Tab groups are on by default in nightly, and though I haven’t been on the computer much this past week they seem well realised. This article suggests they can be enabled in other versions, but I only really use nightly so I haven’t tried it..
For those wondering, this uses cloud-hosted AI models. But it's completely free to use so Mozilla is just paying the cloud bills out of pocket. Maybe not the best idea given their increasingly precarious financial situation?
I never understood why Mozilla doesn't offer paid privacy respecting services (outside basic sync that already exists) like email, cross platform password storage etc.
Because everyone who cares about it is already using some sort of a service like that, and it’s incredibly niche market so it’s not worth the effort to develop and support such projects.
Mozilla should have been doing what Proton is doing. But considering how far Proton has gone I wouldn't be surprised if Proton spun up their own browser at this point.
Only within the confines of web usage, unless you know something about Firefox Sync injecting passwords into Android apps or environment variables
Also, like it or not, I think Passkeys/WebAuthn is going to be pushed more and more, so without a user focused way to own them, that's another reason not to try and use a browser (any of them!) as a password store
Not using services funded by Google is pretty high on my personal privacy checklist. Also, they offer some services (like VPN, Relay, Pocket) that usually end up as annoying bloatware in Firefox.
> The minute Mozilla starts being relevant in search, Google will cut their billions USD/year sponsorship, killing Mozilla in the process.
The incentives are not there.
Google needs mozilla more than anyone. It would be a big win for the web if google stopped financing mozilla and had to deal with the consequences.
Why does Google need Mozilla at this point? Firefox has 2.6% market share. Sure, I imagine the cost/benefit ratio is slightly positive for Google, but Mozilla needs Google since that's the majority of their revenue; the reverse is definitely not true.
As likeabatterycar said, it's about avoiding anti trust / monopoly issues. Just imagine mozilla died and google had to cut lose chrome. That would really be interesting.
> Orbit is a Firefox add on that uses AI to summarize and answer queries about web content such as articles and videos.
> When a user asks Orbit to summarize or query content, Orbit gathers the context (eg. text, images, videos, etc.) of the page the user is viewing and provides a summary or answer. Orbit works on websites including Gmail, Wikipedia, NY Times, YouTube, and more.
> For the current version, we are using a Mistral LLM (Mistral 7B) hosted within Mozilla’s GCP instance.
It's interesting they're going with Mistral 7B. Is anybody else using Mistral 7B in production? And in what role?
I've considered using it for general knowledge type questions, and as a way to classify information, but would have never considered it for summarization type tasks due to it's limited context size (8k).
Its summarisation, who cares if its right as long as you feel confident after reading it? /s.
In my experience, even GPT-4o is terrible at revealing information from things longer than a few pages.
It might be an issue with dimensionality reduction in general though. If you think about it, you can't really take away much of what is contained within any given amount of text with text, unless the source was produced extremely inefficiently.
Producing outlines or maybe a form of abstract, it seems to be okay at, but you would never really know where it fails unless you read the entirety of the source text first to begin with. IMO, its not worth risking unless you plan to read the source anyway or its not really important.
A key thing to remember in this specific moment in history is that the vast majority of people will be as lazy as they can get away with being. People want to leave work and go home and if an LLM lets them do that faster, who cares if it's accurate? It can absorb the blame.
> Why are you confident to use such a tiny model on something so critical?
Don't use any LLM for something critical. They can't be trusted innately due to their design, why on earth would you use them for something where you need reliability?
I don't use Mistral 7B alone, this is just a component in a RAG-based system. A system that is 1) not clinical facing, 2) not used in clinical decision making, 3) provides in-line references sources for end users to validate information, and 4) is inherently human-in-the-loop.
Buried as the last sentence in a collapsed box at the bottom of the page:
> For the current version, we are using a Mistral LLM (Mistral 7B) hosted within Mozilla’s GCP instance.
And why is it "...Mozilla's GCP instance", not "We quietly send all your data to Google servers, and everyone pinkie-swears that's totally privacy-respecting"?
I don't know what bit of news you are talking about. Unless you are just talking about the news in general which still doesn't prove anything. Any news that Google is stealing data out of GCP is not something you could just sweep under the rug in 24 hours.
Even on the linked page, Mozilla is arguably being evasive about the fact that they're sending the data out the Internet at all.
We don't know whether this is another time that Mozilla execs have sold out users, or shipped something half-baked and vulnerable.
I'm not saying they're leaking the data (by agreement, or negligence), but Mozilla has mediocre credibility in recent years, and there's nothing on this page that improves that reputation.
Regarding Google, for a long time, their thinking seemed to be "We're Google, so of course anything we do is privacy-respecting", not as guidance, but to justify whatever they wanted to do. Also, every time Google gets caught with their hand in the private information cookie jar, it just mints a new industry standard practice.
> I'm not saying they're leaking the data (by agreement, or negligence), but Mozilla has mediocre credibility in recent years, and there's nothing on this page that improves that reputation.
I think you read too much HN and aren’t aware about all the stuff going on in the background at Mozilla.
If there’s one company I would trust, it would be them. Their marketing has been mediocre and I’m not 100% sure about if I like their future decisions, but I trust them 100%.
If you aren't running an encrypted disk on any cloud provider you should absolutely fundamentally understand that your data has been scanned and that your VM data is "business data" so a copy gets sent to whomever wants it, in bulk.
Maybe it's good to assume that but at this point that is not going to purposely happen at a company like Google or Amazon. The risk, which is a near certainty to bear out if they have any decent employees among the tens of thousands (esp. with the weekly "I'm leaving because I hate this company" screeds these companies yield), isn't worth whatever little reward they might find in your data.
In this case, usually the infrastructure provider owns the keys, and if not, they would have easy access to them. So I don't see how encrypted disk really solves anything besides accidental leakage to a peer infra user, or someone sneaking into the datacenter and physically removing the disks.
Is there a profit to Google digging around in the information people send them?
Are there long and vague terms of service documents backed by a pile of lawyers?
There you go, incentive and means. I'm not even confident companies would see that as a problem when it was raised with them directly, in much the same way that Microsoft hosting all the corporate email seems to be just fine.
The disincentive is far higher than the incentive, and the TOS have been scrutinized deeply by some of the biggest enterprises on the planet. GCP is not a consumer service like Gmail or Maps.
As the comment above suggested, any information to the contrary would be business-destroying for GCloud. Many of their enterprise users have strict requirements about access to and use of their data.
Re the example of Microsoft corporate email, much the same situation applies. If Microsoft were mining that corporate customer data and using it or reselling it, enterprises would dump them in a heartbeat.
Can confirm. I worked in gcloud for years. There are so many policies in place to keep customer data secret, even when you're on-call and trying to solve customer issues, it's actually annoying.
It makes sense. Some gcloud customers are banks. Some are federal govt agencies. Some are foreign governments. Google would not only destroy it's cloud business, but also probably get fined and sued out of existence if it was poking around in cloud or gsuite data.
You get what you pay for (in terms of privacy) at Google. Regular users never pay Google a dime, so they don't get much privacy. Cloud and gsuite users fork over mountains a cash directly, and their data is kept about as safe as can be as a result.
- Currently using Mistral 7B, but ability (by Mozilla) to swap the model used to another open source at any point.
- Hosted "by Mozilla" on their GCP instance.
- No obvious info about what it will cost them to run this since it is free to use.
- No training on user data.
Like others here, I'm very curious about the cost for Mozilla to run this service. It may be less than it initially appears given the 7B model they chose. I do wish they would focus their efforts on creating a very long-term endowment to pay devs for continued Firefox development in lieu of projects like this given the tenuous situation with their Google funding.
I'm not against this kind of thing in theory, but I hope it's being done in a cost-sustainable way.
If you scroll to the bottom of the page it appears like this is made by Fakespot, https://www.fakespot.com/our-mission. They explain that they make money through ads, for their other similar product, wouldn't be far to guess that's the future strategy here as well.
Pretty much, about 85% of Mozilla's revenue comes from Google in exchange for making them the default search engine. The ongoing antitrust case against Googles search business is threatening to make that deal illegal though, so ironically the attempt to break up one Google monopoly might incidentally kill Firefox and make another Google monopoly stronger than ever.
In a separate comment, I mentioned how Mozilla should have been more like Proton with their cloud storage, VPN, password manager, and cloud office suite.
In fact, they should have done that a decade a ago.
Mozilla has been around since the late '90s and should have evolved beyond just being a browser company. They launched a VPN service when VPNs were already everywhere, and they did the same with a bookmark manager when others were already offering similar solutions. Mozilla is always catching up, never leading, and that's a common issue with many big open-source and free software companies. They often pretend to be a business that isn't heavily propped up by big tech donations.
If I were leading a browser company, my focus would have been aggressively directed towards small business software. I’d create an internet and privacy-focused affordable minimal business software suite that lives within the browser — a combination of Proton and Zoho. And I’d strongly avoid building things that should be browser extensions.
Firefox needs to be forked and owned by the people. I'd pay $1 a month for a browser run by the people. Management and extension options could be rolled out for government, business, education, etc. There are so many models where the browser thrives, the org that shepherds the browser thrives, and the people thrive.
Firefox is like the shitty best option that camps out in its niche, it sucks but it is really hard to push out of the way.
And who are 'the people'? It's an open source product, developed by a non-profit organisation. You want the government to release a browser? I can tell you how popular that will be...
>The Mozilla Corporation is a wholly owned subsidiary of the Mozilla Foundation that coordinates and integrates the development of Internet-related applications such as the Firefox web browser,
>Unlike the non-profit Mozilla Foundation, and the Mozilla open source project, ... the Mozilla Corporation is a taxable entity.
> I'd pay $1 a month for a browser run by the people.
You and approximately 5 other people. Paid browsers in 2024 are not a path to mainstream success, especially if you are what Firefox is - a fully independent tech stack down to the browser engine and not a barebones reskin of Chrome. A reskin of Chrome would have very low development costs. Firefox does not.
That is simply not true. I donate a little under 1k a month, there are many like me. Mozilla is not an org that I would feel comfortable funding given the perverse incentives that they have aligned themselves with.
> I'd pay $1 a month for a browser run by the people.
The bad news is that 12,000 other people would need to similarly pay per month to have a dev team of just 10. I know Ladybird is lean and mean but finding that big of an audience (or, of course, bigger) who would pay for a browser, per month, is likely a non-starter
It would be a much more interesting proposal to start a bug bounty for the damn near infinite Bugzilla queue, although as I understand it some of the hassle of a bug bounty program is evaluating submissions. And don't say "but we'll use an LLM" or I'll throw up in my mouth
I generally frown on calculating market size like this, but 10% of the top 1B is 100m, if 1% of those paid a dollar a month, that is a million a month. Good for 100 fang eng.
I hear you, and ironically I'm actually already in the target demographic of willing to pay for software I use and enjoy. Then again, I find that I am often an outlier in the "lengths folks will go to for good experiences" camp
Anyway, at this moment in my life I don't have the emotional energy to launch this project, but if someone does then please tag me and let me know
This project seems especially ripe for success because it doesn't need product market fit, it doesn't need requirements gathering, it just needs execution and a non-retarded bug tracking product
I installed it and it's a giant floating popup that's permanently on your screen. You can enable the "minimal" theme in the settings to turn it into a smaller-but-still-big pill shape. It doesn't look like there's a way to hide it in a context menu, at least not one I can see.
I don't fancy having a random floaty object in the way of my webpages, no thank you.
Edit: It appears to go away occasionally. This UX is unclear to me.
Yeah, I can't imagine what lunatic thought it would be a good idea to have this stupid orb floating inside the page. Heck, it would be fine if you at least had an option to move it into the toolbar where it belongs...
If you pin the extension in the toolbar, there's an option to "disable" the extension. The floating popup goes away and using the extension to summarize still works.
Why does Mozilla always spin their wheels releasing random side projects? Cloud-hosting an outdated tiny LLM that you can't swap out or run locally, to do basic summarization? This just doesn't feel like an area of strategic focus that makes sense for Mozilla. GPUs are expensive, talent to do inference well is expensive, and the actual product they're shipping seems pretty marginally useful at best.
If they shipped vertical tabs I'd probably switch back to Firefox. But that would require focusing on actual browser UX instead of random offshoots. What's their actual product differentiation from Chrome these days?
*uMatrix is unmaintained, and uBlock Origin can do fine-grained control – it just requires the “advanced user” setting for some reason, even if you expand the panels all the way. https://github.com/gorhill/ublock/wiki/quick-guide:-popup-us... (see “I am an advanced user!” expanding section at the bottom)
And I can actually build and package it from source in about 2 hours, unlike the other monstrosity, which surely must lower the barrier to contribution. Almost as much as Bugzilla raises it, I guess :-(
> Why does Mozilla always spin their wheels releasing random side projects?
Because some project manager had "AI" on his performance goals this year.
Firefox is buggy as hell - which is incomprehensible given its age, but older brother Netscape had the same problem 20-ish years ago. The Netscape 4.x days were absolute hell and you could go hardly a day without the browser crashing.
Despite this, it's packed with nonsense no one asked for like Pocket. Which is a coincidence because "AI assistant for Firefox" is the dictionary definition for redundant things no one asked for, with better alternatives preexisting.
At this point Firefox needs to die and something new and manageable - with energetic developers - needs to take its place. Maybe Microsoft could open source the original Edge engine? The one before they bent over for the long dick of Google Chrome.
Don't you worry, I'm sure their new AI Assistant will generate a ton of bugfix code!!11 AI gonna take all teh jobs, or so I hear
But, as a more serious contribution: the sentiment "At this point Firefox needs to die and something new and manageable" is the same one which generates the infinite JavaScript treadmill akin to: "I don't like all those bloated JS frameworks, I want one slim and fast and manegable ... well, except this one other feature ... and this other ... oh, shit, this needs to die and have something new and manageable take its place ..."
It's not that I think Gecko is the bees knees, but I do think it has stepped on more than its fair share of real life landmines, and the Great Rewrite Theory means someone needs to spend all that time re-discovering them
As a counterpoint I'm a heavy firefox user and haven't had crash in many months, and even that was because I was testing some experimental webgpu thing which I had to manually enable. It has its fair share of odd bugs but what kind of big piece of software nowadays hasn't? It's honestly no less cromulent than Chrome.
There is, but hiding the tabs on top — IMO the entire point of vertical tabs, since almost all screens have more horizontal than vertical space — isn't supported without maintaining your own custom CSS. It's a pain.
They seem to feel that they will be marginalized if Google stops paying them to set their default search engine to Google but, the way they have handled it is to focus on everything else other than their browser. At this point since browser engines are dominated by Blink and Webkit what exactly does Mozilla have? Their market share just keeps on going down.
Mozilla laid off everyone they had working on it. The project is still going on a volunteer basis, but it's obviously not progressing anywhere near as quickly as it was when it had people working on it full time.
Some parts of it, the parts that were production-ready, were merged into Firefox years ago and live on there. Other parts that weren't production-ready were canned and the staff laid off.
> from Google to survive. If Google stops the payments, Mozilla probably goes out of business.
> I think it’s pretty clear why they keep on doing this type of side project.
It’s not clear to me. I agree they would have some problems if Google declined to pay them because the next best offer would be lower.
But the best way to keep these payments, or to increase them, is by making a better browser and giving people a reason to use Firefox. After spinning off Servo I lost the last hope I had.
It seems everyone is stuffing AI summary tools into everything, is this something that will retain users or bring in new users? I doubt it.
Browsers are a means to an end for other major providers excluding for Opera. They can be loss leaders even, as long as they do their job of steering users towards proprietary ecosystems
Most non-tech "normies", which is to say 95% of the population, barely know what a browser is. They couldn't describe the function of one, or discriminate what makes one better than another. They certainly won't give a crap about Servo, or extremely marginal improvements in page load time (at best).
Given that web browsers are the heaviest application most users are running, and they are running them on low end 10yr old laptops with 8gb of RAM, I think an ultra modern lightweight web browser would be noticeably better.
Web browser crawl on these low end laptops.
This is how Firefox became popular in the first place, by being better.
The “we need an alternative to the “WebKit/blink/chromium” monopoly is what the majority of people will never care about.
Web slowness has much more to do with sites than with browsers. Where sites have accumulated Everest-sized balls of JavaScript mud, web engines have only become more optimized. If you visit “old web” style pages (Macintosh Garden for example) on old machines with modern browsers there’s no speed problem at all.
In the face of all that JS, there’s only so far a spiffy new browser can improve the situation, aside from maybe drop large chunks of legacy web standards but then you’re breaking large chunks of the web.
Most of the new features on the web are corporation driven. The only web site I reluctantly give microphone access is Google Meet and this should have been a desktop app. We have so many layout engines in a browser while layout for both documents and GUI have been solved for decades. Every new feature is just reinventing the wheel to solve a non problem.
From my experience observing and interacting with “normal” non-tech users, slowness of apps and long loading times are simply ignored. I would think “how the hell can you live like that?!”, and they would at best say “yeah it’s kinda slow”.
They also wouldn't use Firefox - they'd use Chrome, Edge, or (more likely) Chrome/Safari mobile. People who use Firefox are already tech people or the family of.
The least cynical view, in my opinion, is that Google is an ad company, which ultimately means they are a data company. They don't need Mozilla to build a better browser, they need Mozilla to increase the amount of user data Google can collect and ad spots Google can sell.
The more cynical view is that Google doesn't care at all about Mozilla because the investment is nothing more than a hedge against regulatory pressure.
You bring up a great example. Mozilla poured tons of resources and had very smart people working on Rust, Servo, and related tech projects to improve Firefox. Where was the surge of market share?
We’re at a point where the core functionality of browsers is very mature. It’s unlikely that any amount of investment will produce a browser that is significantly faster at things like JS execution or rendering compared to Chrome.
So alternative browsers add things like better ad blockers, more privacy protections, or maybe LLM summaries to enhance the core browser experience instead.
For some of the stated reasons this seems like a terrible way to diversify and add any revenue. What if they take a page from Silicon Valley? If the US Government makes Hooli, I mean Google, divest Chrome could Mozilla acquire it?
I'm sympathetic to the viewpoint but the idea that Mozilla would attract more users via vertical tabs than by an AI assistant strikes me as flat out wrong.
Mozilla's brand is "pro privacy", it does make sense for them to launch an AI product with that brand position. I doubt it'll be successful because I don't think enough people actually care about privacy, but still.
I feel like it's a common HN sentiment to say "why don't Mozilla just focus on the browser?!"... the answer is because barely anyone is using it and there's very little they can do to move the needle on that. IMO they're an organization looking for a purpose.
Depends. If they start with this and then use it as iterative development before going local-only as the models and hardware improve, that could be a good move.
For me, the product differentiation of Firefox is a bunch of small convenience features which Chrome in its monolithism refuses to provide, such as:
Allowing Backspace to go back a page.
The built-in screenshot tool.
Being able to turn off smooth scrolling.
Support for a menubar, so that I can navigate the features I want quickly.
Being able to choose page encodings (I use a non-Latin charset language).
A usable/useful bookmark manager with things like sorting, tags/labels, timestamps, etc.
"Restore Previous Session" feature.
These are just a few features off the top of my head, I know there are many more.
Of course, for certain sites like Google Docs and YouTube, I have to use Chrome for obvious reasons, but for most browsing I use Firefox (and qutebrowser.)
Out of curiosity, what "obvious reasons" issues have you had with Docs and YouTube. I use Firefox for everything, including those, without problems. (Though not in any kind of advanced way.)
I use Firefox on my HTPC, running a 6700K with 24GB of RAM. Not new, but not “slow”. Clicking a YouTube video will cause a three second page load, even if the actual “page” says it’s finished loading. Videos will start to play audio before the page is rendered. Navigating back and forth causes issues like this too. Sometimes I can get it to show a video but not change the title or comments that it renders. If I accidentally click the “Shorts” hyperlink I basically have to close the tab to stop it from endlessly playing shorts in the background of the SPA.
It’s awful. The best example I experience on a day to day basis that the SPA as a concept is utterly flawed. YouTube is a fucking webpage that fails to work like a webpage and an app that behaves like a students rough alpha. An utterly painful experience, continually made worse by likely skilled devs who are managed by complete bozo losers. But at least the progress bar has an ugly pink hue now.
Docs just almost entirely does not function for me on Firefox on Linux. As in, I've had it crash the entire browser while just trying to type in a document. In general, Google just aggressively seems to be hostile to any non-Chrome browser using any of their sites. I'm sure they cloak it in "well Firefox just hasn't implemented this spec yet" but when it's functionally enforcing their browser monopoly I have to assume malice at this point.
Too many Google sites behave worse on any non-Chrome browser for it not to be intentional.
I wouldn't call them "obvious" reasons but I recently discovered that in Google Sheets under FF I couldn't duplicate a tab or copy a tab to a new sheet. I had to fire up Chrome to do that.
Oddly enough, Chrome had somehow lost my settings since the previous time I started it a few months earlier, as if it were a new install. :-?
I've not noticed any problems under YT Premium so far.
You may not like it, but I don't think it's a good reason to remove the option of enabling it for those who rely on it for daily work, which is what Chrome did.
Can disable as easily as Firefox -- chrome://flags/#smooth-scrolling
Actually, Firefox lately introduced a bug that will reset smooth scrolling everytime I remote desktop to my Windows machine. Which is annoying AF and they haven't fixed it for months.
This only works on pages where extensions are enabled, and only after the extension is successfully activated, so about 30% failure rate for me.
>Can disable as easily as Firefox -- chrome://flags/#smooth-scrolling
This has not worked for me reliably, and the flag has been renamed several times.
For example, on my Mac, it reads "Not available on your platform."
>Actually, Firefox lately introduced a bug that will reset smooth scrolling everytime I remote desktop to my Windows machine. Which is annoying AF and they haven't fixed it for months.
I use all the extensions I mentioned for years, claiming it has 30% failure rate is bullshit. They don't work on internal special pages, sure, but they work flawlessly on any "normal" webpages with close to zero load time.
And no, disabling smooth scrolling works totally fine as I use it for more than 10 years.
If you're going to exaggerate your points to make a statement instead actually trying to find solutions, I have nothing for you, then.
What is more likely that they are lying or it doesn't work as well in their configuration as yours? The good faith answer is that it is probably the latter.
Well, you're in luck, sort of. Mozilla has vertical tabs in a new sidebar experience. It's the worst implementation of vertical tabs and a sidebar I've seen in a browser. Complete with typical Firefox UX, it's completely inconsistent and unintuitive to disable. A complete farce compared to Sidebery.
- continued support for manifest V2 (primarily because ublock origin would stop working if forced to V3 only)
- the firefox address bar is way smarter for any given string i type in than chrome's. it's ability to surface the most relevant deeplinks from my history, vs top level site, vs perform a web search, is night and day difference from the randomness that other browser search bars offer.
- I have the opportunity to use Zen (a Firefox fork) [0] and it's 100% interoperable with my vanilla Firefox instances across devices -- i can even send tabs from my Firefox Nightly on Android to my Zen instances on Windows or Mac. BTW Zen has vertical tab support, (more) first-party multi-profile support, and preserves the address bar behaviors of vanilla Firefox.
Building a browser is hard. Building a proof of concept of the current tech fad is easy and fun. Sometimes developers need to do easy and fun things to keep themselves motivated and happy.
You could build an AI Assistant, or you could spend a month bikeshedding some design details of vertical tabs.
>Sometimes developers need to do easy and fun things to keep themselves motivated and happy.
I tried this line on my boss. Didn't fly.
Back to bikeshedding with the rest of the team.
I wonder if the difference is we have to earn our money, and our customers expect an roi.
Or, at least something that doesn't mess up what already exists.
Installing the extension enables a floating widget in the webpage. Horrible implementation. Couldn't it have been integrated into the browser better, like the Reader button?
I guess I'm really not this extension's target audience, because somehow Mozilla managed to make me uninstall it in just a few minutes.
First, they have forced telemetry. Okay, it's an early release, it's very basic information, they want to understand how they're doing - I don't like it, but I can understand it. Sets a wrong vibe, though - I had to check if it was from Microsoft and not Mozilla. ;-)
Then, I figured there is no option to use locally-hosted LLMs, which can be something as minimal as simply allowing to configure custom API URL. Somehow, less and less things about Firefox are tinkerer-friendly than they used to be.
That made me wonder if Mozilla used OpenAI-like API, or if they invented their own unique thing for some reason. So I went to look and according to the extension page, it's proprietary ("All Rights Reserved") and I'm too lazy to bother deminifying code from the xpi or remembering how to debug extensions.
Finally, '00s have called and said they wanted their weird floating round thingy UI back, and so I had to return it to the store. (I'm kidding, I don't really mind the visual style. The real issue with that thingy was how it floats on the page. Like, why on Earth it isn't normal right-click menu option that doesn't obstruct the view until it's needed? Or a menu on that toolbar button? It's not even a paperclip to be worth it.)
And then I realized I somehow missed the big "AI you can trust" header, which should've already been a huge red flag.
> For the current version, we are using a Mistral LLM (Mistral 7B) hosted within Mozilla’s GCP instance.
Based on this I would assume they are using GCP Vertex AI as that's going to be WAY cheaper than rolling it all themselves and hosting the model on a GCP server instance. I would also assume they'd be using the gcloud SDK for Vertex AI/Model Garden, which I believe means they can't just provide for a different endpoint and payload shape if you had a service elsewhere.
Eitherway, at the (presumed) scale they'll probably also be using GCPs API management service, so I would expect further abstraction between what the extension is sending and what the model/Vertex AI expects as a payload. This means providing that kind of "bring your own endpoint" experience would require more bespoke build-out time.
BUT who knows? Maybe this is just straight up hitting the out-of-the-box GCP Vertex AI REST API directly from the extension like some hobby project.
It's Google Cloud Platforms "AI" service[0], so actually more analogous to what is now called Azure AI Foundry[1], and what used to be called Azure OpenAI Studio.
Microsoft Copilot Studio[2] (formally Power Platform Power Virtual Agents) imho is unique in it's enterprise AI offering. I truly think Copilot Studio is going to be Microsoft's "killer app" when it comes to companies utilizing AI at scale internally, and not it's Azure service.
Given the broader Mozilla foundation's political biases, will these assistants be censored or heavily curated? I am reminded of when Mozilla chose to ban Dissenter, a free speech plugin powered by Gab, from its store. I've never used it but I found it distasteful that a company working on a basic utility program decided to become political. I don't understand why they cannot just focus on the basics and get those right. Still waiting for proper vertical tabs.
What is a free speech plugin? were there political discussions in these threads? If so it seems like they did the correct thing removing such political plugin right?
> basic utility program decided to become political
It's around 8KB now – so text bigger than 8 thousand characters will return: "414 Request-URI Too Large".
Anyway the document.body.innerText contains all things on the site, including links, menus, buttons etc just 1 per newline. LLM will only recognise if it previously scanned the same website and it did not change much since the last training set. Some arbitrary websites it will not recognise this way and start hallucinating one because innerText removes all the structure from it.
416 comments
[ 3.1 ms ] story [ 284 ms ] threadhttps://github.com/mbnuqw/sidebery
https://www.ghacks.net/2024/12/03/how-to-enable-tab-groups-i...
I've been using Simple Tab Groups [1] for many, many years.
1. https://addons.mozilla.org/en-GB/firefox/addon/simple-tab-gr...
Gee, fucking wow, it's almost as if it's plain a day why they've sucked as a corp, non profit, and culturally for a decade at this point.
Also, like it or not, I think Passkeys/WebAuthn is going to be pushed more and more, so without a user focused way to own them, that's another reason not to try and use a browser (any of them!) as a password store
The minute Mozilla starts being relevant in search, Google will cut their billions USD/year sponsorship, killing Mozilla in the process.
The incentives are not there.
Google needs mozilla more than anyone. It would be a big win for the web if google stopped financing mozilla and had to deal with the consequences.
Google needs Mozilla to exist to prove on paper they don't hold a monopoly with Chrome.
Also, does anyone know if we'd be able to point it to our own LLM instance for the guarantee of our data being secure?
> Orbit is a Firefox add on that uses AI to summarize and answer queries about web content such as articles and videos.
> When a user asks Orbit to summarize or query content, Orbit gathers the context (eg. text, images, videos, etc.) of the page the user is viewing and provides a summary or answer. Orbit works on websites including Gmail, Wikipedia, NY Times, YouTube, and more.
> For the current version, we are using a Mistral LLM (Mistral 7B) hosted within Mozilla’s GCP instance.
It's interesting they're going with Mistral 7B. Is anybody else using Mistral 7B in production? And in what role?
I've considered using it for general knowledge type questions, and as a way to classify information, but would have never considered it for summarization type tasks due to it's limited context size (8k).
I would never use anything smaller than a 70B model for anything even vaguely medical related!
In my experience, even GPT-4o is terrible at revealing information from things longer than a few pages.
It might be an issue with dimensionality reduction in general though. If you think about it, you can't really take away much of what is contained within any given amount of text with text, unless the source was produced extremely inefficiently.
Producing outlines or maybe a form of abstract, it seems to be okay at, but you would never really know where it fails unless you read the entirety of the source text first to begin with. IMO, its not worth risking unless you plan to read the source anyway or its not really important.
Don't use any LLM for something critical. They can't be trusted innately due to their design, why on earth would you use them for something where you need reliability?
As very best I can tell the answer is "because bags of cash"
> Commitment to privacy
Buried as the last sentence in a collapsed box at the bottom of the page:
> For the current version, we are using a Mistral LLM (Mistral 7B) hosted within Mozilla’s GCP instance.
And why is it "...Mozilla's GCP instance", not "We quietly send all your data to Google servers, and everyone pinkie-swears that's totally privacy-respecting"?
We don't know whether this is another time that Mozilla execs have sold out users, or shipped something half-baked and vulnerable.
I'm not saying they're leaking the data (by agreement, or negligence), but Mozilla has mediocre credibility in recent years, and there's nothing on this page that improves that reputation.
Regarding Google, for a long time, their thinking seemed to be "We're Google, so of course anything we do is privacy-respecting", not as guidance, but to justify whatever they wanted to do. Also, every time Google gets caught with their hand in the private information cookie jar, it just mints a new industry standard practice.
I think you read too much HN and aren’t aware about all the stuff going on in the background at Mozilla.
If there’s one company I would trust, it would be them. Their marketing has been mediocre and I’m not 100% sure about if I like their future decisions, but I trust them 100%.
Are there long and vague terms of service documents backed by a pile of lawyers?
There you go, incentive and means. I'm not even confident companies would see that as a problem when it was raised with them directly, in much the same way that Microsoft hosting all the corporate email seems to be just fine.
As the comment above suggested, any information to the contrary would be business-destroying for GCloud. Many of their enterprise users have strict requirements about access to and use of their data.
Re the example of Microsoft corporate email, much the same situation applies. If Microsoft were mining that corporate customer data and using it or reselling it, enterprises would dump them in a heartbeat.
It makes sense. Some gcloud customers are banks. Some are federal govt agencies. Some are foreign governments. Google would not only destroy it's cloud business, but also probably get fined and sued out of existence if it was poking around in cloud or gsuite data.
You get what you pay for (in terms of privacy) at Google. Regular users never pay Google a dime, so they don't get much privacy. Cloud and gsuite users fork over mountains a cash directly, and their data is kept about as safe as can be as a result.
- Currently using Mistral 7B, but ability (by Mozilla) to swap the model used to another open source at any point.
- Hosted "by Mozilla" on their GCP instance.
- No obvious info about what it will cost them to run this since it is free to use.
- No training on user data.
Like others here, I'm very curious about the cost for Mozilla to run this service. It may be less than it initially appears given the 7B model they chose. I do wish they would focus their efforts on creating a very long-term endowment to pay devs for continued Firefox development in lieu of projects like this given the tenuous situation with their Google funding.
I'm not against this kind of thing in theory, but I hope it's being done in a cost-sustainable way.
Mistral has released Codestral under a new license, but that's not the one used here. https://mistral.ai/news/mistral-ai-non-production-license-mn...
In fact, they should have done that a decade a ago.
Mozilla has been around since the late '90s and should have evolved beyond just being a browser company. They launched a VPN service when VPNs were already everywhere, and they did the same with a bookmark manager when others were already offering similar solutions. Mozilla is always catching up, never leading, and that's a common issue with many big open-source and free software companies. They often pretend to be a business that isn't heavily propped up by big tech donations.
If I were leading a browser company, my focus would have been aggressively directed towards small business software. I’d create an internet and privacy-focused affordable minimal business software suite that lives within the browser — a combination of Proton and Zoho. And I’d strongly avoid building things that should be browser extensions.
https://connect.mozilla.org/t5/discussions/try-orbit-by-mozi...
Firefox is like the shitty best option that camps out in its niche, it sucks but it is really hard to push out of the way.
Mozilla Corporation is a for-profit.
https://en.wikipedia.org/wiki/Mozilla_Corporation
>The Mozilla Corporation is a wholly owned subsidiary of the Mozilla Foundation that coordinates and integrates the development of Internet-related applications such as the Firefox web browser,
>Unlike the non-profit Mozilla Foundation, and the Mozilla open source project, ... the Mozilla Corporation is a taxable entity.
You and approximately 5 other people. Paid browsers in 2024 are not a path to mainstream success, especially if you are what Firefox is - a fully independent tech stack down to the browser engine and not a barebones reskin of Chrome. A reskin of Chrome would have very low development costs. Firefox does not.
The bad news is that 12,000 other people would need to similarly pay per month to have a dev team of just 10. I know Ladybird is lean and mean but finding that big of an audience (or, of course, bigger) who would pay for a browser, per month, is likely a non-starter
It would be a much more interesting proposal to start a bug bounty for the damn near infinite Bugzilla queue, although as I understand it some of the hassle of a bug bounty program is evaluating submissions. And don't say "but we'll use an LLM" or I'll throw up in my mouth
If Wikipedia can do it, a browser can do it.
Anyway, at this moment in my life I don't have the emotional energy to launch this project, but if someone does then please tag me and let me know
This project seems especially ripe for success because it doesn't need product market fit, it doesn't need requirements gathering, it just needs execution and a non-retarded bug tracking product
I don't fancy having a random floaty object in the way of my webpages, no thank you.
Edit: It appears to go away occasionally. This UX is unclear to me.
If they shipped vertical tabs I'd probably switch back to Firefox. But that would require focusing on actual browser UX instead of random offshoots. What's their actual product differentiation from Chrome these days?
Not being the progenitor and linchpin of surveillance capitalism.
Because some project manager had "AI" on his performance goals this year.
Firefox is buggy as hell - which is incomprehensible given its age, but older brother Netscape had the same problem 20-ish years ago. The Netscape 4.x days were absolute hell and you could go hardly a day without the browser crashing.
Despite this, it's packed with nonsense no one asked for like Pocket. Which is a coincidence because "AI assistant for Firefox" is the dictionary definition for redundant things no one asked for, with better alternatives preexisting.
At this point Firefox needs to die and something new and manageable - with energetic developers - needs to take its place. Maybe Microsoft could open source the original Edge engine? The one before they bent over for the long dick of Google Chrome.
Same reason Logitech did it, it's a totally arbitrary waste of resources
Don't you worry, I'm sure their new AI Assistant will generate a ton of bugfix code!!11 AI gonna take all teh jobs, or so I hear
But, as a more serious contribution: the sentiment "At this point Firefox needs to die and something new and manageable" is the same one which generates the infinite JavaScript treadmill akin to: "I don't like all those bloated JS frameworks, I want one slim and fast and manegable ... well, except this one other feature ... and this other ... oh, shit, this needs to die and have something new and manageable take its place ..."
It's not that I think Gecko is the bees knees, but I do think it has stepped on more than its fair share of real life landmines, and the Great Rewrite Theory means someone needs to spend all that time re-discovering them
I would, however, agree FF has many "odd bugs".
Truly one of the most self-sabotaging companies I’ve ever seen.
0. Install the Tree Style Tab extension (or whatever vertical tabs extension you prefer).
1. Enable userChrome.css: set toolkit.legacyUserProfileCustomizations.stylesheets=true in about:config.
2. Set browser.tabs.inTitlebar=0 in about:config so the title bar buttons (and, on some OS's, the title bar itself) remain visible.
3. Create =chrome/userChrome.css= in your Firefox profile folder and write the following to it:
I think it’s pretty clear why they keep on doing this type of side project. They are trying anything they can think of to diversify revenue.
> I think it’s pretty clear why they keep on doing this type of side project.
It’s not clear to me. I agree they would have some problems if Google declined to pay them because the next best offer would be lower.
But the best way to keep these payments, or to increase them, is by making a better browser and giving people a reason to use Firefox. After spinning off Servo I lost the last hope I had.
It seems everyone is stuffing AI summary tools into everything, is this something that will retain users or bring in new users? I doubt it.
Web browser crawl on these low end laptops.
This is how Firefox became popular in the first place, by being better.
The “we need an alternative to the “WebKit/blink/chromium” monopoly is what the majority of people will never care about.
In the face of all that JS, there’s only so far a spiffy new browser can improve the situation, aside from maybe drop large chunks of legacy web standards but then you’re breaking large chunks of the web.
The more cynical view is that Google doesn't care at all about Mozilla because the investment is nothing more than a hedge against regulatory pressure.
We’re at a point where the core functionality of browsers is very mature. It’s unlikely that any amount of investment will produce a browser that is significantly faster at things like JS execution or rendering compared to Chrome.
So alternative browsers add things like better ad blockers, more privacy protections, or maybe LLM summaries to enhance the core browser experience instead.
Though in the long term, maybe they could use the market share to make money in other ways.
But if they can’t manage to make money in other ways with Firefox, I’m not sure that they’d be able to do it with Chrome either.
Providing monopoly protection to Google.
They have no interest in actually competing for market share.
about:config, sidebar.revamp = true, sidebar.verticalTabs = true
It's getting there.
Mozilla's brand is "pro privacy", it does make sense for them to launch an AI product with that brand position. I doubt it'll be successful because I don't think enough people actually care about privacy, but still.
I feel like it's a common HN sentiment to say "why don't Mozilla just focus on the browser?!"... the answer is because barely anyone is using it and there's very little they can do to move the needle on that. IMO they're an organization looking for a purpose.
Allowing Backspace to go back a page.
The built-in screenshot tool.
Being able to turn off smooth scrolling.
Support for a menubar, so that I can navigate the features I want quickly.
Being able to choose page encodings (I use a non-Latin charset language).
A usable/useful bookmark manager with things like sorting, tags/labels, timestamps, etc.
"Restore Previous Session" feature.
These are just a few features off the top of my head, I know there are many more.
Of course, for certain sites like Google Docs and YouTube, I have to use Chrome for obvious reasons, but for most browsing I use Firefox (and qutebrowser.)
It’s awful. The best example I experience on a day to day basis that the SPA as a concept is utterly flawed. YouTube is a fucking webpage that fails to work like a webpage and an app that behaves like a students rough alpha. An utterly painful experience, continually made worse by likely skilled devs who are managed by complete bozo losers. But at least the progress bar has an ugly pink hue now.
Too many Google sites behave worse on any non-Chrome browser for it not to be intentional.
Oddly enough, Chrome had somehow lost my settings since the previous time I started it a few months earlier, as if it were a new install. :-?
I've not noticed any problems under YT Premium so far.
I mostly agree, but whoever came up with this shortcut should get choked.
Easily solved by https://chromewebstore.google.com/detail/go-back-with-backsp... their first party webext.
> Being able to turn off smooth scrolling.
Can disable as easily as Firefox -- chrome://flags/#smooth-scrolling
Actually, Firefox lately introduced a bug that will reset smooth scrolling everytime I remote desktop to my Windows machine. Which is annoying AF and they haven't fixed it for months.
https://bugzilla.mozilla.org/show_bug.cgi?id=1912246
> Being able to choose page encodings (I use a non-Latin charset language).
https://chromewebstore.google.com/detail/set-character-encod...
> "Restore Previous Session" feature.
You can do it since.. forever?
Settings -> On Startup -> Continue where you left off
Or just press Ctrl+shift+T when you restart Chrome to restore it manually.
There are some more feature-rich session manager extensions, but they're usually available across Firefox and Chrome.
This only works on pages where extensions are enabled, and only after the extension is successfully activated, so about 30% failure rate for me.
>Can disable as easily as Firefox -- chrome://flags/#smooth-scrolling
This has not worked for me reliably, and the flag has been renamed several times.
For example, on my Mac, it reads "Not available on your platform."
>Actually, Firefox lately introduced a bug that will reset smooth scrolling everytime I remote desktop to my Windows machine. Which is annoying AF and they haven't fixed it for months.
I'm blessed to not have experienced this.
>https://chromewebstore.google.com/detail/set-character-encod...
Same as the other extension comment above, only works once extension activates successfully.
>Settings -> On Startup -> Continue where you left off
This is not the same feature.
And no, disabling smooth scrolling works totally fine as I use it for more than 10 years.
If you're going to exaggerate your points to make a statement instead actually trying to find solutions, I have nothing for you, then.
Yes, obvious.
But left unsaid is that those reasons should lead to the breakup of the (deeply evil) Google ad machine.
I struck a case where Googe Meet is the only website I have found that will not work with Firefox and my headset.
I am so disappointed by Google, I can taste it
which bookmark manager can do that ?
Agreed that it would be nice if they had better focus though.
- continued support for manifest V2 (primarily because ublock origin would stop working if forced to V3 only)
- the firefox address bar is way smarter for any given string i type in than chrome's. it's ability to surface the most relevant deeplinks from my history, vs top level site, vs perform a web search, is night and day difference from the randomness that other browser search bars offer.
- I have the opportunity to use Zen (a Firefox fork) [0] and it's 100% interoperable with my vanilla Firefox instances across devices -- i can even send tabs from my Firefox Nightly on Android to my Zen instances on Windows or Mac. BTW Zen has vertical tab support, (more) first-party multi-profile support, and preserves the address bar behaviors of vanilla Firefox.
[0]: https://zen-browser.app/
You could build an AI Assistant, or you could spend a month bikeshedding some design details of vertical tabs.
I tried this line on my boss. Didn't fly. Back to bikeshedding with the rest of the team. I wonder if the difference is we have to earn our money, and our customers expect an roi. Or, at least something that doesn't mess up what already exists.
First, they have forced telemetry. Okay, it's an early release, it's very basic information, they want to understand how they're doing - I don't like it, but I can understand it. Sets a wrong vibe, though - I had to check if it was from Microsoft and not Mozilla. ;-)
Then, I figured there is no option to use locally-hosted LLMs, which can be something as minimal as simply allowing to configure custom API URL. Somehow, less and less things about Firefox are tinkerer-friendly than they used to be.
That made me wonder if Mozilla used OpenAI-like API, or if they invented their own unique thing for some reason. So I went to look and according to the extension page, it's proprietary ("All Rights Reserved") and I'm too lazy to bother deminifying code from the xpi or remembering how to debug extensions.
Finally, '00s have called and said they wanted their weird floating round thingy UI back, and so I had to return it to the store. (I'm kidding, I don't really mind the visual style. The real issue with that thingy was how it floats on the page. Like, why on Earth it isn't normal right-click menu option that doesn't obstruct the view until it's needed? Or a menu on that toolbar button? It's not even a paperclip to be worth it.)
And then I realized I somehow missed the big "AI you can trust" header, which should've already been a huge red flag.
Finally someone admits that BonziBuddy was 25 years ahead of its time.
Based on this I would assume they are using GCP Vertex AI as that's going to be WAY cheaper than rolling it all themselves and hosting the model on a GCP server instance. I would also assume they'd be using the gcloud SDK for Vertex AI/Model Garden, which I believe means they can't just provide for a different endpoint and payload shape if you had a service elsewhere.
Eitherway, at the (presumed) scale they'll probably also be using GCPs API management service, so I would expect further abstraction between what the extension is sending and what the model/Vertex AI expects as a payload. This means providing that kind of "bring your own endpoint" experience would require more bespoke build-out time.
BUT who knows? Maybe this is just straight up hitting the out-of-the-box GCP Vertex AI REST API directly from the extension like some hobby project.
You metnion GCP Vertex AI, is that something like MS Copilot Studio?
Microsoft Copilot Studio[2] (formally Power Platform Power Virtual Agents) imho is unique in it's enterprise AI offering. I truly think Copilot Studio is going to be Microsoft's "killer app" when it comes to companies utilizing AI at scale internally, and not it's Azure service.
0. https://cloud.google.com/vertex-ai
1. https://ai.azure.com/
2. https://copilotstudio.microsoft.com
Everything else is a waste of time and money and energy.
> basic utility program decided to become political
looks like they wanted to avoid it, isn't it?
Tested on Firefox and Chrome.
Some websites block ChatGPT and can't be summarized this way.
Works in incognito/anonymous mode and doesn't require a ChatGPT account.
You can probably use another AI service with this idea.
I cannot access external links directly. However, if you provide the text or key points from the page, I can help summarize it for you!
and clicking the model selection drop-down produced "Log in to try advanced features"
edit: you're right it requires the user to be logged in to crawl websites. Somehow in my test while writing it I was logged in. My bad.
So while it's handy, it's not perfect.
I just wonder if browsers will limit the amount of characters in URLs.
If memory serves me, there was a limit. But it might be high enough to work fro most pages.
Anyway the document.body.innerText contains all things on the site, including links, menus, buttons etc just 1 per newline. LLM will only recognise if it previously scanned the same website and it did not change much since the last training set. Some arbitrary websites it will not recognise this way and start hallucinating one because innerText removes all the structure from it.
But the URL bar was not the problem.
Personally I’d worry about using this accidentally and with some sensitive data (eg logins).
I do like the idea though, I’d use this with a local model.