This is the shell script it runs on Mac/Linux: https://github.com/corbindavenport/just-the-browser/blob/mai...
For FireFox it downloads this: https://github.com/corbindavenport/just-the-browser/blob/mai...
{
"policies": {
"DisableFirefoxStudies": true,
"DisableTelemetry": true,
"DontCheckDefaultBrowser": true,
"FirefoxHome": {
"SponsoredStories": false,
"SponsoredTopSites": false,
"Stories": false
},
"GenerativeAI": {
"Enabled": false
},
"SearchEngines": {
"Remove": [
"Perplexity"
]
}
}
}Stupid to run random scripts you find online, but browser makers push users into it.
My son wants to eat "Chinese" food with chopsticks, but he can only really use a fork, so we adapt the chopsticks. He'll be able to use them eventually, but not everyone has a) the desire, nor b) the dexterity.
Making it easier to do what users want with a computer without telling them 'just learn to program' (or script in this case) is actually a good thing imo.
It was chrome, downloading a multi GB file without any sort of UI hints that it was doing so. A generative AI file.
Is this why chrome uses so much ram? They’ve just been pushing up the memory usage in preparation for this day, hoping I wouldn’t notice the extra software now running on my (old, outdated) system?
I’m going to attribute this to stupidity instead of malice, but it’s pretty much the height of SV arrogance to assume that unannounced multi-GB files are okay for any purpose, let alone generative AI that the user hasn’t opted into.
Some people still have shitty internet and/or bandwidth caps.
[0] "small" in comparison to ChatGPT, but still a bulky download
I can see the use of LLMs and machine learning tools like TTS, translators and grammar checkers to be integrated to browser, but only depending on local models or better, like Firefox's case to CPU optimized local models.
A lot of anti-AI backlash seems to exempt machine translation, which as far as I can tell is just because it's been around for so long that people are comfortable with it and don't see it as new or AI-y, which imho spells doom for a lot of this- in ten years automatic tab groups will seem just as natural and non-intrusive as machine translation.
A local LLM that I explicitly bring up to ask a question and dismiss (ie no CPU or RAM usage) when I'm done consulting it is nice. A piece of software I'm using interrupting what I'm doing to ask me a useless and annoying question or to make an unsolicited change to my workspace leaves me thinking about permanently uninstalling it.
I will never want automatic tab groups or automatic anything else. I don't even want an "integrated" desktop environment - I use i3 to get away from that. I hate all the useless bullshit half baked features that are constantly shoved in my face.
If the modern web was compatible with it I'd use a text based browser for 90% of what I do online. And if that were the case I'd still welcome a built in machine translation feature because it's an incredibly useful tool.
It's much more efficient on system resources than the larger LLMs downloaded by browsers for other tasks.
Do you also want to remove the ML that gets the results you want at the top of the address bar autocomplete. That's been around for 15 years and it's "AI" so might as well get rid of it right?
This "all ML sucks" because generative AI LLMs suck has to end. It's entirely a garbage take.
Mozilla decided to give businesses the ability to turn off generative AI features about 6 months ago.
There is Harper as local grammar checker; an amazing project, but it is only in English and not yet able to replace the mentioned tools: https://writewithharper.com/
Also there's absolutely zero need to be sudo to put a JSON config file for Firefox on Linux.
You're basically bash/curl'ing the kitchen sink, with all the security risks that entails, executing a shell script as root (which may or may not be malicious now or at some point in the future), just to...
Put a 12 lines JSON file in a user's Firefox config folder.
Way to go my "fremen" brothers [1].
[1] the "fremen" in Dune as those who adore the Shai-Hulud
Sure, Authenticode signing certificates aren't always cheap, and signing your script doesn't protect the script from compromise without other good security practices, but it would still show some attention to detail on PowerShell and some attempt to avoid malware compromising your script.
[1] https://learn.microsoft.com/en-us/powershell/module/microsof...
https://news.ycombinator.com/item?id=46616033
Interesting project.. and props for spending the time to figure out all those settings and how to flip them off (for all 4 major browsers too!)
I like the goal of stripping browsers back to basics, but I'm not sure why I'd run a third-party script to flip low-level browser and system settings I can change myself.
From a security point of view, that feels, not great?
This might work better as a simple guide with screenshots, so people can see and control exactly what’s being touched.
The need for this is mainly on work machines that are locked down; if admin mode is necessary then it's DOA...
A local MITM proxy that doesn't require elevated rights and which filters out everything unwanted, starting with ads, would be nice I think.
I grew up on DOS, and my first browser was IE3. My first tech book as a kid was for HTML[1], and I was in absolute awe at what you could make with all the tags, especially interactive form controls.
I remember Firefox being revolutionary for simply having tabs. Every time a new Visual Basic (starting with DOS) release came out, I was excited at the new standardized UI controls we had available.
I remember when Tweetie for iPhone OS came out and invented pull-down refresh that literally every app and mobile OS uses now.
Are those days permanently gone? The days when actual UI/UX innovation was a thing?
[1] Can someone help me find this book? I've been looking for years. It used the Mosaic browser.
Some stuff has been solved. A massive number of annoyances in my daily life are due to people un-solving problems with more or less standardized solutions due to perverse economic incentives.
The labyrinth of ways to interact with the temporal path between pages is a cluster. History, bookmark, tab, window,, tab groups.
There are many different reasons to have a tab, bookmark, or history entry. They dont all mean the same thing. Even something as simple as comparison shopping could have a completely different workflow of sorting and bucketing the results, including marking items as leading candidate, candidate, no, no but. Contextualizing why I am leaving something open vs closing it is information ONLY stored in my head, that would be useful to have stored elsewhere.
Think about when you use the back button vs the close tab button. What does the difference between those two concepts mean to you? When do you choose to open a new tab vs click? There is much to be explored and innovated. People have tried radical redesigns, havent seen anything stick , yet.
There's a similar amateurs-do-too-much effect with typography and design. I studied typography for four semesters in college, as well as creative writing. The best lessons I learned were:
In writing, show, don't tell.
In typography, use the type to clarify the text - the typography itself should be transparent and only lead to greater immersion, never take the reader out of the text.
Good UI follows those same principles. Good UX is the UI you don't notice.
The UX is also awful.
But I think this is a compounding problem that spans generations of applications. Consider the page convention — a great deal of the writing content we typically publish, at a societal level, will be digital-only so why are we still defaulting to paper document formats? Why is it so fucking hard to set a picture in?
And it's that kind of ossification and familiar demand that reinforces the continuum that we see, I think. And when a company does get creative and sees some breakthrough success it is constrained to nascency before it gets swallowed by conglomerate interests and strangled.
And Google's alternative ecosystem has all of these parallels. It's crazy to see these monolithic companies floundering like this. That's what I don't understand.
I'm forced to use WhatsApp for a local group, and for some reason, when in the group chat, when I pull up to ensure that I see the latest message, that stupid app opens an audio-recording thingy at the bottom as if I wanted to send an audio note to the group.
Who designed that? Has that person been fired?
Also, I wish that on Windows "windows" weren't able to provide their own chrome and remove the title bar. Add some things to it yes, but fully replace it? No thank you.
I think "yes" and "a bit", in that order. The early days of the web and mobile, where everything was new, are gone. In those days, there was no established pattern for standard UX. Designers had to innovate.
It makes sense that we have a lot less innovation now. There's probably room for a lot more than we see, but not for the level that was there in the early days of the web.
There's no reason to "learn" a UI or use shortcuts on most sites, because they change everything around every few months.
I see people reminiscing about tabs in firefox, well today a majority of the top websites don't even allow you to open links in new tabs! The links aren't even real links anymore, and everything's a webapp. ( and by top websites, I mean social media, not the top sites used by the HN crowd. Sites like YT, FB, IG, and TT ).
I try to interact with the "UI" of websites as little as possible these days. I use RSS readers for as much as possible. Any time I get a popup on any site, I get mad. I don't care about news updates, software updates, or offers. Anything that pops up at me, or moves around before I can click it, looks like a scam to me. Even if it's "legitimate". The modern web feels like an arcade game that's trying to waste my time.
Would it happen to be HTML Manual of Style: Clear, Concise Reference for Hypertext Markup Language by Larry Aronson? [1]
From the description:
> This book introduces HTML, the program language used to create World-Wide Web "pages", so that users of Mosaic and other Web browsers can access data. Forty to 50 new "pages" are being added to the WWW every day and this will be the first book out on the subject.
Forty to fifty new "pages" per day! </Dr. Evil air quotes>
To an extent, yes. The ecosystem has matured. The things that work have been discovered, the things that don't have been discarded.
I think it'll take another big leap in hardware form factor (Apple Vision being an example of an attempt at it) for us to see meaningful UI changes.
Browsers could start by simply improving the controls they do ship with, such has date pickers and selects. They’re all shit. Slightly more complex perhaps would be a combobox. LMAO, we don’t even have comboboxes.
And if you really want to get fancy, rich-text textareas that return standardized, semantic HTML. Also, decent tables with sorting/filtering wouldn’t go amiss.
Standardize some HTMX features into HTML while you’re at it, you’ve got a full-blown revolution.
But it would be funny if it's this: https://archive.org/details/teachyourselfweb00lema/page/n9/m...
- https://www.goodreads.com/book/show/11177063-creating-cool-w... - https://www.goodreads.com/book/show/1097095.HTML_for_Dummies...
Why would it be funny though? Am I missing something?
I agree mostly with your sentiment. But I still think there is still some work being done. For example the Arc and Zen Browsers. I never used Arc because it is closed source. But it sure looked beautiful. And Zen I tested, but it seemed laggy. I think I might give it another go to see if some of the performance issues have been fixed.
There is more than enough of it. Now it is, of course, AI agents. Before that, Material Design was quite innovative. Interestingly, with the raise of search engines and later LLMs, we are getting back to the command line! It is not the scary black window where you type magic incantations, it is a less scary text field where you type in natural language, but fundamentally, it works like a command line.
It is a good thing? For me, it is a mixed bag, I miss traditional desktop UIs (pre-Windows 8), but I like search-based UIs on the Desktop, an I am not a fan of AI agents: too slow an unpredictable, and that's before privacy considerations. When it is not killing performance, I find Material Design to be pretty good on mobile, but terrible on the desktop. That there is innovation doesn't mean it is all good.
Yes. When coming from DOS, all the UI/UX that could have been created has been created. What we have now is a loop of tries to refresh the existing but it's hard, mainly because it's now everywhere and it has reached maturity.
As an example, the "X" to close and the left arrow for back won't be replaced before a long time, just like we still have a floppy to represent save.
Cars have tried to refresh their ui/UX but they failed and are now reverting back to knobs and buttons.
It seems that VisionOS is a place where innovation could come but it's not really a success.
Designers decided that scrollbars that shrink to super-thin columns when not in use were better. Maybe... but often it results in shrunken scrollbars that require extra work to accurately hover over and expand.
Designers decided that gray text on gray backgrounds were easier to read, and there was even a study to "prove" it... which resulted in idiots picking poor contrast choices of gray-on-gray, without understanding the limits on this idea.
I will say that the current push for accessibility is forcing some of these "innovations" back onto the junk heap where they belong. I was annoyed the first time an accessibility review complained about the contrast of my color choices on a form once... but once I got over my ego, I have to admit they were right; the higher-contrast colors are easier to read.
In some ways, this is happening at the moment with AI and LLMs. The tools available, how we prompt them, etc are all "UI/UX innovation" if you believe these things have a use.
If we have a huge platform shift in the future (LLMs, AR/VR, ???), we may start from zero and go through "inventing tabs" again until that platform becomes maximally optimised.
No. You just need to look outside of desktop computing, and computing in general.
For example, I'm getting into CAD and 3d printing. Learning it reminds me of when my father learned to program in the late '80s, or when my grandfather telling me about how he got his Model A up to 50 mph.
Remember: Desktop computers and the web are ultimately tools for a purpose, and that purpose isn't always "nerd toy." We (the nerds) need to find and invent our toys every generation or so.
Paradigms for existing forms of computer interaction (keyboard, mouse, touch) are pretty much solved.
It's still a thing but it went off the rails, see Apple and their latest no-contrast UI.
Apple has the unfortunate burden of needing to shepherd millions of developers over to this new paradigm (AR) before it really exists, and so is shoving Liquid Glass onto devices that don't really benefit from it.
But in practice people are generally not happy about lots of new experimentation going on. By definition, most of the results suck. In retrospect we get to stand in awe of those that survived the evolutionary battle and say "wow browser tabs" and "wow pull to refresh" and forget the millions of other bad ideas that we tried.
Like the AOL browser, come to think of it.
Tabs in Firefox were such an unfamiliar thing.
New apps were announced in blogs, and people downloaded them to try them out. I remember downloading Opera, using it for a few days or weeks, and then going back to Firefox.
Chrome's Whats New seems like half AI stuff and half UI features for people who have tons of tabs.
The standard affordances for most well-known problems are long settled. Unless you're solving an entirely new class of problem, maybe you don't need to reinvent a large number of wheels, again. We're all tired of the triangular wheels coming out.
Which makes it funny that the request for UI innovation is prefixed with a quote that amounts to "but what if browsers were permanently frozen ca. 2012?". Mind, I can sympathize with some of the thoughts behind the request, even if I disagree - but you can't ask for a stop in new features & problem classes to be accompanied by continued UI innovation.
That is, as my art teacher used to say, "intellectual wankery in the disguise of creativity".
I don't think these are permanently gone, but the corporations failed us, and also the "not for profit" fakers such as Mozilla.
We need a new web - one developed by the people, for the people. Whenever corporations jump in, they try to skew things to their favour, which almost always means in disfavour of the people.
IMO this should never have existed. if X or whatsapp or some site wanted pull-down to refresh they can implement it. 99.99% of sites do not need it.
Now, WhatsApp have a pull-down feature that starts a voice note or voice chat or something ... it's awful, if you scroll down in a chat it is really easy to trigger by accident.
They also have a big button at the bottom right to start some sort of recording. Were they trying to get people to start recordings by accident? Does that help them somehow?
- Use of analytics tends to replace user trials/interviews entirely, trading away rich signals for weaker ones
- Analytics can be used to justify otherwise unpopular or ill-advised changes
- When combined with certain changes (e.g. making features harder to access), the numbers can be “steered” in a particular direction to favor a particular outcome and better enable the last point (“Looks like nobody’s using that thing we hid behind an obscure feature flag! Guess we’re safe to remove it entirely now!”).
In theory telemetry/analytics have strong potential for improving software quality, but more often than not they’re just massaged and misused by product managers bent on pushing the software a particular direction.
Telemetry in the hands of stakeholders whose stakes are business/career KPMs will probably serve those, and the software experience will follow.
All this to say, I don't think Mozilla is doing much with all the telemetry data it's gathered all these years
These Edge changes are important if you value your browsing privacy. Not sure any of the major browsers completely protect you, but Microsoft has just gone all out in finding ways to scrape your browsing habits.
I would however download a new browser that promises to not have all these bad features and has stripped them straight from the source code. For example, I switched from Chrome to Brave because it blocks ads.
Is there a way to persist the file even after updates?
Sounds like the beginning of a nice ClickFix campaign: https://it.lbl.gov/the-clickfix-attack-a-new-threat-to-your-...
This is cool! I was expecting a script, which tend to be brittle. This is a great way to do it.
Helium seems to be trying to be the same thing for Chrome - it’s replaced Brave as my go-to for the sites that have issues with non-Chrome browsers.
"Something that hasn't been made clear: Firefox will have an option to completely disable all AI features. We've been calling it the AI kill switch internally. I'm sure it'll ship with a less murderous name, but that's how seriously and absolutely we're taking this."
“AI must be clear and understandable. Controls must be simple. AI should always be a choice — something people can easily turn off.”
Should be Ctrl+Shift+V
Search for the Terminal in your applications list and open it.
Next, copy the below command, paste it into the window (Ctrl+V or Cmd+V), and press the Enter/Return key:
& ([scriptblock]::Create((irm "https://raw.githubusercontent.com/corbindavenport/just-the-browser/main/main.ps1")))
I guess then, the browser and AI just serve different purposes now?
For anyone else on firefox, save yourself some effort and just download this https://github.com/corbindavenport/just-the-browser/blob/mai...
source: https://en.wikipedia.org/wiki/Chromium_(web_browser)#Free_an...
I haven't found any way to programatically create profiles.
And you might as well just fork chromium for that purpose.
Yes browsers should be used for browsing, those half websites can run on something else.
Google and others really ruined the web.
I also today tried Qwant and for the first time, in a long while, the results Qwant delivered were objectively better than from Google Search. What the heck is Google doing?
It seems like Qwant is ad supported[0], yet I don't see any ads in my first couple searches. I wonder if this is a, "the first hit is free", situation, or my ad blocker just took care of it. I do wonder how this will play out long-term.
Qwant did bring up a page when I tried the second search to make me slide something to verify I'm human. That was enough of an annoyance that I will stick with Kagi.
[0] https://help.qwant.com/en/docs/overview/how-does-qwant-make-...
Inflating stock prices.
This is an extremely aimless rant. Simply claiming group policies are not enough for an average user or at the very least is not a good start, is misleading. Unless you can back it up with data, your comment is in bad faith.
The whole "average user" agenda is already a smell. Nice to see you writing your first non-question here.
"Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."y
Browsers are currently incentivised to add a bunch of new features outside their traditional role. Some people prefer to keep the browser's role simple. It's not ideological and it's not "hating".
Microsoft and Google I can understand, they have AI products they desperately need to monetize or push to as many users as possible, because management bonuses are tied to CoPilot or Gemini adoption.
I don't see it as hating on AI, just because it's AI. It's not wanting pointless AI features in products that don't need them. I've pretty much disabled anything in the ml namespace in about:config in Firefox, because the features are distracting, but provide absolutely no value to me.
why not? All things being equal non-AI solution is better. "it is current hyped thing" should bring some downward correction
and of all things to hate, AI hate is harmless and at least partially justified
Microsoft shoving LLMs into literally everything, including Notepad, is what people are currently hating, because it isn't quite ready.