Unfortunately none.
Developing a rendering engine that can handle css, html, javascript, while also rendering a website in the exact same way as Chrome and Firefox is a huge tasks, and not something a hobby programmer can whack out in a few weeks. Thats the reason why even Microsoft abandoned their own rendering engine, because things did always look and work different in IE.
This is not true. Pale Moon, Ice Weasel, Librewolf....
Developing a rendering engine that can handle css, html, javascript, while also rendering a website in the exact same way as Chrome and Firefox is a huge tasks
It doesn't have to be from scratch. Not even Apple did this with Safari (they based in on KHTML, the rendering engine of KDE's Konqueror.)
librewolf is a firefox fork, anything thats a fork of firefox/chrome is automatically not counted, because it is inherently bulkier than the original (though maybe more secure)
anything thats a fork of firefox/chrome is automatically not counted
Says who?
because it is inherently bulkier
How is "being bulkier" relevant at all? But let's just go down that route and say that a fork does not necessarily end up in a bulkier product. A dev team could decide to fork, then remove unwanted features from the original project; which is what's happening with Librewolf as far as I know (e.g. no Pocket bs.)
Finally, let's remember that both Safari and Chrome have their roots on Konqueror's KHTML rendering engine. By your metric, we should be saying that they don't count either; because they're "(definitely) bulkier forks" of KHTML.
How is “being bulkier” relevant at all? But let’s just go down that route and say that a fork does not necessarily end up in a bulkier product. A dev team could decide to fork, then remove unwanted features from the original project; which is what’s happening with Librewolf as far as I know (e.g. no Pocket bs.)
now you just have a patched together, disjointed, mess of a browser, on top of a second dev team, who now needs to unpatch it together, re patch it together, and then somehow repackage that. It's just hopeless. It's like trying to turn a full size pickup into a small lightweight town car. It's just not going to happen.
Finally, let’s remember that both Safari and Chrome have their roots on Konqueror’s KHTML rendering engine. By your metric, we should be saying that they don’t count either; because they’re “(definitely) bulkier forks” of KHTML.
It's worth noting that when a fork is building on top of something, there is a point where the original roots are no longer present, or no longer significantly present. It's like saying that android is linux. Which doesnt stop the charts from displaying android separately to linux, or chromeos for that matter. Even if it did i don't like the browsers because they're too bulky so it's not like it influences my opinion anyway lol.
now you just have a patched together, disjointed, mess of a browser, on top of a second dev team, who now needs to unpatch it together, re patch it together, and then somehow repackage that. It’s just hopeless. It’s like trying to turn a full size pickup into a small lightweight town car. It’s just not going to happen.
You are assuming way too much. As if Apple and Google did all this with KHTML. Which lead us to:
It’s worth noting that when a fork is building on top of something, there is a point where the original roots are no longer present, or no longer significantly present.
And what's your point by saying this? What does it matter if the roots "disappear," if the product is good enough for competition?
Even if it did i don’t like the browsers because they’re too bulky so it’s not like it influences my opinion anyway lol.
I mean yeah, but it's my opinion on the matter. Even then my original claim is based on the fact of something being an active fork of another browser. Which is still going to line up with my point just fine.
You are assuming way too much. As if Apple and Google did all this with KHTML. Which lead us to:
assuming too much if you think modern applications are programmed/designed well. Ultimately no matter what you do, having a product be around for a decade, let alone multiple of them, is going to incur substantial tech debt, and significant feature creep. There is nothing you can do about this. It happens in EVERY industry. In fact the only thing that helps to prevent this is an almost religious and fervent dedicated to pure minimalism when it comes to what your software is doing. Look at something like DWM for example.
And what’s your point by saying this? What does it matter if the roots “disappear,” if the product is good enough for competition?
My point is that beyond a certain point, a fork is no longer a fork, but more like a competing piece of software. You see this all the time, look at android or chromeos. Technically "based" on linux, but so far gone that almost nobody considers it linux, i only ever see it mentioned in jokes. Something like prism which is a fork of poly, which is a fork of multimc is starting to get to the point where it's more of an alternate piece of software, than a direct fork. It's twice independently maintained, it's feature set is focused differently.
If you need more examples why dont we have a look at a COW filesystem? When you make a change to a file, a fork is created, and that change is then saved on that forked path, so now you have multiple different versions, throughout the chronological history of that fork. If you have auto-deletion enabled for old forks, as you should, at some point you will have "orphaned" forks. Which no longer represent in anyway the original file, but exist as an independently separate instance of that file, in a different state. It's a similar idea, in a different scale, on a different system. There is also a point where it no longer exists as a fork, but as an implementation on top of that original piece of software. How that's defined is a little more complicated though.
It's a little bit philosophical, and semantical, but my point is simple, if your piece of software exists as a fork on top of another piece of software, you don't get to call yourself "faster" or "leaner" or "more optimized" than the original. Your base browser is still a piece of shit, you've taken a bad car, and repainted it, now it looks a little bit better. But it's still a shit car. You turn a beater into a race car by completely stripping it to bits, at a certain point, it's not really a fork anymore. In the same way that putting a body on a different frame isn't the same as the original.
What bulky browsers don’t you like?
it's not like i've literally named them or anything.
assuming too much if you think modern applications are programmed/designed well. Ultimately no matter what you do, having a product be around for a decade, let alone multiple of them, is going to incur substantial tech debt, and significant feature creep.
I still don't understand what this has anything to do with "forking makes a product bulkier," the original claim. At most, what you're saying is that the fork will have its own set of tech debt. But that doesn't make it bulkier by default. Again, a fork of Firefox without the Pocket and "experiments" crap will be lighter.
My point is that beyond a certain point, a fork is no longer a fork, but more like a competing piece of software.
Well, yeah, isn't that the point of forking? I still don't see why a forked browser being "yet another competing browser" is a bad thing. It's the opposite!
if your piece of software exists as a fork on top of another piece of software, you don’t get to call yourself “faster” or “leaner” or “more optimized” than the original.
I completely disagree with you, and I think I know why you think the way you think. It seems like you assume that all forks:
Must always follow the development of the original software. Nope. Not true. It can happen, but not with all forks.
Are inherently bulkier because devs add features on top of it. Which again, it's not true for all forks. Some forks solely exist to remove crap in the original software.
Your base browser is still a piece of shit, you’ve taken a bad car, and repainted it, now it looks a little bit better. But it’s still a shit car.
Man, have you never seen TV shows about mechanics taking shitty cars and making them awesome? Yes, they strip it to pieces, and reassemble said pieces. That's part of engineering practices. It appears that you have a narrow way of seeing how software development works. Devs don't need to take in the whole "shitty project" and be resigned to deal with it. They can take the good parts, and rewrite the bad parts. And that's just one example.
it’s not like i’ve literally named them or anything.
You haven't mentioned any browser that's a fork from Firefox and that is also bulkier than Firefox. Librewolf? Bulkier than Firefox? Really?
I still don’t understand what this has anything to do with “forking makes a product bulkier,” the original claim. At most, what you’re saying is that the fork will have its own set of tech debt. But that doesn’t make it bulkier by default. Again, a fork of Firefox without the Pocket and “experiments” crap will be lighter.
I mean yeah, removing two features removes two features, that still doesn't optimize the entirety of the browser, all of the rest of the browser will behave the exact same with no difference (unless, somehow, those features are actually so badly implemented they actively impede performance) Thats like taking a corolla and removing the entirety of the interior to strip weight, and doing literally nothing else to it. It's just marginally faster now. Handles a little better maybe. Everything else is still stock though.
Well, yeah, isn’t that the point of forking? I still don’t see why a forked browser being “yet another competing browser” is a bad thing. It’s the opposite!
I dont inherently have an issue with forks, i have an issue with stuff like thorium, you forked chrome, that's great, chrome is faster than firefox by most accounts. You made it maybe 40% faster in some instances? Cool. It's still basically chrome though. They describe it as The fastest browser which, if that's true, that's great! It's still basically chrome though. The issue here is that the modern web, and the web browsers designed around it are just massively overbuilt and bloated. We're solving problems that shouldn't exist, and we're adding features that do almost nothing other than cause problems half the time. That's not a good starting point. Unless you completely rip everything out, and rebuild it. Which is inherently not what a fork is.
Man, have you never seen TV shows about mechanics taking shitty cars and making them awesome? Yes, they strip it to pieces, and reassemble said pieces. That’s part of engineering practices. It appears that you have a narrow way of seeing how software development works. Devs don’t need to take in the whole “shitty project” and be resigned to deal with it. They can take the good parts, and rewrite the bad parts. And that’s just one example.
Yes, you wanna know what they do most of the time? Completely strip it down, and then rebuild it. If you have done that with either chromium or firefox, you wouldn't be calling it a fork of chrome/firefox, and everybody would ALL over it. As far as im concerned, any fork of either of those browsers is just removing the most egregious garbage, which is a good thing, but it's still just a bad browser underneath the removed garbage.
Let's compare forks, firefox and librewolf, both browsers i have installed, and both browsers i use. As far as i can tell they're effectively the same thing. Librewolf probably has some cruft removed and some good defaults compared to firefox, but other than that, nothing inherently different.
Lets look at chrome and chromium why dont we, this is actually just the reverse, but wouldn't you be surprised to discover that i dislike chromium equally as much as chrome because they have equal design decisions? It's almost like 90% of the feature base is going to be identical between them or something!
Thorium? I've not used that one yet, i assume it's just chrome, equally annoying to use, but with the slight added benefit of having marginally less time to ponder my bad life choices in between bouts of loading heavily ad bloated sites, and JS infested messes of web design. Plus all the ram that it probably still consumes. Because it's a web browser, why wouldn't it.
I don't even where to start, but let's just say that I now see where you're coming from. You seem to have an issue with this Thorium browser, then project your perspective over other projects that are also forks. Just because one implementation sucks doesn't mean they all do.
And we have different ideas of software engineering. To you, features are just the things that the user can interact with. When you say things like "that's like someone stripping the interior of a Corolla and doing nothing else." Except that I was thinking, precisely, of working on the whole car, including tweaking the engine, the electrical system, the fuel pump, etc. Sounds like a lot of work? Maybe. But it's better than building a car from scratch.
including tweaking the engine, the electrical system, the fuel pump, etc. Sounds like a lot of work? Maybe. But it’s better than building a car from scratch.
that would be more of a rebuild than a fork. If your end product is more similar to the OEM car than it is to the end product. It's more like a fork. If the car is more akin to a custom built racer, than the OEM, it's a rebuild.
There are plenty of browsers. Dillo, NetSurf, surf, w3m, Lynx, Links, Via, Midori, Pale Moon although it's based on a fork of Gecko, Tunnel, qutebrowser. And there are even options for a search engine, although the only one worth considering that isn't just a layer on top of other search engines is Kagi which costs $10 a month, and I wouldn't exactly call it minimalist.
The problem is that no browser can allow you to escape the horror that is web standards & practices that have been developed over decades and are almost unchangeable, without sacrificing basic web functionality and just making it a worse experience than it needs to be at least. The fact is that practically the entire web is reliant on JavaScript, on top of HTML and CSS which take a lot more resources to utilize/display than it looks, meaning 3 interpreters constantly running that must be sandboxed to each tab you have open with a lot of overhead to manage security.
In an ideal world we'd all just be using provably-safe high-performance compiled WASM-but-stronger (from functional languages or more likely Rust or something less boiler-platey but similar), without having such a complex and fucked dependency situation*, where we wouldn't need to sandbox interpreted languages and slaughter performance. Of course, in an ideal world, we also wouldn't have to be concerned about aggressive tracking, ads, clickbait, SEO abuse, scams, or even malware, so there's not much use in imagining a reality where we actually have quality web browsing.
The actual answer to using the web without the fucked-ness of browsers is to not use a web browser at all for sites you use frequently. Use stuff like this instead.
*seriously, you can write the most basic website with JavaScript and it'll probably rely on tens of thousands of expressions of code which realistically should just be expressable in like a small page or two, you do webdev and you'll probably accidentally be implicitly committing a sacrifice to some Aztec God in order to check if a number is even or odd
Also just imagine if all of web dev was just ML/Scala/Rust/Swift/Erlang without compiling to JavaScript 🤤 That is the definition of a perfect universe
The problem is that no browser can allow you to escape the horror that is web standards & practices that have been developed over decades [...] practically the entire web is reliant on JavaScript, [...]
I've been saying it for a while: continuing to play catch is a losing move for Mozilla or for any independent browser maker.
The real move, is to switch to or at least integrate an alternate internet, something that uses a protocol that is simpler and more limited by design - just get rid of Javascript (or of "remote execution", really) and you instantly get a much leaner, much securer internet design.
I've heard pretty good things about the Gemini protocol, but IMHO they went too far too extremist into the "text internet" philosophy, and as a result is a raw downgrade from Gopher. Gopher could actually be a good option.
I'll definitely have to check out a few of those browsers at some point. It's kind of insane how much tech debt we've accrued over the years.
I think honestly we just need to start waning off half the shit we support. Minimize the amount of support required, and somehow manage to provide a smaller attack window so that way we can stop writing protections for problems that honestly shouldn't even exist to begin with. Bonus points to microsoft for creating security certs that don't do their jobs because hahafunneemalware.exe is signed by fucking oracle of all people, and i guess we should just blindly execute that file because it says it's trustworthy!
Though it would be interesting to have a sort of "web browser" which is actually just an application based on plugins for different frontends, for stuff like yewtube, we do only use a handful of sites from time to time. Plus maybe a basic web fronted for stuff that isn't JS because honestly who wants it anyway.
I mean, let's be real, what major function has Mozilla implemented into Firefox that hasn't been opt-out? And no, UI doesn't count, I'm talking features.
The problem isn't the existence of AI. The problem is the inescapably of it and how, under Microsoft or Google, it will harvest your data whether you like it or not. When you tell them "fuck off, leave me alone, and keep my words out of your AI's mouth", they're not going to listen. Profit motive requires them to invade.
Mozilla is a non-profit, and they've long been very good about letting you opt out things, and listening. I'm not worried about them putting AI into Firefox, because I can be reasonably sure it will be optional, in a way I know the others won't.
I'd rather they didn't go chasing this car at all, to be honest, because they're not likely to catch it, but whatever. They're renewing focus on the browser and I'm taking that as a win.
I can't contest the first point cause I'm not a firefox junkie, so I won't.
What I will contest is that the existence of AI, or, deep learning, or LLMs, or neural networks, or matrix multiplication, or whatever type of shit they come up with next, I'll contest that it isn't problematic. I kind of think it is, inherently, I think it's existence is not great. Mostly because it obfuscates, even internally, the processing of data, it obfuscates the inputs from the outputs, the works from the results. You can do that with regular programming just fine, just as you can do most of the shit that AI does with normal programming, like that guy who made a program that calculates the prices for japanese baked goods and also recognizes cancer, right. But I think AI is a step further than that, it obfuscates it more. I kind of am skeptical of it's broad implementation.
For trivial use cases, it's kind of fine, but I think maybe use cases we might consider trivial, otherwise are kind of fucked, maybe. AI summary of an article? I dunno if that's good. We might think, oh, this is kind of trivial because the user should just not really trust what the AI says, but, as with all technology, what if the user is an idiot and a moron? They might just use it to read the article for them, and then spout off whatever talking points and headlines it gives them. I can't really think of a scenario where that's actually a good thing, and it's highly possible. It might make it easier to parse an article, like that, but I don't think that's actually a good or useful tool, it's just presented a kind of illusion of utility, most especially because it was redundant (we could just write a summary and have it at the top of the article, like every article on the face of the earth), and it was totally beyond our control, at least, in most circumstances.
Also, the Mozilla Foundation is nonprofit, but the Mozilla Corporation is not. The Foundation manages the Corp, which manages Firefox development. So depending on which one you're referring to, it might be a non-profit, or it might not be. In any case, the nonprofit is a step removed from Firefox development, which I think is an important side-note, even if it's not actually that relevant to whatever conversations about AI there might be.
Perhaps, comically, it is the perfect representation of the world as it is now: “knowledge” in people’s brains is created by consuming whatever source aligns with the beliefs that they think are theirs. No source or facts are required. Only the interpretation matters.
One of 70 flavors of Chromium including the "privacy centric" Opera who run Chinese loan shark gangs for some reason, Edge which is Microsoft Chromium and aside from hardware acceleration capabilities is pretty meh, and Brave which despite operating their own separate search engine index are one of the most likely to sell your data and/or kidneys
Rapidly Enshitifying Firefox
Safari - no comment
Whatever the fuck Gecko is...
Tor Browser (for people with infinite time to wait for pages to load, or maybe just drug dealers)
RE: point 1, I'm a fan of Vivaldi, a privacy-focused highly modified chromium build developed by former Opera developers who were disillusioned by the direction that company went in.
Librewolf is built on Gecko, people often accredit it as a "firefox fork".
Tor Browser seems cool, it's what I use on my phone whenever I have spare time to let it load before searching things which don't require a lot of bandwidth. I'll edit the above list.
Mullvad? Is that some kind of slur? I've never heard of that but searches say it's a VPN client. ¯\(ツ)/¯
i literally dont open the steam client anymore, that's how bad it is, it regularly consumes an ENTIRE gigabyte of ram doing literally nothing in the background, the UI is buggy, messy, and just generally hard to navigate. It's also just not a very good platform, steam doesn't have a particularly good linux release binary.
"Doing nothing" is probably downloading an update. There's also a difference between reserved RAM and actually used one.
For example .NET applications grab RAM when they need it, but they don't just free it afterwards if not necessary (Like it needs 1 GB, uses that, but when the work is done your task manager keeps showing 1 GB). This helps performance, if the application needs RAM again a short time later it's already reserved and ready to go.
The whole behavior changes when Windows is low on free RAM, then applications are forced to free up their reserved RAM so you don't start swapping too much.
Overall this means: The more RAM your system has the higher the perceived RAM usage of your system. Unused RAM is wasted RAM and it's easy to free up some if you actually hit the limit. As long as your RAM is not full applications will happily use more and hold onto it to be more responsive.
everybody says this in response to my statement. Steam is doing NOTHING. I've checked, it's not downloading an update, it's not pre compiling shaders, it's not caching them, it's not doing ANYTHING. I don't know if people just don't understand how obscene this is, or think im just wrong.
Heroic, a launcher for both epic games, and GOG. idles similarly to steam uses a bit less ram though, launches multiple times faster, and is much more usable. And this is ANOTHER web app.
I use linux, it reports as used ram, not cached ram. Again, im not wrong. I understand the concept of caching ram, i understand the concept of actively used ram, this is not cached ram. That's also not a very complete explanation of ram caching, ram caching helps in the event that you use that same information, that was already cached. For example, you open a game, or a project, and then close it, it's pretty likely that some of that will be cached, so that way when you open it again, it launches quicker (particularly if you open and close it multiple times)
again i use linux, i literally hand formatted my swap partition, i understand how this works. Also generally, how swapping works, is that it actually swaps cached ram into swap, and only upon swap being filled or almost full, does it actually start to clear cached ram. This may not be the default behavior on windows though, since solidstate drives handle different these days. But this is the default on linux (configurable obviously)
The last tidbit is not quite true, it's true to a point, your system will idle at a higher memory usage, the fundamental problem here is different, actually unused ram is wasted ram, having too much ram, does actually just waste ram. (though im sure linux would absolutely love to use it for cache) Caching everything is an obscene proposition, considering that most people don't have a lot of ram. Chances are, if you have 16 gb of ram, and upgrade to 32, you will see a bump in max used ram, and overtime cached ram. However when we upgrade from 32 to 64 in this same scenario, you probably won't notice a change at all, except for the outliers in the data. Though i suppose you might cache more things, but at that point it really doesn't matter tbh.
It's compounded by applications being heavily bloated and stupidly non performant, i would argue it matters more to have more efficient usage of ram application wise, than it would be to have better ram management OS wise. This should be fairly simple to understand why. An application using 1GB of memory, when in reality it should be capable of using as little as 250MB for instance, is the single worst form of wasted memory you can possibly create, because that memory CANNOT be used for anything else. Period, until the application is no longer running.
That said, again to reiterate my original point here, steam on idle, closed, in the background, not in the foreground, no updates, no game updates, etc... Consumes an entire gigabyte of ram. Why? Because the web front end runs at ALL times, for some reason. Steam is running an entirely separate web browser installation, 24/7 because, why not i guess? Fun fact, you used to be able to disable it under linux, and steam ram usage would drop to under 200MB.
Here's another funny pain point of ram caching, when dedicated applications like discord, and steam, start using web backends, you compound this with software bloat, they all use a web backend, and instead of running on a single web browser like all of your tabs, they now run in THREE separate web browsers, thats THREE times the idle wasted ram, because you have three separate web browsers, all running, and all individually sandboxed. This is actually just bad ram management, inherently. It's more secure i suppose, provides a development benefit, technically. But to the end user, and the ram itself, harms it actively.
Ah, I didn't expect it to be actually used RAM. Maybe this is a Linux issue with the Steam build then? Here is my Windows 11 task manager, Steam just downloaded 10 different game updates (so did plenty of work) and is now idle:
In total 516.5 MB RAM on a machine with 32 GB (22 GB free at the moment), if there was any pressure on RAM usage it would probably go down further.
Either way, since upgrading to 32 GB RAM nearly a decade ago I haven't had a single issue with RAM usage (While with 16 GB I actually had games in the past where I ran out of memory). So it's no big deal as far as I'm concerned and if I'd actually run any applications that needs tons of RAM I'd quickly upgrade to 64 GB and be done with it.
The only way this would be annoying is on low-end machines, like 4 or 8 GB RAM in total, but those have plenty of issues anyway in regards to games (otherwise why would you install Steam?). On a high-end machine complaining about 1 GB of RAM is a waste of time in my opinion, there are a ton of better topics you can rage at.
it could very well be a linux build issue, it wouldn't surprise me honestly. The main telling thing for me though is that heroic uses the same if not more ram, and is actually many times more performant.
My main problem with the ram usage is that steam takes equally as long to launch as it does to boot a game, which is super annoying, not including any updates it hasnt performed yet. Heroic launches faster than my web browser does, even though its literally an electron app.
I wouldnt really care how much ram it used if i could just close it when i was done with it, and have it go away, but it's such a mess that's not really feasible.
The whole "just buy more ram" is not really a solution im a fan of. My system has 16GB. which is fine most of the time, it gets stretched sometimes, most of that ram is used by browsers, (because three different containerized browsers run simultaneously for some reason) so my idle ram quickly becomes 8GB. 8GB is still a lot of available ram though, if steam didn't use an additional gig on top of that it would only be beneficial. Maybe i'm just too jaded in general. But saying just get more ram is kind of like saying "just repair a cracked back glass on a phone" When i never wanted to have a piece of glass on the back of my phone which could get broken in the first place.
Although to preface this, i AM a linux user, and i can routinely enjoy a machine with 4GB of ram through the magic of non shit software. i3 + debian cooks. Idle ram usage under 100M is trivial when you aren't running any bloat. In fact, my server actually on average, uses less ram than my workstation. It's probably sitting at like 4GB util right now, running a handful of services, and a handful of game servers.
I'm not sure if you remember, but site rings were what you used instead of Google. They were useful.
And I've seen some guest books with lots of people at some point in my childhood, but about half a year after that everybody firmly chose in favor of hierarchical boards.
And I don't share that hate for <marquee>, it served the purpose of showing you a long line in a small space, implicitly saying that it's secondary temporary information, a bit like on TV.
And what's wrong with animated GIFs, animation is nice.
It really grinds my gears. Why does my bank insist on installing an app to approve transactions, and why does that app have a huge background video playing every time i open it? It really should consist of an MFA code generator.
LoL, and I had been contemplating switching to Firefox on my phone. Fucking nope! Not gonna board a ship that has decided to follow the pack into the ice...
I'm using chrome on phone, because it's basically part of the operating system, but I did like Fennec. It's a fork of Firefox mobile with a few more privacy features (or so they advertise)
I did address that. I am fully aware of Brave's faults. But the benefits to me outweigh the negatives. Besides which, there isn't a browser in existence that doesn't collect data on you. Not unless you go ahead and compile your own. So I choose the one that is blatantly not following the rules and allows me some leeway to enjoy the web the way I like. It's also not the only browser on my phone. Chrome is obviously still here but so is Duckduckgo and I do have Firefox on here even though I never actually use it.
Cromite and Fennec I've seen and don't trust their security at all. Better the devil you know and all that. But Iceraven is new to me and will have to look into that one. So thanks
Jumping ship means trying to go in a different direction. I'm not gonna upheave my entire online presence just to get onboard with a company who is not only going the same way, but is woefully behind in the race to go that way.
I get that people are upset, because this fucking buzzword is haunting us. I'm just hoping that they don't jump on that BS-bandwaggon and create something actually useful. But we'll see I guess ...
Yeah, it's a hate-train for AI, I definitely get it, but Mozilla seems to be using it for actually useful things. Offline translation and fake reviewing checking for Amazon are pretty cool, in my opinion. Don't get me wrong, I'm not brand loyal, and I'm ready to jump ship to a FLOSS alternative as soon as they do something stupid. I'll just keep using Firefox until they do.
The paradox of tech right now "we are going to build the most complex technology known to man into our product in the next 12 months. Are we hiring record numbers of people to get it done? No. We fired a bunch of people and everyone else will just have to be extremely hardcore."
I bet I know much more on the topic than you, but please enlighten me on which part of this is complex?
The core concepts of DNNs are taught in high-school, and putting them together can done by a Bachelor student. Shit, people often advise writing a NN libraries as a good learning exercise when picking up a new programming language.
I think mathematically illiterate people assume that incredible results necessarily imply complexity, but that's simply not the case here. Or the idea that unknown things are necessarily complex, maybe.
The main reason DNNs are popping up is because we finally have the hardware for it. And the second reason is that tech companies have the resources (both financial and in terms of available data) to throw at it.
They're refocusing on Firefox and continuing the ai stuff they were already doing. They fireded people who were working on fediverse and metaverse platforms. Did you even read the article?
It depends on what they mean by AI. I can think of oodles of great uses:
An AI-powered adblock that removes all trackers, cookie confirmation popups, those annoying “please subscribe” popups, etc. would be badass. It would be virtually invisible but it would make the internet usable again.
A content filter that magically extracts the recipe you’re looking for out of the stupid blog post they write for SEO
Or to expand on that, an AI that goes through the page of search engine results and removes the ones that are SEO spam instead of actually useful content
An AI that can review at a page or email and determine if it’s a scam would save a TON of people by pointing out suspicious features.
Basically anything that requires you to copy data from one context to another is a good use of AI. You could probably have a nice resume-filling feature, for example.
But yeah, Mozilla will probably just go for a “chat with your browser” feature. Total waste of space.
All of those could be terrible to be honest, because AI is a data tracking vacuum. An AI adblocker or content filter sounds cool at first, but it would mean it reads and analyzes your data, just like the shit you do with chatbots too. Reading your mails? That's basically what Google does for years with gmail, that's why they have such a good spam filter. I agree that a chatbot would be kinda useless though, even if privacy friendly, which in of itself would be great but I just don't see the use. This could simply be outsourced to a website.
If they're local they'd be basically useless due to a lack of computing power and potential lack of indexing for a search engine chatbot, so I doubt it. It would also have to be so polished that it wouldn't require further user knowledge / input, and that's just not a thing with any local LLM I've come across. Mozilla can gladly prove me wrong though. I certainly wouldn't mind if they generally can make the whole process of local LLMs easier and more viable.
The requirements to run good local LLMs have really been shrinking this past year… I have a lot of faith that there is a generally useful yet tiny AI tool within the grasp of Mozilla.
I can understand your thinking, but it could be as simple as giving the user the option to outsource the computation to a secure something or other, if their machine can’t handle it.
And yeah, the requirements are still quite high, but they are being reduced somewhat steadily, so I wouldn’t be surprised if average hardware could manage it in the long term.
Edit: For the record, Mozilla is one of the only companies I would trust if they said “the secure something or other is actually secure.” And they’d likely show actual proof and provide and explanation as to how.
Yes, but what would a local model do for you in this case? Chatbots in browsers are typically used as an alternative / more contextualized search engine. For that you need proper access to an index of search results. Most people will also not have enough computing power to make use of any complex chatbot / larger context sizes.
Pennomi wrote a whole list of potential ideas. And honestly, while I agree that local LLMs on typical hardware are underpowered for most tasks, it's possible they would have the option for those that can run it.
People are getting all upset over this announcement without even knowing what their plan actually is, like the word "AI" is making them foam at the mouth or something. I'm just saying we should reserve judgements for when we have an idea of what's happening.
But it's a hellishly expensive thing that seems to not attract enjoyment from current Firefox users, and seems unlikely to bring new users, and (again) seems to be prioritised over other things that could better use the money, like developers, so...
to be fair, most people aren't using physical keyboards with their phones, this mostly applies to tablets (which are much less popular) and android tv (where firefox isn't even officially supported anymore)
anyway, even without that issue, the android version of firefox is kinda janky and regularly gets stuck while loading websites
bt keyboards can be absolutelly critical for some users with disabilities. most of the developing world can't afford to own several devices like a computer and a phone.
They're not adding ai to Firefox, at least that's not what the memo said, the memo said they were refocusing on Firefox (firing people who worked on their metaverse thing for example), and doing so on the side.
They specifically stated that Firefox mobile is a big focus moving fowards
I know there is currently a massive PR campaign for a power grab to consolidate control over AI software. They want to control the means of generation. Only MozillAI can save us from King GhAIdorah!
Sorry I'm upsetting you. I know we're entering an acceleration of technology at a time where our institutions globally are in an absolutely horrendous state. People on all sides are brainwashed as hell. The AI watchdogs are insane as well. What's left but gallows humor? I do hold out some hope though.
You cannot upset me more than the current common misunderstandings that everyone has about AI already does.
I don't think you understand the implications of undetectable AI to shift social conversation or the kind of world that those AI owners want to create.
That might actually be the kind of thing where open source AI could help. At least I hope. To detect bias, lies or AI powered filtering / sorting of content.
Ok so this is one of the naive thoughts that makes me upset.
The open source community can't even make a distro of linux that is out of the box functional for everyday users and you think somehow they are going to be able to outcompete billion dollar companies that can afford the best gear and devs?
Look, I bought in heavy to open source early on in the 90s, and have done my best to go open source for every tool I can, but the simple fact is that even the 'best' open source projects are severely lacking in aspects and YOU CAN'T TRUST DEVELOPMENT OF AI TO THAT.
Compare The Gimp to Photoshop. It isn't even close, why? Because Adobe has a fucktonne of cash to throw at their projects and they have clear direction and motivation.
I don't like it
I'd prefer a fully open source world
But it isn't going to happen, and open source AI will always lag behind corporate AI, and considering how fast it has been developing, even being 3 months behind renders a tool useless as an AI detector.
We aren't prepared for this and 90% of what everyone on the internet says about AI is poorly informed and full of confabulation, and WORST of all, when you try and explain this to them they get antagonistic.
We have already seen the threat AI can pose in 2016 with Cambridge Analytica helping to hand trumpty dumpty the election by using AI to focus target vulnerable facebook groups.
AND THAT AI WAS A FUCKING INFANT compared to what we have now.
It's going to be so bad and almost none of you have the slightest clue.
Honestly, most of what Cambridge analytica did was blackmail, illegal spending, and collusion between campaigns that were legally required to be separate.
Much of the data processing/ml was intended as a smoke screen to distract from the big stuff that was known to work and consequently legislated against. The problem is that they were so incompetent that the distraction technique was also illegal.
Maybe the machine learning also worked, but it's really not clear.
You already have AI in Firefox - local translations for example. Developing local AI aligns perfectly well with Mozilla's goals, but it seems people panic as soon as they see the two letters together.
