hckrnws
I had a look at what it actually does in the Firefox settings and all it seems to do is to disable one AI feature flag, change the default search engine, and then set a few other flags that are changes that you may or may not want to make, unrelated to AI. Not sure you want to run a 3rd party shell script just to do that…
For anyone interested
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"
]
}
}
}The second sentence in the Getting Started section invites you to follow the manual guides instead of running the 3rd party shell scripts. I think this is a good way to do it -- have both options and tell people about them right at the start of the process. Is there some other way you wish they'd share this info?
There seems to be quite a lot missing from the Chrome configuration too
It's a pretty simple project, and the license is MIT. Wouldn't be hard to let them know what's missing: https://github.com/corbindavenport/just-the-browser
Are all options available to the group policy? Since this is not directly modifying your app and merely creating a group policy for the browser to use, there might be some things not able to be set there. I have not experience with these group policies. Just thinking of why something might be missing as you stated. It could also be considered out of scope for the dev making this project.
Seems like this is for the people that need to execute random powershell scripts they don't understand in order to turn of telemetry and copilot on Windows because reading about the registry and group policy is too much for them.
The Getting Started section invites you to follow the manual guides instead of running the scripts. That's what I did, and I really appreciate the site/guides.
I mean, yes ...?
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.
> users want with a computer without telling them 'just learn to program'
A computer is meant to be programmed by the user. That is its raison d'être from the very beginning and why it is called like that.
Some people want computers, some people just want to use them like appliances. The bigger problem is those companies who want to control other people's computers no matter which type of person they are.
A mobile phone is meant to be used to make and receive telephone calls. That is its raison d'être from the very beginning and why it is called like that.
Do you see how stupid that sounds?
It... doesn’t? PDAs won, there are nearly no mobile phones being made anymore. The thing you call a mobile phone is a PDA with an incidental value-add of placing phone calls. (I had been wondering for some time why the UX of talking into a headset had been on the decline for the last 20 years until I encountered this idea; sadly, I don’t remember where.)
That's exactly my point?
Seems like you're using your computer wrong by posting here then. In fact, 99% of people are using their computing devices wrong all the time. The computer must be the most misused tool in the world.
With all due respect, I hope you never touch the development of any piece of software any of my relatives or friends ever has to use.
Good UX is one of the most important-yet-underserved areas in the tech industry (the topic of this site), and this sort of attitude goes beyond being smug and naive to being actively harmful. Your goal should always be to make things easier and with as little friction as necessary.
Comment was deleted :(
It's not hard to search for a few keys in the about:config menu or to set a group policy. If you can't be bother to do this you have zero business running random scripts that update your system configuration that you have no idea how it works.
Normie users would be better off reading some detailed step-by-step instructions on how to do it by hand using built-in methods than to run random code from the internet that can be malicious.
My mom is 75 years old and barely knows how to use a web browser to begin with. There is zero chance I encourage her to run random pwsh scripts from the internet.
God forbid we're going to start giving them AI agents to do this kind of stuff for them. God help us.
Yes, if someone wrote a guide on how to do it and why, that’d be great.
Knowing where to look and which settings are relevant, for yourself, is a crazy ask of even very computer savvy users.
> a crazy ask of even very computer savvy users
Why?
When Mozilla updated Firefox with the AI chatbot feature the first thing I did was look in settings in how to remove it. When that failed I just googled it, which pointed me to about:config and which keys to look for.
Much easier to figure out with your intuition what `browser.ml.chat.enabled` could possibly mean than running pwsh script.
All it takes is a bare minimum of curiosity.
If you're this paranoid then you can't really trust any piece of software. Many "random" shell scripts that update your system config are more well vetted than 90% of the software you run on your computer daily.
You should trust software that you can verify yourself as safe, or software written by people who you trust not to abuse the power you're giving them over your device by allowing them to modify it.
Personally, I don't trust most popular software either, but its easy to see why people would be fooled into thinking that software written by a major corporation used by millions of people might be more trustworthy than a script uploaded by a random anonymous person who couldn't be held accountable if their software infested your system with malware.
I agree. But I'm just surprised that you'd be extremely wary of running a sub 100 line open source script as a one time operation that you can easily audit yourself but on the other hand are likely using a browser that no one in the world (not even the developers) has fully audited.
There are "detailed step-by-step instructions" linked in the second sentence of the Getting Started section. I'm not sure what more you could want, besides perhaps making it more foolproof against people who can't be bothered to read.
Presumably you grow all your own food, cook from scratch; manufacture your own tools; refine your own fuel; mine your own lithium for batteries; produce your own fibres to weave at home, make cloth, which you tailor your own clothes from; grow trees, fell them, cut them to boards, and make your flooring and furniture; and so on?
Otherwise, you're the sort of normie carpenter who doesn't even do their own land clearance ready for seeding. For whom you must express your utmost contempt! /s
As far as I can see all these Firefox options can be set through the UI as well.
Comment was deleted :(
A few weeks ago I noticed some mysterious app was killing my (poor) internet downloading a large file.
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?
It's an AI model file, part of an upcoming local AI API. It can be very useful for on-device AI though, stuff like parsing and analyzing text from an image such as a receipt of groceries, which is actually an app I'm making currently, so it helps if everyone has such a model.
It would be absolutely lovely if Google asked users to opt-in rather than force a large file download on them.
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.
This is why I use Firefox. In case where I have to use something chromium based (because Cloudflare hates Firefox, apparently), I use Brave.
Firefox also downloads and uses AI models: https://support.mozilla.org/en-US/kb/on-device-models
I don't believe it ever does so without first asking for permission?
It did it without my permission, it's a small model to rename and group tabs, though, no more than 50M in size.
wait, hang on, I can't be the only person with 300 tabs open for whom that would actually be useful
Use seamonkey or palemoon
Does seamonkey still work on modern internet? Last time I tried it to open Notion, nothing was displayed. Works great on old world stuff though.
It should, it has some issue rendering some sites (Google's AI overview acted funny but if you still use google you aren't their intended audience anyway) but otherwise works fine on content focused sites. Youtube works iirc, idk about adblock though (theres uBlock legacy but that hasn't been updated in a bit)
Interesting. Do you have more specifics? I don't use Chrome or it's derivatives, but this is the first I've heard of it doing that.
Chrome can run small[0] models and can auto-download them: https://developer.chrome.com/docs/ai/built-in
[0] "small" in comparison to ChatGPT, but still a bulky download
Yeah, seems to be about 4gb or so in disk space for the models in my situation.
Don't ever use Chrome. Go for Brave or Vivaldi, they're far better than Chrome.
Doesn’t Brave has their own ‘Leo’ AI built in?
Yeah but I never use it. Also you can disable it easily.
Those are propietary, actually as worse as Chrome if not more.
Does it also remove Firefox's translation models that uses local CPU? I find that feature very useful and totally obliterated my dependence on Chrome's translate features. Models are surprisingly good, especially for languages like English, Spanish and German.
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.
It explicitly doesn't, though they don't explain why not. It's not an on/off device distinction because it disables Firefox's automatic tab groups too.
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.
It's not mere familiarity. Machine translation is immediately useful to me. I was going to pull up google translate anyway; keeping it local to my device improves both convenience and privacy.
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.
Firefox's translation by default does pop up to interrupt and ask if you want to translate a page that it's detected is in another language. We're just more used to that and it's a more reliable signal that you probably want to run a tool than most.
It's still relatively new in FF and I don't think I've seen anyone complaining about it annoying them with popups, even though it absolutely does throw up an interrupting overlay, especially on mobile.
I definitely complain about this one. I can read a few languages, and rarely if ever browse a page in a language I don't understand, so popups with "do you want to translate this" are unwelcome here. It doesn't help that in the first iterations Firefox didn't offer a quick way to turn the whole thing off.
You can disable the popup but still invoke the tool manually from the main menu. I reaffirm my previously expressed dissatisfaction with modern software "features" and add that there are plenty of defaults in Firefox that I personally dislike. That includes anything that pops up unsolicited without good reason.
From a UI perspective, auto tab groups are just an extra button as far as I can tell, so it's not clear why it's getting the axe from this site, just from a pure "this is annoying" point of view.
The flow is 1) you drag a tab over another tab and it suggests a name for the tab group and 2) you click on a tab group and another button offers to suggest more of your tabs you can add to the group. That's less intrusive than Firefox Translations are by default.
The local translation in Firefox is an NMT model, not an LLM: https://github.com/mozilla/firefox-translations-models?tab=r...
It's much more efficient on system resources than the larger LLMs downloaded by browsers for other tasks.
AFAIK the model for tab group suggestions is about the same size? (And I think that model is/can be used for other features as well?)
Thanks for clarification!
It's still a transformer architecture though, right?
These are tiny purpose built models with simple and safe use cases.
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.
It doesn't because the enterprise group policy doesn't which is what this sets.
Mozilla decided to give businesses the ability to turn off generative AI features about 6 months ago.
Indeed. For translation, local LLMs are a straightforward privacy win. Not that they're getting much credit for that.
Isn't it? I am totally looking forward to a grammar checker that can compete with DeepL or Grammarly that can run locally and not heavy on resources. This will complete the holy trinity of local and free natural language editing; translations, grammar/spell checker, thesaurus/dictionary!
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/
Suggesting bash/curl'ing to get a 12 lines JSON file is just... Not great. We've seen a shitload of developers account getting compromised (with all the supply chain attacks) and developers account turning evil.
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
Administrator access or sudo is required because the configuration paths (C:\Program Files\Mozilla Firefox on Windows, /etc/firefox/policies/ on Linux) are protected. The browser guides explain the manual install and uninstall process for anyone who doesn't trust the script.
I have firefox installed on Linux. There is no /etc/firefox/policies/ dir, nor indeed even an /etc/firefox/ dir. Therefore, no need for sudo.
The /etc directory and everything under it is protected.
Of course it is. But there's no point trying to put policies into a directory in a tree which doesn't exist and by extension which Firefox won't be reading.
In Linux (and in any sane system) there is no need for elevated privileges just to alter your browser settings.
Firefox will read it if it exists[1]. You could use the /usr/lib/firefox/distribution directory (or whatever the installation directory may be), but that may be overwritten by an update.
There doesn't seem to be any way to set per-user group policies, so unless you're installing firefox in a user-controlled directory, it will require elevated privileges.
[1]: https://support.mozilla.org/en-US/kb/customizing-firefox-usi...
I feel this way even more about this approach to PowerShell scripts. At the very least you could code sign [1] your script. Most PowerShell devs run with the execution policy RemoteSigned for multiple reasons. Instead of pulling your script down from the web and dangerously converting it to a string to call (with &) you can download it as a real file and Invoke-Command (icm) it. Dangerous eval versus correctly running a script.
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...
There's even talk in the PowerShell world about Invoke-Command directly accepting HTTPS URLs to scripts, but in addition to some security questions, it is caught in a catch-22 that not enough of these scripts are Signed so there's not enough demand for it, but if more tools like this were doing code signing there's a world where the code golfed instructions are just `icm https://yourdomain.com/some/script.ps1`, and it is more secure than these examples with (`&` or `iex`) and `iwr`.
Copying my comment from last time
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.
> Windows: Open a PowerShell prompt as Administrator
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.
> aims to remove: Most AI features, Copilot, Shopping features, ...
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.
I feel like wishing for UI innovation is using the Monkey's paw. My web experience feels far too innovative and not enough consistent. I go to the Internet to read and do business not explore the labyrinth of concepts UI designers feel I should want. Take me back to standards, shortcuts, and consistency.
Yes! I don't want a car with an "innovative" way of steering. I don't want a huge amount of creativity to go into how my light switches work. I don't want shoes that "reinvent" walking for me (whatever the marketing tagline might say).
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.
> I don't want a car with an "innovative" way of steering.
99.5 % agree, because I would love to try SAAB:s drive-by-wire concept from 1992: https://www.saabplanet.com/saab-9000-drive-by-wire-1992/
The thing why this was only a research project and never came into mass production was regulatory stuff, IIRC? (most EU countries require, still until today, a "physical connection between steering wheel and wheels" in their trafic regulation)
This was a few years before Sweden joined the EU, but yes, I think the lack of a physical connection was one of the main problems.
From what I've read the test drivers also thought the car was too difficult to drive, with the joystick being too reactive. I wonder how much of that could be solved today with modern software and stability control tech.
I can't find it now, but I do remember a similar prototype with mechanical wires (not electrical) that was supposed to solve the regulatory requirements. That joystick looked more like a cyclic control from a helicopter.
Having played enough video games that use joysticks for steering I don't want to drive a real car with a joystick. Crashing in Mario kart or Grand theft Auto because I sneezed is fine but not in real life.
Exactly. The control needs to have both an intentional and major motor movement from the driver. Modern steering wheels have the same benefit as the original iPod wheel. Easy for small movements, even accidental ones; possible for big movements.
Also funny that they had the ability to swap to the passenger to drive it. So acceleration/break for one person, steering for another? Really not a good idea.
I think there's a ton of innovation left to be done regarding steering and light switches.
You're right that it's not going to be better designs, but paradigm shifts.
We still don't know what it means to provide input to a mostly self-driving car. It hasn't been solved and people continue to complain about attention fatigue and anxiety. Is the driving position really optimal for that? Are accident fatalities reduced if the driver is sitting somewhere else? Even lane assist still sucks on traditionally designed cars. Is having to fight a motorized wheel to override steering really all that safe?
Light switches may be reliable and never go away, but we have many well-established everyday examples of automatic lights: door switches, motion sensing, proximity sensing, etc. You never think about it and that's the point.
> Yes! I don't want a car with an "innovative" way of steering.
You might, but you'll never really know.
I mean, steering wheels themselves were once novel inventions. Before those there was "tillers" (a rod with handle essentially)[0], and before those: reigns, to pull the front in the direction you want.
I highly doubt there's a steering input device so superior to the current wheel shape that it's worth throwing out the existing standard. Yes, at one point how steering should work (or how you should navigate the Web) was uncertain, but eventually everyone settled on something that worked well enough that it was no longer worthwhile to mess with it.
Although, one thought I had is that there's nothing wrong with experimenting with non-standard interfaces as long as you still have the option to still just buy, say, a Toyota with a standard steering wheel instead of 3D Moebius Steering or whatever. The problem is when the biggest manufacturers keep forcing changes by top-down worldwide fiat, forcing customers to either grin and bear it or quit driving (or using the Web) entirely.
I sympathise with the frustration, but I think the issue isn't innovation itself: it's that we've lost the ability to distinguish between solving actual problems and just making things different.
Take mobile interfaces. When touchscreens arrived, we genuinely needed new patterns. A mouse pointer paradigm on a 3.5" screen with fat fingers simply doesn't work. Swipe gestures, pull-down menus, bottom navigation—these emerged because the constraints demanded it, not because someone thought "wouldn't it be novel if..."
The problem now is that innovation has become cargo-culted. Companies innovate because they think they should, not because they've identified a genuine problem. Every app wants its own navigation paradigm, its own gesture language, its own idea of where the back button lives. That's not innovation, that's just noise.
However, I'd have to push back on the car analogy: steering wheels were an innovation over tillers, and a crucial one. Tillers gave you poor mechanical advantage and required constant two-handed attention. The steering wheel solved real problems: better control, one-handed operation, more space for passengers. It succeeded because it was genuinely better, and then it standardised because there was no reason to keep experimenting.
The web needs more of that approach: innovate when there's a genuine problem, then standardise when you've found something that works. The issue isn't innovation, it's the perverse incentive to differentiate for its own sake.
Leaving aside the externalities of constantly breaking everyone's workflow and potentially introducing disastrous bugs, there's an opportunity cost to innovating where there isn’t a clear need. Google and others are wasting massive resources endlessly tweaking browsers and the Web because that's all they know how to do, their users are locked in and without recourse, and they don't feel threatened by any competitors or upstarts. I would argue the web and smartphones and similar tech are boring now but because the market is controlled by only a few huge companies, the tech hasn't been allowed to become low-margin, standardized cookie-cutter commodities. Instead these attempts to make this old boring tech seem exciting is getting to the point where it's sad and comical.
Your last paragraph reminded me of HTML5 and the WHATWG which led to official W3C adoption.
Back when that started W3C was still strongly embedded in the XML hellhole.
You need to be careful here, because we have a real tendency to get stuck in local maxima with technology. For instance, the QWERTY keyboard layout exists to prevent typewriter keys from jamming, but we're stuck with it because it's the "standardized solution" and you can't really buy a non-QWERTY keyboard without getting into the enthusiast market.
I do agree changing things for the sake of change isn't a good thing, but we should also be afraid of being stuck in a rut
I agree with you, but I'm completely aware that the point you're making is the same point that's causing the problem.
"Stuck in a rut" is a matter of perspective. A good marketer can make even the most established best practice be perceived as a "rut", that's the first step of selling someone something: convince them they have a problem.
It's easy to get a non-QWERTY keyboard. I'm typing on a split orthlinear one now. I'm sure we agree it would not be productive for society if 99% of regular QWERTY keyboards deviated a little in search of that new innovation that will turn their company into the next Xerox or Hoover or Google. People need some stability to learn how to make the most of new features.
Technology evolves in cycles, there's a boom of innovation and mass adoption which inevitably levels out with stabilisation and maturity. It's probably time for browser vendors to accept it's time to transition into stability and maturity. The cost of not doing that is things like adblockers, noscript, justthebrowser etc will gain popularity and remove any anti-consumer innovations they try. Maybe they'll get to a position where they realise their "innovative" features are being disable by so many users that it makes sense to shift dev spending to maintenance and improvement of existing features, instead of "innovation".
> For instance, the QWERTY keyboard layout exists to prevent typewriter keys from jamming, but we're stuck with it because it's the "standardized solution" and you can't really buy a non-QWERTY keyboard without getting into the enthusiast market.
So, we are "stuck" with something that apparently seems to work fine for most people, and when it doesn't there is an option to also use something else?
Not sure if that's a great example
Sometimes good enough is just good enough
> the QWERTY keyboard layout exists to prevent typewriter keys from jamming
even if it is true (is it a myth by any chance?), it does not mean that alternatives are better at say typing speed
As someone that makes my own keyboard firmware, 100% agree. For most people, typing speed isn't a bottleneck. There is a whole community of people that type faster than 250wpm on custom, chording-enabled keyboards. The tradeoff is that it takes years to relearn how to type. Its the same as being a stenographer at that point. Its not worth it for most people.
Even if there was a new layout that did suddenly allow everyone to type twice as fast, what would we get with that? Maybe twice as many social media posts, but nothing actually useful.
I'd imagine at this point that most social media posts are done by swiping or tapping a phone's virtual keyboard (if one is used at all).
One don't need to be a scientist to take a look at own hands and fingers, to see that they are not crooked to the left. Ortholinear keyboard would be objectively better, even with the same keymap like QWERTY, but we don't produce those for masses for a variety of reasons. Same with many other ideas.
> to see that they are not crooked to the left
how it makes ortholinear keyboards better?
If I recall correctly, QWERTY was designed to minimize jamming. The myth is that it was designed to slow people down.
Whether it does slow people down, as a side effect, is not as well established since, as another person pointed out, typing speed isn't the bottleneck for most people. Learning the layout and figuring out what to write is. On top of that, most of the claims for faster layouts come from marketing materials. It doesn't mean they are wrong, but there is a vested interest.
If there was a demonstrably much faster input method for most users, I suspect it would have been adopted long ago.
Comment was deleted :(
It's been debunked by both research (no such mention at the time) and practice on extant machines.
These days QWERTY keyboards are optimal because programs, programming languages and text formats are optimized for QWERTY keyboards.
Depends on the language no? Qwerty isn't great for APL.
I have a QWERTZ keyboard!
Is my digital life at a natural end now?
If you mean the default German keyboard layout then, yes, putting backslashes, braces and brackets behind AtlGr makes it sub-optimal in my book. Thankfully what's printed on the keys is not that important so you too can have a QWERRTY keyboard if you want.
I wish for browser ui innovation.
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.
If you expect the browser to help you manage your various workflows beyond generic containers (tabs, tab groups), then you become tied into the browser's way of doing things. Are you sure you want that?
I'm not saying your hopes are bad, exactly. I'm interested in what such workflows might look like. Maybe there _is_ a good UX for a web shopping assistant. I have an inkling you could cobble something interesting together quite fast with an agentic browser and a note-taking webapp. But I do worry that such a app will become yet another way for its owner to surveil their users in some of the more accurate and intimate areas of their lives. Careful what you wish for, I reckon.
In the meantime, what's so hard about curating a Notepad/Notes/Obsidian/Org mode file, or Trello/Notion board to help you manage your projects?
shopping assistant was a specific example, but in the process of research, brainstorming, etc theres a bunch of different ways id like to see visualization and record of how i got somewhere, what was discarded, summary of what was retained, whats coming next, options for branching.
the web is a document structure, but browsing it doesnt need to be linear.
We had that ability in Firefox, through XUL. Then it was removed. Tree Style Tab addon doesn't work properly to this day because of this.
We had that ability in Chrome, through Chrome Apps. You could make a browser app, load pages in webviews, with the whole browser frame customizable. Then it was removed.
We had an ability to make a new innovative browser, until Google infested all the standartization committees, and increased complexity of standards on a daily basis for well over a decade. Now they monetize their effort on making Chrome by removing adblockers and enforcing their own ads, knowing full well that even keeping a fork that supports manifest v2 is infeasible for a free open-source project.
There is no way forward with the web we have right now. No innovation will happen anymore.
Kinda yeah, kinda no. Big-thinking drastic UI experiences are usually shit. But small, thoughtful touches made with care can still make a big difference between a website that just delivers the data you need and one that's pleasant to interact with.
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.
It definitely feels like it is gone. Of course I'm largely talking about the applications that I use, e.g. MS Word which is still using the searchless 1980s character map and has a crazy esoteric add-on installation process. It's hilariously bad when we consider the half-screen UI which obscures a considerable amount of the ribbon.
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.
> invented pull-down refresh that literally every app and mobile OS uses now
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.
You need to try Telegram. Not because it’s better but because it’s more insane. The QR code you use to add people when you meet them is in Settings
No it isn't. Hamburger menu > My Profile > QR button is next to your username.
On iOS it's in Settings (in the context of meeting people, you're not going to pull out a desktop)
We don't want an insane experience, it's an HOA group. That's all I use WhatsApp for.
Also, I despise telegram (just as much as X), because in Germany both are rotten to the core in terms of user base, worse than WhatsApp.
Signal or Threema would be great, and I voted for Signal, but the majority uses WhatsApp.
I used to use Telegram, but ever since Covid and the whackos that found their "truth" over there I say no thank you.
You don't use an IM service, because a subset of its users are "wackos"? I've used Telegram for many years now, and aside from the occasional "hot singles in your area" type spam, I've never been bothered by anyone.
Well, the subset feels like the majority. Which does not mean that they are socially a majority, just on those platforms. Telegram and X has been taken over by them, IN GERMANY. I don't know how it is in other countries, maybe it's more civilized there.
The actual reason why I use an IM is the HOA, as I said. Else I do not have and do not need an IM, except for my ejabberd server + Conversations clients so that devices can send me status updates, like "backup completed successfully".
To get them onto Telegram, where group chat is not encrypted, is a win by whatever three-letter agency pulled that off. ;-)
Same here. Telegram does not suggest contacts, groups or anything else to you. If you see wackos it's because you have them in your contact list.
It's the only IM I have used that works most of the time, is not Google or Meta, is free and is easy enough to get working for normies.
I'd use IRC, XMPP or Matrix but then I cannot contact the non-tech friends I want to chat with from time to time.
Except Telegram requires you to pay to have the option of only allowing your contacts to message you.
And even then, if the spammer buys premium, they can still message you!
True, but then the spammer still needs to have your phone number or your @username.
Sadly there are too many data leaks so those phone numbers are out there.
Phone numbers aren't secret. There's too few of them. It's feasible to send messages to every phone number.
The spam messages I receive don't address me by name. They're generic.
This is a somewhat recent new "feature" to force group calls, even if they're accidental. It's not what most people I know want, and there is no way to disable it for a group, just as there is no way to disable audio messages anywhere. WhatsApp is made to the lowest common denominator, UX is secondary to market share.
> Are those days permanently gone? The days when actual UI/UX innovation was a thing?
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.
Only speaking for myself, but I have "front end exhaustion". Text based sites like this are the only ones I spend any time on anymore.
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.
Comment was deleted :(
> Can someone help me find this book? I've been looking for years. It used the Mosaic browser.
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>
> Are those days permanently gone? The days when actual UI/UX innovation was a thing?
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.
Good, then what’s stopping browsers from shipping these standard controls? It’s ridiculous that everyone needs to invent or depend on large amounts of JS to get anything decent that goes beyond a simple textarea.
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.
[1] Sounds difficult without any other detail.
But it would be funny if it's this: https://archive.org/details/teachyourselfweb00lema/page/n9/m...
This is a strong contender. Other candidates (hard to find links to the first editions):
- https://www.goodreads.com/book/show/11177063-creating-cool-w... - https://www.goodreads.com/book/show/1097095.HTML_for_Dummies...
That actually might be it! It's hard to tell, I was only 10. Still skimming it.
Why would it be funny though? Am I missing something?
https://www.ebay.com/itm/257059686708
Since you mentioned Mosaic, this came to mind. There are also a handful of OReilly books. I think it'd be funny because you gave us basically nothing to work with, so for her to pluck the exact right one from the thousands of possibilities would be impressive, lol.
Funny because it would mean the guess is a hole in one when essentially blindfolded. Glad it looks like it!
Fun fact: Opera had a tab functionality before Firefox. In fact a little-known browser called InternetWorks from the 90s is thought to be the first that had them.
I was an Opera user. They were the innovators in the browser space back in the day. Eventually it just felt too bloated, and sadly now they are essentially another Chromium fork.
If one thing will always be true on the internet, it will be forum posters saying "Opera had it first".
Firefox also had it when it was called Firebird, and I'm sure Mozilla had tabs.
According to their own blog [0], Opera implemented tabs in 2000. The earliest version of what became Firefox was released as "Phoenix" in 2002 [1]. Phoenix was renamed to Firebird in 2003 and again to Firefox in 2004 [2], in both cases due to name clashes.
[0] https://blogs.opera.com/news/2026/01/opera-one-r3-new-browse...
[1] https://en.wikipedia.org/wiki/Firefox_early_version_history
Mozilla later added them after the concept became popular thanks to Firefox. Mozilla and Firefox browsers coexisted for quite a while and Firefox was the lite version of Mozilla that didn't include E-mail client and other such features that Mozilla did.
Are those days permanently gone? The days when actual UI/UX innovation was a thing?
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.
> The days when actual UI/UX innovation was a thing?
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.
> Are those days permanently gone?
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.
Moreover, designers keep trying to justify their own jobs by changing fully functional interfaces, and then claiming post-hoc that the new UIs are better because they are better.
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.
> scrollbars that shrink to super-thin columns
Honestly, I could endlessly vehemently express my frustration to any designer that find this "cool".
/* rant /
Those designer never had to scroll to a long, long scrollable section of a page to reach the end and sadly discover that the "end" button doesn't work, because of course the browser goes to the end of the page, not the end of the scrollable section.
And of course, the scrollbar is 2 pixels wide (I took a screenshot to measure it) and it's only visible if I put my mouse in the section.
And of course, it's right next to the scrollbar that the dev decided to put the Action Icons for each item in the scrollable section.
1 Pixel left, open the popup to delete the item, 1 pixel right, scrollbar.
And of course, if I increase the zoom on my browser, everything grows, except the scrollbar.
I can have icons the size of my fist on a 27" screen but those scrollbar stay thinner than an uncooked spaghetti.
/ end of rant */
I've seen it mentioned somewhere, probably here on HN, that the original thin scrollbar was implemented in Gnome, and that it only applied until you moved your cursor to it. If true, this sounds like a great innovation. Unfortunately, nobody bothers to expand the scrollbar when you get near it. And of course, scrollbars also didn't start to disappear until long after the square-ish monitors that made them so annoying were mostly replaced by widescreen monitors that didn't have that issue.
Anecdotally, I do recall using an extremely obnoxious viewer application (PDFs or ebooks or some such) on the Windows 8 store that reversed scroll entirely. The scrollbars were still there (though I can't remember if they had handles), but scrolling the mouse wheel down moved the document up.
The best UI is no UI--the program anticipates what you need and does it when you need it. Unfortunately, this is impossible in practice for anything other than the simplest systems (eg. pause music playback when the phone rings).
Eventually we reach some kind of local maximum for UI/UX. So much of these things are a function of the relative immaturity of the platforms. They're all also pretty low hanging fruit.
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.
> Are those days permanently gone? The days when actual UI/UX innovation was a thing?
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.
New UI paradigms will revolve around how to best interact with AI.
Paradigms for existing forms of computer interaction (keyboard, mouse, touch) are pretty much solved.
> The days when actual UI/UX innovation was a thing?
It's still a thing but it went off the rails, see Apple and their latest no-contrast UI.
Good example because Liquid Glass is obviously preparing for the next paradigm shift in computing which will actually require/open up a lot of innovation on the UI front again.
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.
> Good example because Liquid Glass is obviously preparing for the next paradigm shift in computing which will actually require/open up a lot of innovation on the UI front again.
Bruh, I just want to be able to read the text on my phone.
Yeah: most experiments fail and even the ones that ultimately succeed have rough edges.
That's my point about people swooning about the days of UI experimentation. There's a reason we don't do it once we figure out good solutions to problems (experimentation is hard and mostly bad).
> > Apple ... is shoving Liquid Glass onto devices that don't really benefit from it.
> Yeah: most experiments fail and even the ones that ultimately succeed have rough edges.
Vista / Aero 2.0 already did Liquid Glass. At least they had the decency to ship a "turn this shit off" toggle that actually worked.
Vista/Aero 2.0 was purely for aesthetics. Liquid Glass is obviously to enable UIs overlaid on top of uncontrolled content (i.e. camera input from the real world, or be used through fully transparent displays).
Apple really has to bite the bullet somehow here if they want to get everyone over to what they see as the next computing paradigm.
Much like transparent glass tablets in sci Fi movies, this looks pretty cool but I think makes text hard to read and gets old immediately. Is it really a compelling new paradigm?
I think if I had a really improved version of Apple vision I would still want non transparent windows that are clean and easy to read, not floating holograms with glass like distortion?
All important questions to answer and problems to solve.
It would be interesting if someone had a way to throw a couple hundreds thousand designers and developers into an environment where they have to find solutions so we could get a head start before the relevant hardware goes fully mass-market...
The iphone is kinda fully mass market.
Right, which is why they're pushing the developer community to solve the problem on the iPhone before the next transition to a form factor that's totally dependent on this probably being solved.
And I don't want the fucking notifications displayed on my glasses!
Oh wait, I have them all off. So what will AR do for me?
I already have a physical keyboard! So what will a touchscreen do for me?
Turns out that interaction shift actually enabled a lot.
IMO any individual (like you or I) are unlikely to immediately conjure up every possible high-value idea that AR makes possible.
Not saying those ideas necessarily exist (though I suspect they do), just that your lack of imagination isn't evidence against them existing and being discoverable in the next 10-20 years.
> I already have a physical keyboard! So what will a touchscreen do for me?
Replace a keyboard only in space constrained situations. Otherwise I'll use a keyboard thank you.
Just real-time text translation and annotating faces with names would be cool.
GNOME seems to try, though people hate them for it
People don't hate them for trying new things, they hate them for making the new things the only option.
I think UI innovation requires truly novel interaction mediums. Likely, the only innovations left are predicting user behavior using AI, so essentially putting what you're looking for right in front of you before you even knew you wanted it. I haven't seen anyone do this well yet.
I remember what it was like before tabs, when there was that Multi Document Interface (something like that) instead, so you had the main parent window but then each page was its own window within it that you could resize, minimise, maximise…
Like the AOL browser, come to think of it.
Tabs in Firefox were such an unfamiliar thing.
MDI was rightfully seen as a complete failure, but there was also SDI, where each open thing is a separate window. I don't know how we got from MDI in office apps being completely terrible, to MDI in browsers being the accepted norm.
MDI in browsers isn't enforced though, you can still use each tab in it's own window, and (depending on your browser) you can open other sites in a sidebar (? Used to be possible) or split-tab (I think I've seen this in some desktop browser? I know chrome does it on mobile). Office documents though? SDI is now enforced on office. You can't hide that taskbar entry for every single document you have open, even though there's only 2 important ones each open in their own window and the rest could all share a window for all you care.
MDI wasn't a failure, the requirement was a failure. SDI requirement for office is also a failure. More so than the escapable (multiple window support) MDI of browsers on desktop.
Actual MDI was so much worse than browser tabs, unrelated tabs can be merged into the same window or split apart into their own, instead of floating on top of an awkward background.
The question is why aren't they a feature of the window manager instead of the application. We should be able to have windows with tabs from different applications.
Tabbed MDI is effectively just a better interface to SDI (for most situations)
Actual MDI applications feel so dated. It made more sense when there wasn't a unified task bar kinda thing (which when you think of it, is kinda like tabs as well)
Well websites and documents are not the same thing so it makes sense that a paradigm that works for one doesn't necessarily work for the other. I do find web-based document editors very annoying to use when they are in the same window as other tabs - at least web browser MDIs allow you do effortlessly separate tabs into a new window these days.
I went through the same (or at least very similar) experience. I loved that.
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.
New things do come around. Chrome's new split view tabs is pretty slick new "UI control".
Chrome's Whats New seems like half AI stuff and half UI features for people who have tons of tabs.
Can we stop innovating on UI for existing problems?
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".
Vivaldi has pretty good default UX features. They are following Opera advances and keep pushing for UX innovation.
> Are those days permanently gone? The days when actual UI/UX innovation was a thing?
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.
We tried with gemini but nobody jumped on
Adding an AI button is not an innovative feature
Not the button alone, but integrating AI is.
VR gaming had this experience. AR is about to.
If pull-down refresh were invented today it would definitely be called an anti-pattern and the evidence of the regression of Apple.
Agree. I know some people want that. Me though, I'm 85% done filling in some format, I scroll up to reference something that's off the screen, the effing 'pull-down to refresh' triggers and I have to re-enter everything I just frustratingly spent time entering with mobile input being so shit.
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.
I can't really see how? It doesn't have any affordance and discoverability is rather low; but there's feedback and a modicum of discoverability. It's useful [to me].
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?
My grandfather is continually unintentionally starting group voice chats because of that ridiculous design choice.
He has one going right now.
???
BS claims about a universe that doesn't exist, to sound nihilistically cool.
Except in the early days of smartphone, people pushed back against pull-to-refresh[0][1][2]. Android devs were confused why it was a thing. It's a design with zero discoverability - how do you know what would happen when you pull down? Perhaps the app would show a search bar. Or pinned posts if it's a forum board. Or ask you to review the app? How do you know pulling down is a gesture at all?
The only reason pull-to-refresh got accepted is that it came so early that the UX of smartphone app wasn't well established. Before pull-to-search or pull-to-whatever had a chance.
> nihilistically
It's quite nihilistic to think history doesn't exist and things were born as they are currently.
[0] https://web.archive.org/web/20201204045158/https://www.fastc...
[1] https://web.archive.org/web/20120331181045/http://android.cy...
[2] https://www.reddit.com/r/androiddev/comments/vbt6d/pull_to_r...
If it has zero discoverability, nobody would be using it today.
Discoverability is more than simply visual cues. Seeing other people do it counts.
> Zero discovery
It's on a desire trail. Users discover it by scrolling up. Which, presumably, most users do?
For Firefox, I think that disabling the telemetry and the studies is not going to help Mozilla improve the browser.
I'm not convinced that's their goal any more. The number of users turning off or ignoring AI features is probably considered a problem, not a signal.
Every single thing for the past 10 years has had (opt-out, which most people didn't) telemetry and that correlates with a decline in quality, not improvement.
My suspicion is that this is due to three things:
- 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.
> Use of analytics tends to replace user trials/interviews entirely, trading away rich signals for weaker ones
Yeah, this is huge. The 30-day A/B test is a scourge on the industry.
Telemetry in the hands of software craftsmen with a supporting business model will probably support improved software.
Telemetry in the hands of stakeholders whose stakes are business/career KPMs will probably serve those, and the software experience will follow.
And not giving IKEA access to cameras in your home won't let them improve the furniture.
But it might be accepted if the furniture cost $0.
One could hope if it happens enough they'd be jostled out of the McNamarra fallacy tarpit they've ended up in, though maybe that is too optimistic.
They don't care about improving the browser anymore. They just want to make it into an AI browser.
For Firefox they had a decade+ to stop making useless wasteful choices that killed any good will towards the company.
I still use Firefox but I, frankly, feel it's stagnated. On mobile I'm in the process of changing habits to something else (auto reflex sometimes still opens Ffox, but lately I'm circling back to opera, which I stopped using on desktop what... 20 years ago?)
All this to say, I don't think Mozilla is doing much with all the telemetry data it's gathered all these years
Edge has so many places that it can send telemetry to Microsoft. Any of the features that intend to "help" your browsing, including safe URL checks, shopping enhancements, etc.
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 think a "Just A Browser" approach would be better for people like me, who don't really want to patch configuration files for their existing browsers - sounds messy.
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.
Vivaldi, Librewolf, Palemoon and Zen Browser (the latter I'm not 100% sure about) have all committed to add no AI features, for some definition of AI.
If you go through the manual steps on mac os, the file gets deleted when Firefox is updated.
Is there a way to persist the file even after updates?
That's a nice idea, but I'm hesitant to use automatic installation scripts like this, even if I review them carefully. I prefer manual steps – downloading the file, placing it in the correct directory. Also, I use custom profile directories for my web browsers, so this script wouldn't apply to my setup.
do you carefully review those files you've downloaded and placed into said correct directory, or do you just double click an YOLO your way through the install? at least a bash script allows for you to read it vs a pre-compiled binary installer. either way, you are trusting code created by someone else.
Honestly, I'm reacting a bit against the "YOLO" install culture common on Windows and macOS – just double-clicking and hoping for the best. I specifically use CachyOS, an Arch-based Linux distribution, because it emphasizes a more deliberate, package-managed approach. I prefer relying on the maintainers' work and pre-defined steps within packages whenever possible. It allows me to understand what is being installed and how it's being configured, rather than an all-in-one script potentially making changes I didn't anticipate.
Manual steps are linked to in the second sentence of the Getting Started section.
Not for Linux. I am using Linux on my desktop.
It does if you look into the README for Firefox
https://github.com/corbindavenport/just-the-browser/tree/mai...
Ah, it looks like they have Linux instructions for Firefox but not for Chrome or Edge. How odd.
Who uses Edge on Linux?
I do. I use it, alongside other browsers, because I prefer different browsers for different types of tasks.
> Open a PowerShell prompt as Administrator...
Sounds like the beginning of a nice ClickFix campaign: https://it.lbl.gov/the-clickfix-attack-a-new-threat-to-your-...
> No. Just the Browser uses group policies that are fully supported by web browsers, usually intended for IT departments in companies or other large organizations
This is cool! I was expecting a script, which tend to be brittle. This is a great way to do it.
I like how the bash script to install this starts with getting sudo access :)
I’ve been using Waterfox recently, which feels almost nostalgic in how much it’s just a plain goddamn browser. It’s really delightful.
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.
Firefox:
"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."
I hope someone tells the Mozilla CEO.
https://blog.mozilla.org/en/mozilla/leadership/mozillas-next...
“AI must be clear and understandable. Controls must be simple. AI should always be a choice — something people can easily turn off.”
And then in the same article he goes to write that Firefox "will evolve into a modern AI browser", which makes AI sound like an intrinsic trait. Doesn't exactly inspire confidence if you ask me.
You could turn Pocket off, you just had to do it every time you updated because they decided to be user-hostile and keep jamming it down users throats (I'm still baffled as to why).
I'm not hopeful.
That wasn't my experience with Pocket. It stayed off for me on Firefox for Linux.
Very neat, just installed it. I would say having someone download the profile and install it on macOS is honestly just as easy as running bash. It's also safer and provides more understanding.
I noticed that Safari is not mentioned - is it because is not relevant on Desktop or because it didn't go through the same enshittification process as the other two major browsers?
That's an interesting observation. Apple doesn't really have an AI offering, so they have nothing to stuff into Safari. In general Apples failure to produce an AI offering might turn out to be an advantage long term.
Probably both? I did find its omission spoke loudly. I use it every day on desktop. The only enshitification I have to worry about is Alan Dye’s hit and run crimes against usability.
> Mac and Linux: 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:
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")))
(in case anyone is looking to exploit this: user `corbindarvenport` is available if you want to do some typosquatting)
It'll be good to just use the browser again, so I will def be trying this out. But I can't help but feel that for simple dumb questions it's a lot easier to just ask AI bots instead of searching on a web browser. Does this just depend on the context? Example most recently I wanted to know how many miles would a pair of running shoes last. AI can answer this instantly (hooray instant gratification) and googling something like this would take longer. And of course this is why they shove this stuff on the browser.
I guess then, the browser and AI just serve different purposes now?
there are already a bunch of electron and chromium projects that give you just a simple and highly performant browser sandbox.
I dont understand the need for an entire website and shell script installer when all it does is download 1 file and put it somewhere.
For anyone else on firefox, save yourself some effort and just download this https://github.com/corbindavenport/just-the-browser/blob/mai...
The website explains each setting that would be turned off, which I find rather informative.
Just give me a Chromium based browser make it open source and a verifiable build and I’ll pay you twenty dollars.
Free and open-source browsers based on Chromium: Brave, Dooble, Falkon, Otter, qutebrowser, Supermium, ungoogled-chromium
source: https://en.wikipedia.org/wiki/Chromium_(web_browser)#Free_an...
I don't want all the extras of Brave, Double seems to really care about it's 2017 release https://textbrowser.github.io/dooble/ and seems either unmaintained or fake, and I want to ensure the developer can eat.
I wish Firefox had a per-user location for policies, so I can just carry it around with dotfiles.
Aren't preferences stored in per-user JavaScript (text) files?
Not quite: they're per-profile, but the path for the profile is not deterministic by default. So you need to first create profiles, exit, and then manually paste the preferences.
I haven't found any way to programatically create profiles.
What difference does this have compared to something like ungoogled chromium?
Nice touch - seeing the Windows 95 IE favicon took me back for a while.
can you remove webrtc, localstorage, web workers, and customize fonts?
Half of the websites will stop working if you did that.
And you might as well just fork chromium for that purpose.
Half of the webapps maybe. Actual websites don't have a reason to use any of these features and most don't (except for fonts maybe, but removing those doesn't prevent the website from working).
> Half of the websites
Yes browsers should be used for browsing, those half websites can run on something else.
Security experts when AI in browser: `curl | sudo sh`
launching firefox automatically deletes the distribution folder from Content/Resources; and nothing changed in the browser
How intrusive is AI in a browser that you feel you need another browser that advertises no-ai? Is it a privacy thing? Like for me in edge, it is completely out of the way.
It's not a privacy thing because it also disables on-device AI features too.
would be great to block all cookie popups. Pry would need to be a chrome extension
What is sad is that we need anti-AI measures.
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?
I've been using Kagi for several years and find it much better than Google. I've not been tempted to go back, like I always was with DuckDuckGo and others.
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-...
> What the heck is Google doing?
Inflating stock prices.
Really cool!!
[dead]
[flagged]
> I don't believe someone can understand the problem, and make _this_ in good faith.
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.
Open Wireshark and see the traffic. Read documentation of two mentioned projects.
The whole "average user" agenda is already a smell. Nice to see you writing your first non-question here.
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
"Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."y
What a rude thing to say. No one is forcing you to use it.
I mean this is an Anti-AI move. I am not saying you should join the pro ai but hating on AI just because its AI is not a good look
It's silly to treat this like a totalizing partisan issue where everything must be clearly "pro-ai" or "anti-ai".
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".
Most of the "AI" features added in Firefox makes no sense. They provide very little value to most people, but they are unreasonably hard to disable. Other than jumping the AI bandwagon, I have yet to understand why Mozilla keeps pushing AI features.
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.
> but hating on AI just because its AI is not a good look
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
I gaurantee you there will be a very popular niche that focuses entirely on being anti-AI, and it will always be around.
This niche will get smaller over time. The key hurdle right now is that most "AI" is just LLMs. People currently prefer to go to a website or open a dedicated application for AI inference. As better integrations with other workflows are made and people see them, the resistance will weaken.
Microsoft shoving LLMs into literally everything, including Notepad, is what people are currently hating, because it isn't quite ready.
Crafted by Rajat
Source Code