Been using as my sole web browser on a daily basis since it was called Qupzilla, maybe for 10+ years as I switched to it from reKonq.
I really like it because it's fully and truly integrated with KDE, not needed a whole sort of patchs to integrate to it like Firefox needs to.
It's actually great. Not sure about the "Qt doesn't upgrade WebEngine often enough" in other comment (what is "ofteh enough"? I got several updates for it over the year) and of course it can be lagging from stuff from mainstream, but I think for 99% of users it's just fine.
Granted, you can't use Chrome/Firefox plugins, which may it seem not worthy to some people, but there's a basic adblock and greasemonkey extensions shipped with it with default which blocks most of stuff and even you can install a script to speed up youtube ads so that annoying ad will run out in a sec or less. Apparently you can write your own plugins for it but last time I wrote to one of its devs the api wasn't even documented.
There are some quirks on it, though, like the user agent thing - I set up as the latest chrome user agent for every website except accounts.google.com where I left it as the one shipped with Falkon so it lets me sign in, and yet it shows a warning about "upgrading" to another browser.
Ironically, such "warning" also shows up when browsing discuss.kde.org. Yes, the very KDE discussion board warns you against using KDE's own web browser.
Since some years ago I have a silly idea about a plugin that transforms tabs into some sort of Vim buffer list thing that can be filtered by the url bar, but am too incompetent about C++.
With kdeconnect when I have some sound playing in firefox, e.g. a youtube video, I get audio controls directly from my phone. There's a million features, check it out here: https://kdeconnect.kde.org/
KDE has integration with Firefox/Chromium too, via plasma-browser-integration. You can search browser tabs and history via the KDE menus, and share URLs with KDE via the browser context menu.
* The KDE system for network protocols ist called KIO. KParts is for user interface components.
AFAIK, Falkon isn't using KIO because QtWebEngine has no extension points for that (and if it did, somebody would need to write HTTP2 and HTTP3 implementations or Qt wrappers for KIO).
Yeah Falkon is not using KIO. Same for Konqueror for network request. In the old time of QtWebkit and even older time of KHTML, the webview would reuse the Qt network stack, but now it's reusing the implementation from Chromium.
Btw KIO is now using QtNetwork for it's http implementation, instead of having it's own. So http2 is now also supported.
Not really. You can use XDG portals to have native dialogues in Firefox, but GTK developers call this an abuse of portals, so this will go away at some point. Portals are intended as a way out of sandboxes like flatpack, not to plug-and-play file picker dialogues.
Wow, I'm not keeping up with all the new developments in the Linux desktop land, but that surely sounds like a step back.
So let's say I'm building an app like Blender or Reaper - I'm sidestepping the need for most of the OS-specific/native-widget components, because I already need to do a whole lot of very complex and custom rendering, and that is the saner choice when going for portability. But I would still like to maintain a certain level of basic OS integration, for example a native menu bar on macOS, matching the light/dark theme with the OS, or perhaps... a native file picker?
What are my choices on Linux? Link with Gtk, and make the app look out of place on KDE? Link with KDE, and pull in half of it with me when installed on Gnome? Link both? Summon Cthulhu?...
Sounds like we've had a solution for a moment, and now we want to remove it, because think of the yaks?
If you use Qt the framework does its best to make sure the native thing will be used where appropriate. QtCore, Gui and Widgets is like 15 megabytes total and that will already allow 90% of Qt apps to run. A bazillion packages being installed if you want to install one Qt app is just the fault of your distro's policies on how software should be distributed
Yeah, about that... fracturing at every possible level.
- Linux vs Free/Open/Net/Dragonfly BSD
- Linux distro 1 vs X vs A vs Ω vs ...
- glibc (with all its warts like versioned symbols) vs musl vs BSD libc
- systemd vs sysvinit vs rc vs OpenRC vs daemontools/s6/runit/...
- Who "owns" /etc/resolv.conf?
- apt vs (yum / dnf) vs pacman vs apk vs xbps vs emerge vs ...
- Flatpak vs Snap vs AppImage
- X11 (and libx11 vs xcb) vs Wayland (which protocols/extensions are supported?)
- OSS vs ALSA vs PulseAudio vs Pipewire vs sndiod vs ...
- Gtk (2/3/4) (with Gnome or without) vs Qt (with KDE or without) vs Tk vs direct X11 vs SDL/glfw/... vs an obscure toolkit last updated 15 years ago
inb4 it's about choice, the heck I'm supposed to choose as an app developer? inb4 "follow the standard", half of these are not standardised but still in widespread use? inb4 distro policies, which distro - top 10 on distrowatch looks like more work than macOS+Windows combined? Delegate to package maintainers, and my app is "fixed"/patched beyond me being able to debug/support? I choose what seems to work (for example, escaping via the XDG portal) and I'm getting rug-pulled?
Do you have a reference for this portal going away? I was really happy that work was being done in that area, and was optimistic that even Java apps might get sane file pickers in the future...sigh.
> Granted, you can't use Chrome/Firefox plugins, which may it seem not worthy to some people
Why is it granted? WebExtensions is an open spec, it is possible to implement it in any browser if needed. (Orion does this, although I think they don’t cover 100% of the features yet.)
But an ad blocker and userscript support can take you long ways, yeah.
and yet it shows a warning about "upgrading" to another browser.
Ironically, such "warning" also shows up when browsing discuss.kde.org. Yes, the very KDE discussion board warns you against using KDE's own web browser.
Such is the propaganda of Big G, that user-agent discrimination is actively encouraged to further Chrome's marketshare.
> Falkon is a KDE web browser using QtWebEngine rendering engine, previously known as QupZilla. It aims to be a lightweight web browser available through all major platforms. This project has been originally started only for educational purposes. But from its start, Falkon has grown into a feature-rich browser.
While I applaud competition among browsers, I think that the renderer/webengine is such a big external component nowadays (especially if you lump in the JavaScript engine) that it might be more accurate to say that you're providing a skin, rather than a new browser.
Anyway I miss Konqueror and am happy to see any of its descendants carry on its legacy.
Not likely unfortunately thanks to Mozilla, which in all its wisdom, dropped embedding support from Gecko. This means it effectively cannot be hosted outside of Firefox/Seamonkey and forks thereof. This is contrasted by WebKit, which is largely UI toolkit agnostic and can be dropped in just about anywhere.
Funny thing, Epiphany (now known as GNOME Web) was originally built around Gecko, but when Mozilla axed embedding, the project had no choice except to switch web engines and so it’s now built around WebKit instead.
Embedlite is still maintained, even though it's not a Mozilla project nor very active.
Since it's the base for the Sailfish browser, it even does have Qt bindings.
I think that was a fine decision by Mozilla, even if I'm not happy with it. Firefox is already a web browser with barely enough market share to support it. Wasting resources on making even smaller special purpose web browsers integrate with the engine is not a move for a company that has been struggling with a lack of independent funding for years now.
Of course the decision was made back in the Firefox 5 days, but back then Chrome was rapidly growing, mostly at the cost of Firefox. Had Firefox remained embeddable, I think it would've only been held back in important performance areas like threading and rendering because of the API around the web view.
Interesting enough, with the Firefox for Android rewrite, Mozilla actually went back to making Gecko embeddable: https://wiki.mozilla.org/Mobile/GeckoView This design decision is also claimed to be the reason for them to disable about:config in production builds, because misconfiguring geckoview can disconnect it from the app and leave users with a broken browser (apparently).
So now there is an official way of embedding Gecko in your application, but it's limited to Android only.
I still find the decision questionable. No embedding meant that Gecko missed out by default on the wave of embedded WebViews in the early smartphone era as well as the “desktop web stack” boom that Electron/CEF came to dominate.
Additionally, I believe that one of the reasons why Firefox has struggled to maintain market share is because it’s a bit of an acquired taste. Relative to Chrome, Safari, etc it’s kinda quirky and not to everybody’s liking. Alternative browsers wrapping Gecko could’ve helped a lot here.
Camino was a great example — Mozilla couldn’t justify building a Mac-specialized Firefox and that makes sense, so Camino filled that gap instead and became quite popular among Mac users. When Camino had to close up shop following the removal of embedding, most of those users didn’t switch to Firefox but instead to Chrome and Safari because those were better suited to their needs.
"Wasting resources on making even smaller special purpose web browsers integrate with the engine is not a move for a company that has been struggling with a lack of independent funding for years now."
So says the narrative. Mozilla has money, but prefer using it for C-Level salaries : Mitchell Baker raised her salary from 2.5M$ to 7M$ within 4 years, all the while laying-off developers and spending a shitload of money on seminaries and others BS expenses. Meanwhile, the market share of Firefox dropped.
And I assume you have noticed that privacy is not their main concern.
Such a pity because Phoenix/Firebird/Firefox brought a lot to the Web. But I guess this happens when Corporate people take over any organisation.
>raised her salary from 2.5M$ to 7M$ within 4 years, all the while laying-off developers and spending a shitload of money on seminaries and others BS expenses.
Seminaries? Kagi turns up nothing. Could you explain?
any proof that they are losing money on VPN or Pocket? That's my only concern, if they're profitable they should live, if they are cost centers, drop them (obviously they're outside core mission of the browser, internet standards, internet freedom)
Mozilla XULRunner, which was the thing that happened before Electron, wasn't too bad though. It had way more traction than Positron. It died because Mozilla didn't care about it.
By that point Electron had gained too much momentum and public mindshare to compete with without a big marketing push and flagship project at least as significant as Atom and VSCode were. Positron never got either.
Electron apps were popular, not Electron itself. Even now there are only 4-5 dozen popular apps using it, while I probably run across that number of Qt/GTK apps a month. It's a sort of weird dichotomy it finds itself it.
I’m not saying they shouldn’t use it, of course! I just want there to be an alternative, too :-)
(It could be a part of KDE – they do have multiple media players, why not multiple browsers as well? But it could also be an independent project. I might tinker with this idea sometime, although I don’t have a KDE setup right now as my main laptop broke down.)
> While I applaud competition among browsers, I think that the renderer/webengine is such a big external component nowadays (especially if you lump in the JavaScript engine) that it might be more accurate to say that you're providing a skin, rather than a new browser.
Makes me think of NeoPlanet. That was an online trend that ripped through my university[0]. NeoPlanet was billed as the "world's first skinnable browser", and Wikipedia attributes the coining of the term "skin" (in the UI sense) to NeoPlanet's developers[1]. But NeoPlanet itself was pretty much a skin over Internet Explorer's browser engine, embedding a COM component of the browser into itself. So to me at the time, it didn't count as a separate browser, just a UI for internet explorer.
[0] Alongside AllAdvantage, which inspired some of the first mouse-jiggling simulation software because it paid users for active internet time while displaying ads on their screen. There was a whole universe of 90s Windows crudware that I wasn't all that privy to.
[1] I think Winamp's use of "skin" preceded NeoPlanet's, by analogy with Quake "skins", textures that could be applied to your player character in the QuakeWorld online multiplayer service, which could be used to indicate clan affiliation, etc.
I feel so trapped in this current computing paradigm where we're all running Chromium and Firefox is bad. I don't think it's substantially better than the days when sites were built to run in IE, since the solution to incompatibility has apparently involved moving the "browser" is really just a VM now.
On mobile it's even worse, with computing largely being removed entirely, replaced by "apps" that just deliver "content" and monetize their control over the algorithm. Or even worse, don't monetize it, but leverage it as a source of power. Zuck and Elon openly say govs were doing that with their platforms, but they hardly seem reliable. Maybe it's much much worse, maybe they exaggerated.
Urbit is more fun in theory but the community is just a bunch of rw chuds trying to get thiel bucks. Maybe computers just aren't that interesting now. At least, you have to be more creative than before. I've been doing some fun stuff. Anyway, /rant
FWIW, I'm genuinely unsure where the "Firefox is bad" narrative comes from. I use Firefox as my primary browser across Android, MacOS, Windows, and Linux (KDE) and have not had any significant trouble with that setup in years outside of an occasional hiccup with a Google property. Mozilla acts like an ADHD squirrel and can't seem to help but burn through their market share, but for all the terrible management and business execution the engineers at Firefox have been consistently improving the product.
Yeah, Firefox is quite good. It's fast and gets good battery life. It has the occasional bug, but so does Chromium, with any nontrivial usage. I wish Mozilla would focus on Firefox and on their messaging/what their users want (I don't want another chatbot, personally...). The Mozilla management is the only really bad thing about Firefox, but I'd take them over Google's any day.
Even the privacy preserving ads stuff is like... fine to me. I wish they would do other things, but I'm glad they're trying to make an internet monetization framework that's less awful for privacy. I don't have any inherent beef with ads, besides the surveillance apparatus around them. Whether they'll be able to stick to their core mission with the financial allure of ads lurking around the corner, though, I'm not sure.
It's strange, Firefox gets so much criticism because they don't pass various purity tests while Chrome is completely exempt because nobody expects it to not be evil.
Sure but people decide Mozilla is not meeting their standards and then switch to.... Chrome. (Or a browser based on chromium which is just a filtered version of Chrome).
the people complaining about firefox are not switching to chrome! they just keep complaining. don't know where you took that wildly inaccurate idea from.
people who don't complain about mozilla sorry state do move to chrome tho.
There is a certain type of poster that says they use Chrome because of X Mozilla thing, almost like it's a rationalization. Certainly not everyone complaining about Firefox of course!
My least favorite thing about Firefox is the "you must restart Firefox to keep browsing" message when my package manager updates it.
I removed the Pocket button from the toolbar with 2 clicks when I setup Firefox a couple months ago and haven't thought about it since. It takes me about 5 minutes to change the Firefox settings to my liking from a fresh install. Firefox sure isn't perfect by a long shot, but the constant haranguing makes me a bit sad
That doesn't actually work as a parsing of the sentence in English, and besides, they responded to be with a confirmation that they think Firefox is bad:
There is another web browser build on KDE libraries - Angelfish. While it is marketed as "webbrowser for mobile devices", it has a desktop interface too. This one could be more easy to get into and help with development, as it is quite recent and does not have a lot of legacy features.
Another issue with Firefox containers is there's no UI to manually add or remove sites from the container assignment list. Very annoying when so many sites and services use tracking/pass-through domains.
I think add-ons are an added complexity that needlessly increases the attack surface. I'd rather not use them, just like some people don't like to use Javascript. I have a bunch of bookmarklets that do everything I want. Something as essential as managing containers should imho be fixed at the browser level, not by an add-on. I don't like it when essential security features are bolted on top of software as an afterthought.
Do you have a git repo or something for the bookmarklets you use? I haven't used a single bookmarklet in like a decade so I'm curious what you still use them for.
google deal killed containers internally. politics explained it as "user will get confused" and we never got full isolation, just 4 sub profiles (and never even a final UI)
for the longest time there were only 4, named suspiciously exactly like gmail's categories... something like shopping, socials, work... i dont remember exact details.
It feels rather rude to the volunteer developers to accuse them of enshitification. Especially when it's not even true in this case as enshitification implies trading the user experience for shareholder value - there's no share holder value being extracted here. The developers made the decision based on its merits. Furthermore, as I remember KDE4 did still have Konqueror - though it became slowly less maintained throughout its lifecycle as the web browser bit was hard to maintain in the changing web climate and Dolphin was providing a good user experience for many users.
They traded user experience for developer egos. Ego is the equivalent of wealth in the OSS world, just as it is in academia. KDE 4 developers had some grand ideas to push on the world, such as the semantic web, regardless of whether or not their users cared for them.
IMO they're worse than macOS now. But yeah this is why gnome 3 is absolutely not for me and I'll never support it as Linux' single recommended desktop.
it's been almost twenty years since KDE 4 came out. it was terrible, yes. but Plasma 6, which was released this decade, kicks ass, so why are you so mad, Grandpa?
That was a bad time for KDE. They pushed 4 out with a lot of fanfare and it turned out that GUI wise it was a bad windows vista clone that broke every great established KDE 3.5 application.
Definite enshitification. They took something great and replaced it with a craptastic DE that didn't reach feature parity for a decade!
I bailed for XFCE at the time, never gone back to KDE, wouldn't trust them to not do the same thing again.
Konqueror has always been clunky as hell in my opinion. It was like an open source version of Internet Explorer except with tabs. I thought people would've realised mixing files and web pages in one single program was a mistake somewhere around 2000, but I guess it stuck for a while.
Ubuntu still has Konqueror in its repos, though. You can still download it if you want to go back to the early 2000s UI. I think they're using Chromium as a render engine now?
EDIT: nope, they use KHTML and KDEWebKit.
> Then the KDE 4 enshittification came and they had to have a separate file manager with half the features
That sounds rather entitled. As if they're doing it just to spite you, or because they need to maximise shareholder value. There are plenty of services that got enshittified because they saw chances to earn more money over their users' backs, but most people using that word just want to complain that the thing they got for free (especially open source stuff) or for a price that will obviously never cover the basic service costs (Youtube, all kinds of hosting, various streaming services, every AI product I know) now doesn't work like it used to, or tries to seek funding to continue existing.
Clearly the team decided it was better this way, and thanks to the power of open source, everyone could've disagreed and forked KDE 3.5 to stick to the old design.
Falkon (formerly QupZilla[5]) is a free and open-source web browser developed by KDE. It is built on the QtWebEngine,[6][7] which is a wrapper for the Chromium browser core.[8]
Seriously. Without cors, js on any site you visit can make requests as you to anywhere. The only thing protecting you would be security through obscurity.
I also want this, but only to run my own apps which I build for my private use. If you don't assume the user is incompetent, you could build a secure browser that provides permission prompts to opt out of cors.
It should be safer if it doesn't send cookies, but you can still get in trouble for "hacking" because you visited a website that contained code that sent "hacking" requests to another website.
Been using as my sole web browser on a daily basis since it was called Qupzilla, maybe for 10+ years as I switched to it from reKonq.
I really like it because it's fully and truly integrated with KDE, not needed a whole sort of patchs to integrate to it like Firefox needs to.
It's actually great. Not sure about the "Qt doesn't upgrade WebEngine often enough" in other comment (what is "ofteh enough"? I got several updates for it over the year) and of course it can be lagging from stuff from mainstream, but I think for 99% of users it's just fine.
Granted, you can't use Chrome/Firefox plugins, which may it seem not worthy to some people, but there's a basic adblock and greasemonkey extensions shipped with it with default which blocks most of stuff and even you can install a script to speed up youtube ads so that annoying ad will run out in a sec or less. Apparently you can write your own plugins for it but last time I wrote to one of its devs the api wasn't even documented.
There are some quirks on it, though, like the user agent thing - I set up as the latest chrome user agent for every website except accounts.google.com where I left it as the one shipped with Falkon so it lets me sign in, and yet it shows a warning about "upgrading" to another browser.
Ironically, such "warning" also shows up when browsing discuss.kde.org. Yes, the very KDE discussion board warns you against using KDE's own web browser.
Since some years ago I have a silly idea about a plugin that transforms tabs into some sort of Vim buffer list thing that can be filtered by the url bar, but am too incompetent about C++.
What does integrated with KDE mean? I use Firefox, but cannot imagine any type of integration that I really need.
With kdeconnect when I have some sound playing in firefox, e.g. a youtube video, I get audio controls directly from my phone. There's a million features, check it out here: https://kdeconnect.kde.org/
KDE has integration with Firefox/Chromium too, via plasma-browser-integration. You can search browser tabs and history via the KDE menus, and share URLs with KDE via the browser context menu.
https://invent.kde.org/plasma/plasma-browser-integration
I'd like to see the feature of preventing the OS from sleeping before downloading has been finished.
Native file pickers, widgets, settings system . Not sure if it also uses the kparts system for handling protocols
* The KDE system for network protocols ist called KIO. KParts is for user interface components. AFAIK, Falkon isn't using KIO because QtWebEngine has no extension points for that (and if it did, somebody would need to write HTTP2 and HTTP3 implementations or Qt wrappers for KIO).
Yeah Falkon is not using KIO. Same for Konqueror for network request. In the old time of QtWebkit and even older time of KHTML, the webview would reuse the Qt network stack, but now it's reusing the implementation from Chromium.
Btw KIO is now using QtNetwork for it's http implementation, instead of having it's own. So http2 is now also supported.
Aren't these just handled with XDG libraries?
Not really. You can use XDG portals to have native dialogues in Firefox, but GTK developers call this an abuse of portals, so this will go away at some point. Portals are intended as a way out of sandboxes like flatpack, not to plug-and-play file picker dialogues.
Wow, I'm not keeping up with all the new developments in the Linux desktop land, but that surely sounds like a step back.
So let's say I'm building an app like Blender or Reaper - I'm sidestepping the need for most of the OS-specific/native-widget components, because I already need to do a whole lot of very complex and custom rendering, and that is the saner choice when going for portability. But I would still like to maintain a certain level of basic OS integration, for example a native menu bar on macOS, matching the light/dark theme with the OS, or perhaps... a native file picker?
What are my choices on Linux? Link with Gtk, and make the app look out of place on KDE? Link with KDE, and pull in half of it with me when installed on Gnome? Link both? Summon Cthulhu?...
Sounds like we've had a solution for a moment, and now we want to remove it, because think of the yaks?
If you use Qt the framework does its best to make sure the native thing will be used where appropriate. QtCore, Gui and Widgets is like 15 megabytes total and that will already allow 90% of Qt apps to run. A bazillion packages being installed if you want to install one Qt app is just the fault of your distro's policies on how software should be distributed
> the fault of your distro's policies
Yeah, about that... fracturing at every possible level.
- Linux vs Free/Open/Net/Dragonfly BSD
- Linux distro 1 vs X vs A vs Ω vs ...
- glibc (with all its warts like versioned symbols) vs musl vs BSD libc
- systemd vs sysvinit vs rc vs OpenRC vs daemontools/s6/runit/...
- Who "owns" /etc/resolv.conf?
- apt vs (yum / dnf) vs pacman vs apk vs xbps vs emerge vs ...
- Flatpak vs Snap vs AppImage
- X11 (and libx11 vs xcb) vs Wayland (which protocols/extensions are supported?)
- OSS vs ALSA vs PulseAudio vs Pipewire vs sndiod vs ...
- Gtk (2/3/4) (with Gnome or without) vs Qt (with KDE or without) vs Tk vs direct X11 vs SDL/glfw/... vs an obscure toolkit last updated 15 years ago
inb4 it's about choice, the heck I'm supposed to choose as an app developer? inb4 "follow the standard", half of these are not standardised but still in widespread use? inb4 distro policies, which distro - top 10 on distrowatch looks like more work than macOS+Windows combined? Delegate to package maintainers, and my app is "fixed"/patched beyond me being able to debug/support? I choose what seems to work (for example, escaping via the XDG portal) and I'm getting rug-pulled?
This is exactly why we can't have nice things.
There is no Linux desktop.
There are competing, incompatible and incomplete solutions across distributions and software ecosystems.
Do you have a reference for this portal going away? I was really happy that work was being done in that area, and was optimistic that even Java apps might get sane file pickers in the future...sigh.
Download and media playback integrated with the taskbar, mostly
That works with FF as well.
Media through a standard MPRIS implementation, and downloads through.. I don't know what the system is called, but it works
> Granted, you can't use Chrome/Firefox plugins, which may it seem not worthy to some people
Why is it granted? WebExtensions is an open spec, it is possible to implement it in any browser if needed. (Orion does this, although I think they don’t cover 100% of the features yet.)
But an ad blocker and userscript support can take you long ways, yeah.
Granted, as in “conceding a point”. https://en.m.wiktionary.org/wiki/granted
and yet it shows a warning about "upgrading" to another browser.
Ironically, such "warning" also shows up when browsing discuss.kde.org. Yes, the very KDE discussion board warns you against using KDE's own web browser.
Such is the propaganda of Big G, that user-agent discrimination is actively encouraged to further Chrome's marketshare.
> Falkon is a KDE web browser using QtWebEngine rendering engine, previously known as QupZilla. It aims to be a lightweight web browser available through all major platforms. This project has been originally started only for educational purposes. But from its start, Falkon has grown into a feature-rich browser.
While I applaud competition among browsers, I think that the renderer/webengine is such a big external component nowadays (especially if you lump in the JavaScript engine) that it might be more accurate to say that you're providing a skin, rather than a new browser.
Anyway I miss Konqueror and am happy to see any of its descendants carry on its legacy.
(QtWebEngine is derived from Chromium - https://doc.qt.io/qt-6/qtwebengine-overview.html )
This is so funny.
So chromium came from WebKit, which in turn came from KHTML which was ... a KDE web engine :D
I’d really love to see a KDE-native browser based on Gecko instead of Chromium or WebKit.
Not likely unfortunately thanks to Mozilla, which in all its wisdom, dropped embedding support from Gecko. This means it effectively cannot be hosted outside of Firefox/Seamonkey and forks thereof. This is contrasted by WebKit, which is largely UI toolkit agnostic and can be dropped in just about anywhere.
Funny thing, Epiphany (now known as GNOME Web) was originally built around Gecko, but when Mozilla axed embedding, the project had no choice except to switch web engines and so it’s now built around WebKit instead.
Embedlite is still maintained, even though it's not a Mozilla project nor very active. Since it's the base for the Sailfish browser, it even does have Qt bindings.
Relevant repos:
https://github.com/sailfishos/qtmozembed https://github.com/sailfishos/embedlite-components https://github.com/sailfishos/gecko-dev
I think that was a fine decision by Mozilla, even if I'm not happy with it. Firefox is already a web browser with barely enough market share to support it. Wasting resources on making even smaller special purpose web browsers integrate with the engine is not a move for a company that has been struggling with a lack of independent funding for years now.
Of course the decision was made back in the Firefox 5 days, but back then Chrome was rapidly growing, mostly at the cost of Firefox. Had Firefox remained embeddable, I think it would've only been held back in important performance areas like threading and rendering because of the API around the web view.
Interesting enough, with the Firefox for Android rewrite, Mozilla actually went back to making Gecko embeddable: https://wiki.mozilla.org/Mobile/GeckoView This design decision is also claimed to be the reason for them to disable about:config in production builds, because misconfiguring geckoview can disconnect it from the app and leave users with a broken browser (apparently).
So now there is an official way of embedding Gecko in your application, but it's limited to Android only.
I still find the decision questionable. No embedding meant that Gecko missed out by default on the wave of embedded WebViews in the early smartphone era as well as the “desktop web stack” boom that Electron/CEF came to dominate.
Additionally, I believe that one of the reasons why Firefox has struggled to maintain market share is because it’s a bit of an acquired taste. Relative to Chrome, Safari, etc it’s kinda quirky and not to everybody’s liking. Alternative browsers wrapping Gecko could’ve helped a lot here.
Camino was a great example — Mozilla couldn’t justify building a Mac-specialized Firefox and that makes sense, so Camino filled that gap instead and became quite popular among Mac users. When Camino had to close up shop following the removal of embedding, most of those users didn’t switch to Firefox but instead to Chrome and Safari because those were better suited to their needs.
"Wasting resources on making even smaller special purpose web browsers integrate with the engine is not a move for a company that has been struggling with a lack of independent funding for years now."
So says the narrative. Mozilla has money, but prefer using it for C-Level salaries : Mitchell Baker raised her salary from 2.5M$ to 7M$ within 4 years, all the while laying-off developers and spending a shitload of money on seminaries and others BS expenses. Meanwhile, the market share of Firefox dropped.
And I assume you have noticed that privacy is not their main concern.
Such a pity because Phoenix/Firebird/Firefox brought a lot to the Web. But I guess this happens when Corporate people take over any organisation.
Yeah playing the whole corporate game really corrupted Mozilla. When you join them you start to think like them. And then it's game over.
It was US tax issues that forced them to go corporate but as a result of the Google deal. So they tricked them into going corporate.
>raised her salary from 2.5M$ to 7M$ within 4 years, all the while laying-off developers and spending a shitload of money on seminaries and others BS expenses.
Seminaries? Kagi turns up nothing. Could you explain?
Who do you think writes the Book of Mozilla? Making a web browser requires a large supply of priests.
Except for Internet Explorer.
Perhaps they meant seminars?
>Wasting resources
talking about Mozilla wasting resources is quite funny. Do you think having a competitor to Electron would be worse than
+ Pocket
+ LLM
+ VPN
+ https://mozillalifeboat.com/
any proof that they are losing money on VPN or Pocket? That's my only concern, if they're profitable they should live, if they are cost centers, drop them (obviously they're outside core mission of the browser, internet standards, internet freedom)
I think embedded gecko would have been the way they could answer electron, and that would have helped them stay relevant.
Mozilla Positron, their attempt at an Electron-compatible runtime on top of Gecko, did not pick up any traction and was cancelled.
Mozilla XULRunner, which was the thing that happened before Electron, wasn't too bad though. It had way more traction than Positron. It died because Mozilla didn't care about it.
Well, and XULRunner was based in XUL, something Mozilla was trying to get rid of for very good reasons.
That was quite late in the game, though. Kind of like closing the barn doors after the horses have already bolted.
It was released in 2016, only two years after Atom and Slack and a year after VSCode. Not sure I'd call that late.
By that point Electron had gained too much momentum and public mindshare to compete with without a big marketing push and flagship project at least as significant as Atom and VSCode were. Positron never got either.
Electron apps were popular, not Electron itself. Even now there are only 4-5 dozen popular apps using it, while I probably run across that number of Qt/GTK apps a month. It's a sort of weird dichotomy it finds itself it.
What about using GeckoView?
https://mozilla.github.io/geckoview/
I don't know what I'm talking about though. I just found that and have no idea how useful it is
It’s geared towatds Android mostly, though it might be possble to adapt it for desktop.
I mean, WekKit started life as a fork of KDE's web engine so only makes sense they'd use it...
I’m not saying they shouldn’t use it, of course! I just want there to be an alternative, too :-)
(It could be a part of KDE – they do have multiple media players, why not multiple browsers as well? But it could also be an independent project. I might tinker with this idea sometime, although I don’t have a KDE setup right now as my main laptop broke down.)
Qt doesn't upgrade WebEngine often enough. So it is possible to have lingering vulnerabilities in Falkon.
> While I applaud competition among browsers, I think that the renderer/webengine is such a big external component nowadays (especially if you lump in the JavaScript engine) that it might be more accurate to say that you're providing a skin, rather than a new browser.
Makes me think of NeoPlanet. That was an online trend that ripped through my university[0]. NeoPlanet was billed as the "world's first skinnable browser", and Wikipedia attributes the coining of the term "skin" (in the UI sense) to NeoPlanet's developers[1]. But NeoPlanet itself was pretty much a skin over Internet Explorer's browser engine, embedding a COM component of the browser into itself. So to me at the time, it didn't count as a separate browser, just a UI for internet explorer.
[0] Alongside AllAdvantage, which inspired some of the first mouse-jiggling simulation software because it paid users for active internet time while displaying ads on their screen. There was a whole universe of 90s Windows crudware that I wasn't all that privy to.
[1] I think Winamp's use of "skin" preceded NeoPlanet's, by analogy with Quake "skins", textures that could be applied to your player character in the QuakeWorld online multiplayer service, which could be used to indicate clan affiliation, etc.
I feel so trapped in this current computing paradigm where we're all running Chromium and Firefox is bad. I don't think it's substantially better than the days when sites were built to run in IE, since the solution to incompatibility has apparently involved moving the "browser" is really just a VM now.
On mobile it's even worse, with computing largely being removed entirely, replaced by "apps" that just deliver "content" and monetize their control over the algorithm. Or even worse, don't monetize it, but leverage it as a source of power. Zuck and Elon openly say govs were doing that with their platforms, but they hardly seem reliable. Maybe it's much much worse, maybe they exaggerated.
Urbit is more fun in theory but the community is just a bunch of rw chuds trying to get thiel bucks. Maybe computers just aren't that interesting now. At least, you have to be more creative than before. I've been doing some fun stuff. Anyway, /rant
> we're all running Chromium and Firefox is bad.
FWIW, I'm genuinely unsure where the "Firefox is bad" narrative comes from. I use Firefox as my primary browser across Android, MacOS, Windows, and Linux (KDE) and have not had any significant trouble with that setup in years outside of an occasional hiccup with a Google property. Mozilla acts like an ADHD squirrel and can't seem to help but burn through their market share, but for all the terrible management and business execution the engineers at Firefox have been consistently improving the product.
Yeah, Firefox is quite good. It's fast and gets good battery life. It has the occasional bug, but so does Chromium, with any nontrivial usage. I wish Mozilla would focus on Firefox and on their messaging/what their users want (I don't want another chatbot, personally...). The Mozilla management is the only really bad thing about Firefox, but I'd take them over Google's any day.
Even the privacy preserving ads stuff is like... fine to me. I wish they would do other things, but I'm glad they're trying to make an internet monetization framework that's less awful for privacy. I don't have any inherent beef with ads, besides the surveillance apparatus around them. Whether they'll be able to stick to their core mission with the financial allure of ads lurking around the corner, though, I'm not sure.
It's strange, Firefox gets so much criticism because they don't pass various purity tests while Chrome is completely exempt because nobody expects it to not be evil.
More like sanity tests.
https://madaidans-insecurities.github.io/firefox-chromium.ht...
It's not strange. Mozilla Foundation is a non profit and Google is not.
Sure but people decide Mozilla is not meeting their standards and then switch to.... Chrome. (Or a browser based on chromium which is just a filtered version of Chrome).
the people complaining about firefox are not switching to chrome! they just keep complaining. don't know where you took that wildly inaccurate idea from.
people who don't complain about mozilla sorry state do move to chrome tho.
There is a certain type of poster that says they use Chrome because of X Mozilla thing, almost like it's a rationalization. Certainly not everyone complaining about Firefox of course!
My least favorite thing about Firefox is the "you must restart Firefox to keep browsing" message when my package manager updates it.
>I'm genuinely unsure where the "Firefox is bad" narrative comes from
It comes from me, I hate it. The constant begging to use pocket or whatever the frugg its called is a perfect metaphor for the browser as a whole.
I removed the Pocket button from the toolbar with 2 clicks when I setup Firefox a couple months ago and haven't thought about it since. It takes me about 5 minutes to change the Firefox settings to my liking from a fresh install. Firefox sure isn't perfect by a long shot, but the constant haranguing makes me a bit sad
To be fair, I think this is a misinterpretation of what the parent was saying. I read it as: “we’re all running (Chromium + Firefox) is bad”.
That doesn't actually work as a parsing of the sentence in English, and besides, they responded to be with a confirmation that they think Firefox is bad:
https://news.ycombinator.com/item?id=43302877
There is another web browser build on KDE libraries - Angelfish. While it is marketed as "webbrowser for mobile devices", it has a desktop interface too. This one could be more easy to get into and help with development, as it is quite recent and does not have a lot of legacy features.
Falkon's fine, but it's important to remember that it's just Chromium with KDE/Qt UI on top.
yeah but Chromium is just the web engine from KDE with a bunch of things on top :P
but we dont want some of that "things on top" by Google.
It's things on top of other things all the way down.
what
https://en.wikipedia.org/wiki/KHTML
Can I open a KWord document in a tab though?
Those were the good old days.
Probably not... but why would you want to?
Presumably referring to Konqueror, the old KDE web and file browser which could embed all sorts of viewers: https://en.wikipedia.org/wiki/Konqueror#File_viewer
Does QtWebEngine have security support?
I took a look at Falkon, it's not for me. The autoscroll is not as smooth as the regular browsers and their adblock isn't as good either.
Question. What modern browser gets containers right?
(Firefox was almost there, but the containers don't sync over multiple devices.)
Another issue with Firefox containers is there's no UI to manually add or remove sites from the container assignment list. Very annoying when so many sites and services use tracking/pass-through domains.
> (Firefox was almost there, but the containers don't sync over multiple devices.)
I haven't really tried this, but will setting up your container config via an addon like https://addons.mozilla.org/en-US/firefox/addon/containerise/ and then syncing that addon's settings effectively solve this problem?
Sorry, but I don't trust add-ons. They sometimes change ownership, and I don't want to keep track of them.
This is also an area where browsers can differentiate.
If that's your threat model, the cited add-on is MIT licensed so you can generate an id of your own <https://github.com/kintesh/containerise/blob/3.9.0/src/manif...> and publish it for yourself, thus controlling its update and ownership lineage
I agree it would be better if one could just install the .xpi from a server you trust but Mozilla hates its users so here we are
I think add-ons are an added complexity that needlessly increases the attack surface. I'd rather not use them, just like some people don't like to use Javascript. I have a bunch of bookmarklets that do everything I want. Something as essential as managing containers should imho be fixed at the browser level, not by an add-on. I don't like it when essential security features are bolted on top of software as an afterthought.
Do you have a git repo or something for the bookmarklets you use? I haven't used a single bookmarklet in like a decade so I'm curious what you still use them for.
I use them to automate some stuff on niche websites I use, for pure convenience. I don't have them in a git repo, they all live in my bookmarks bar.
Note: it's quite easy to create them using chatgpt, nowadays (of course you have to check them before using).
I've heard that Zen has improved on them over Firefox, but haven't tried it yet, so I can't comment on it myself
google deal killed containers internally. politics explained it as "user will get confused" and we never got full isolation, just 4 sub profiles (and never even a final UI)
How do? I can make a lot more than 4 containers in Firefox
for the longest time there were only 4, named suspiciously exactly like gmail's categories... something like shopping, socials, work... i dont remember exact details.
I'd like that with a portion of vimperator please :)
I still remember fondly the KDE 3.5 times when we had Konqueror. The best browser AND file manager in one. It was amazing.
Then the KDE 4 enshittification came and they had to have a separate file manager with half the features. Bad times.
It feels rather rude to the volunteer developers to accuse them of enshitification. Especially when it's not even true in this case as enshitification implies trading the user experience for shareholder value - there's no share holder value being extracted here. The developers made the decision based on its merits. Furthermore, as I remember KDE4 did still have Konqueror - though it became slowly less maintained throughout its lifecycle as the web browser bit was hard to maintain in the changing web climate and Dolphin was providing a good user experience for many users.
They traded user experience for developer egos. Ego is the equivalent of wealth in the OSS world, just as it is in academia. KDE 4 developers had some grand ideas to push on the world, such as the semantic web, regardless of whether or not their users cared for them.
Yeah just like gnome 3 with its grand ideas that's now more barebones in terms of configurability than macOS.
to be fair, gnome3 unwritten moto was "we macos now", so, mission accomplished?
IMO they're worse than macOS now. But yeah this is why gnome 3 is absolutely not for me and I'll never support it as Linux' single recommended desktop.
it's been almost twenty years since KDE 4 came out. it was terrible, yes. but Plasma 6, which was released this decade, kicks ass, so why are you so mad, Grandpa?
because gramps had to suffer gnome or the tiling wm du jour while kde was fixing itself.
and if you think we were exaggerating in how bad our school route was, get ready for this one...
That was a bad time for KDE. They pushed 4 out with a lot of fanfare and it turned out that GUI wise it was a bad windows vista clone that broke every great established KDE 3.5 application.
Definite enshitification. They took something great and replaced it with a craptastic DE that didn't reach feature parity for a decade!
I bailed for XFCE at the time, never gone back to KDE, wouldn't trust them to not do the same thing again.
Konqueror has always been clunky as hell in my opinion. It was like an open source version of Internet Explorer except with tabs. I thought people would've realised mixing files and web pages in one single program was a mistake somewhere around 2000, but I guess it stuck for a while.
Ubuntu still has Konqueror in its repos, though. You can still download it if you want to go back to the early 2000s UI. I think they're using Chromium as a render engine now? EDIT: nope, they use KHTML and KDEWebKit.
> Then the KDE 4 enshittification came and they had to have a separate file manager with half the features
That sounds rather entitled. As if they're doing it just to spite you, or because they need to maximise shareholder value. There are plenty of services that got enshittified because they saw chances to earn more money over their users' backs, but most people using that word just want to complain that the thing they got for free (especially open source stuff) or for a price that will obviously never cover the basic service costs (Youtube, all kinds of hosting, various streaming services, every AI product I know) now doesn't work like it used to, or tries to seek funding to continue existing.
Clearly the team decided it was better this way, and thanks to the power of open source, everyone could've disagreed and forked KDE 3.5 to stick to the old design.
> thanks to the power of open source, everyone could've disagreed and forked KDE 3.5 to stick to the old design.
And thus was born the Trinity Desktop Environment.
> if you want to go back to the early 2000s UI.
Seamonkey's also a great choice for that.
From wikipedia, the missing important bit of information in falkon.org:
>It is built on the QtWebEngine, which is a wrapper for the Chromium browser core.
From Wiki
Falkon (formerly QupZilla[5]) is a free and open-source web browser developed by KDE. It is built on the QtWebEngine,[6][7] which is a wrapper for the Chromium browser core.[8]
Which Chromium browser core was itself developed from QtWebEngine
What I want: I want a browser without security restrictions, that can run node.js code as well as not CORS restricted fetch.
Wanted to code my own via electron, but damn, it was slow.
Because you want to get hacked?
Seriously. Without cors, js on any site you visit can make requests as you to anywhere. The only thing protecting you would be security through obscurity.
I also want this, but only to run my own apps which I build for my private use. If you don't assume the user is incompetent, you could build a secure browser that provides permission prompts to opt out of cors.
It should be safer if it doesn't send cookies, but you can still get in trouble for "hacking" because you visited a website that contained code that sent "hacking" requests to another website.
Not really. There is still same-origin-policy and CSRF tokens.
You can run chromium/chrome with many security features (eg cors) disabled. If that helps whatever security hell you wanna swim in...
[dead]