
Some users prefer "More flashes of unstyled content, less lag". Others prefer "More lag, fewer flashes of unstyled content". Would it make sense to expect all web browsers to expose a slider between the two extremes?
Bring it on (Score:3)
New/updated things to disable (Score:3)
For your "user.js" file:
user_pref("browser.newtabpage.activity-stream.asrouter.userprefs.cfr.addons", false); // Recommend extensions. // Recommend features.
user_pref("browser.newtabpage.activity-stream.asrouter.userprefs.cfr.features", false);
[ First one changed from ".cfr" to ".cfr.addons". ]
Anyone got anything else?
Memory running low (Score:2)
Next, Firefox 67 detects if your computer's memory is running low (under 400MB)
Can't do much with 400MB :/
Re: (Score:2)
Firefox is the only browser (Score:4, Insightful)
Everything else is a product aligned with the web at large, attempting to extract information from its users like the energy of human bodies were extracted in The Matrix.
At least Firefox cares about openness, privacy and security *first*. IMHO they are the only browser you should ever use.
Re: (Score:3)
Re: (Score:1)
Pale Moon is great and is based on Firefox. I use that as well. They're almost like the FF purists (which is funny because FF began as the "Mozilla purists").
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
"Firefox is the only browser" in the way that "Windows is the only OS". You haven't looked very far have you.
Re: (Score:2)
I guess English isn't your first language?
Warning: New versions bypass hosts file by default (Score:2)
A warning: Recent versions (not just the latest version, but at least the one prior to this) enable DNS-over-HTTPS (to a Cloudfare DNS server) by default. This has the side effect of completely bypassing the local "hosts" file. So, if you have modified your "hosts" file (e.g., to redefine tracking and/or malware domain names as 0.0.0.0), you should turn this off (via "Preferences" -> "Network Settings").
Re:Warning: New versions bypass hosts file by defa (Score:5, Informative)
A warning: Recent versions (not just the latest version, but at least the one prior to this) enable DNS-over-HTTPS (to a Cloudfare DNS server) by default. This has the side effect of completely bypassing the local "hosts" file. So, if you have modified your "hosts" file (e.g., to redefine tracking and/or malware domain names as 0.0.0.0), you should turn this off (via "Preferences" -> "Network Settings").
You can also control this in FF using the following config setting (code below for "user.js" file):
user_pref("network.trr.mode", 0);
Re: (Score:1)
It's sad how tricky this local documentation blurb is, intentionally obscuring DNS-over-HTTPs as if it were an enhancement feature instead of a router bypass. It's bad enough when viruses had JUST started hiding blocking windows update and antivirus sites thru tweaking the hostfile when it wasn't well known to powerusers and routine hostfile checks for spyware weren't a thing.
Now that we've mastered that, Windows itself has been bypassing your hostfile, and even if you're off windows the big multiplatform c