Before you come at me with stuff like Librewolf, Waterfox and IceCat; those don’t count. They are just tweaked Firefox distros with mostly basic low level changes. Not every Chromium browser is super unique either, but I feel like there are more differences between them then there are with Firefox distros. Why is that? Why there aren’t different browsers that use Firefox’s engines but provide a different UX?
To the list of Firefox-based projects, you can also add Tor.
Right, I forgot about Tor.
I’m not rlly well-versed in this subject but from what I can tell, I think it’s got something to do with Chrome dominating the browser market share coupled with development contributions for Chromium from big-name corporations. Big buck businesses, am I right? Given that, some websites have been designed with Chrome in mind rather than the rest of the webiverse itself, possibly making Chromium a more attractive browser to base a separate browser project on (please correct me if I’m wrong).
I’m doubtful; while I can see that being a factor on browsers like Edge and Vivaldi, I think plenty of FOSS enthusiasts would prefer to base their software on Firefox given their anti Google stance and monopoly concerns.
If that were the case, where FOSS enthusiasts would prefer to base their software on Firefox for reasons above, then we would’ve seen more projects use Firefox and less of Chromium by now (outside of the ones listed from the original post above), no?
That’s my point. There could be something stopping them from doing so, Firefox was quite popular at one point and even than I don’t think there were alternatives to mainline Firefox, I feel like there should be a reason for this.
I am not so sure if FOSS enthusiasm plays in here…privacy concerns and “dislike of big companies” might but chrome/blink is a fork of apple’s webkit which is a fork of KDE’s khtml/kjs…under LGPL. If you look at it chrome is a pretty good example of FOSS in action. If khtml hadn’t be LGPL in the first place I have my doubts apple would have made webkit public as they did…and am also not convinced that google had done the same for blink. (But has to be said that webkit adds BSD licensed parts that are not directly based on khtml…that might be a concern for FOSS enthusiasts)
Above would be the primary reason. From a business strategy perspective website compatibility is better with Chromium than Firefox. While Firefox is fast now, there was a period in the mid 2010’s when Firefox was trying to diversify its income streams that resulted in it taking it’s eye off the ball and the browser experience got slower and suffered. We’re still seeing this effect even though they have improved the browser significantly over the past couple years.
deleted by creator
It would be nice if Mozilla could make a project similar to Chromium Embedded Framework, in order to integrate the Gecko engine in apps (because currently Blink and V8 are used and the truth is that the performance is not very good) and even in game consoles Gecko could get a good market share. Servo was an step in the right direction.
For Android there is https://mozilla.github.io/geckoview/. I don’t know if there is anything similar on desktop.
There is still an effort going on in the Sailfish OS web browser to separate the UI from the core browser. But they’re still about 18 versions behind the current Firefox version. The README of the project contains some link to useful resources about how it works.
Embedding chromium/webkit is just way easier. I honestly wonder why SailfishOS went that route. Maybe it’s a continuation of some legacy project form Maemo/MeeGo?
I wonder if this is significantly different with Servo. Servo isn’t really ready for as far as I can tell, but in a couple years it might be. (If there’s some kind of significant volunteer effort, given Mozilla fired everyone working on it.)
I’ve already seen a company announcing a product that sounded a lot like Servo. But it was targeted for embedded use, so something where a limited number of supported webpages may be perfectly fine.
Actually implementing all of the web standards into Servo, that’s just not going to happen. Mozilla and Google already struggle to keep up with web standards in Firefox and Chrome, and those have more than two decades of a headstart.
That’s also why Mozilla ran Servo as a research project. They used it to explore a modern architecture, much of which they were able to retrofit into Firefox, but they didn’t work much on web standards.
And yeah, when finances run thin, research projects are among the first things to go.Servo is now funded by Linux Foundation.
That was my guess too, apparently I was right. Thank you for taking the time to answer :)
deleted by creator
This is probably not an answer to your question but apple faced such a decision 20 years ago already…and picked khtml/kjs over the firefox engine gecko back then because:
When we were evaluating technologies over a year ago, KHTML and KJS stood out. Not only were they the basis of an excellent modern and standards compliant web browser, they were also less than 140,000 lines of code. The size of your code and ease of development within that code made it a better choice for us than other open source projects.
…
…How did we do it? As you know, KJS is very portable and independent…
( https://marc.info/?l=kfm-devel&m=104197092318639&w=2 ) (Sorry, it’s a long time ago, best post I could still find from that time…but why they picked khtml over gecko was discussed extensively back then)
Of course I have no real clue (and firefox changed the engine in the meantime) but a project based on a portable library in the first place then forked into chrome/blink might have something to do with people finding it easier to integrate in their solutions than an engine specifically written for a single browser.
Yeah, I have heard before that one reason is Chromium being more modular.
I also assume Google put somewhat of a focus on that, after Node.js chose Chromium as a base. If I remember correctly, it was initially not possible to run Node.js in a headless mode, which it is now.
Not. Google put effort in making WebKit less modular in blink. In WebKit you can chose the renderer, while in blink you ate stuck with skia.
Blink is a black box widget you van interact only via we extensions API. WebKit is a set of libs you can meddle with.
For instance, there is no way to render a web page to a PNG file in blink, while it’s trivial with WebKit.
EDIT Alas, the first chromium was WebKit with the js engine replaced by V8. Not sure if you can chance blink js engine.
So the Chromium is more modular, I always guessed that it was but not certain. Thank you for your answer
These days, yeah, I’m pretty sure that it is. One of the stated goals of Servo was that it should have a well-defined API to allow embedding into Node.js and similar.
But a few years ago, when people started embedding browsers into everything, that difference might’ve been less big between Firefox and Chromium.
So, maybe there was another motivation, for example Chromium’s JavaScript engine had much better performance around that time, which might’ve been rather important to Node.js.
So the portability is a factor, understood. Thank you for takig the time to answer.
I would say so…but please don’t take my word for it. It’s really not like I have a clue. It’s just what I gathered.
IDK about projects but I do know why I abandoned Firefox a month ago, after using it for a very long time - since about 2007.
- A year or two ago Mozilla waged a war against adblockers and added a blacklist of extensions they don’t like (naturally the adlockers were the first to go) and the browser started removing them without asking for permission first. It didn’t take long for me to “catch the trails” of that function and to silence it for good. IDK if they’re still waging that war but I have no intention to check - the roadblocks I’ve setup on my computer, my router and in the browser itself are still in place.
- Switching to web extensions instead of legacy. Web extensions provide very limited functionality compared to legacy extensions, thus the customization level dropped seriously.
- Removing the option to disable updates of the browser. This was the reason I started referring to Mozilla as Microzilla bc removing the option in question was a typical Micro$hit move.
- Removing the good old statusbar that so many of us had gotten used to it, getting the browser closer and closer to look like Google Chrome.
- Disabled the feedback function, so that the user can’t send bug reports or suggestions. This says “We don’t care about the users, we’ll just do whatever the hell we want, just like Bellamy Blake”.
- Did something to the function layers.acceleration-forced that caused the browser to lose its interface after waking the computer up from sleeping. That option was the ONLY thing in Linux that worked like VSync in games - prevented screen tearing when scrolling up and down in pages. I had it reported to them about 18 months ago or so - still hasn’t been fixed and I doubt they will bc they don’t care about the user experience anymore. There is a substitute in Linux through the nVidia driver (force full composition pipeline) but it doesn’t work with any browser, including Firefox, it only works outside the browser.
- The anti-tracking function they recently introduced is becoming more of a pain in the side than a useful thing. Because of that anti-tracking, a few websites stopped letting me in to my profile - mediafire, gamivo and a few other websites for buying games, even kinguin sometimes refuses to me let me in.
- Removed the right click on image function “View image details” from the context menu. This was the final drop for me. After putting up with their crap for all these years, that was too much for me. If I was gonna use a browser that looks like Chrome, I’d rather switch to that browser. So nowadays I’m using Chromium. At least Chromium has an extension to replace the missing function I just mentioned. Firefox doesn’t have an extension for that function…
No doubt dropping XUL was a terrible move. WebExtensions are just userscripts but worse in almost every way. I never heard about the other issues however, can you link sources for those?
I think dropping XUL was a great move. It was old legacy and expensive to maintain. Mozilla already has limited resources, I don’t think maintaining a redundant and complex technology is a great idea. I say burn XUL.
But I do agree that they should have kept legacy extensions. There are a few major arguments against this:
- The depend on browser internals, so limit what Mozilla can do without breaking extensions. (Like remove XUL).
- The run with full access to the entire browser.
As to 1 I say that is fine. I think the better approach would have been recommending WebExtensions and providing a stability guarantee, but allow legacy extensions and break compatibility with impunity. Yes, it sucks for extensions authors (I speak as a past extension author) but for the minority of extensions that need this level of integration it seems worth it. And it is arguably a benefit to make the WebExtension route more attractive to developers in the cases where it can be used.
For 2 I call this a feature. There are already WebExtenions privileges that are basically full access so I don’t see why legacy extensions couldn’t display to the user some “fake” permissions such as “Full access to my browser and all sites” and allow it. I don’t care if they label it bright red in the store, or even hide it from search results, but please make it possible.
I know because in the slow tightening of restrictions I maintained VimFx for a while. So sure, it was a pain fixing the incompatibilities, but it was manageable. Furthermore in that case there were very few users because you had to jump though many hoops to use the extension even after it was updated. So if this extension type was supported there would be more people helping out, lightening the load.
I’ll try and see what I can find bc the most of these things were done in a long periods of time and I’m not sure what is still findable. The “View image details” in particular you can see for yourself that it’s missing. Simply open any image (except from Instagram - Firefox doesn’t detect these correctly), right click it and you’ll se there’s no such an option. There’s only “Inspect” from which you can get the URL but there’s no more the additional window where you can see image dimensions, size or just click “save as…”.
The disappearing interface bug that was never fixed: https://bugzilla.mozilla.org/show_bug.cgi?id=1536396
The anti-tracking function that causes some websites not to let you in - try logging into your mediafire account. It doesn’t always happen but when it does, it will say “this form was idle for too long” and that’s it. Eventually you’ll have to change the browser in order to login.
The disabled function - you can see that one for yourself as well. Simply open the menu, click “Help” and then “Feedback”. You will see this: https://i.imgur.com/NjvwTjm.png It’s like that for about a year, probably longer.
The other removed features - such as status bar and disabling updates are clearly visible by anyone, so I’m skipping these.
As for the war against adblockers (mostly), here’s an example of my about:config (I still have Firefox but it’s not my default browser anymore): https://i.imgur.com/UEIaAcf.png This was my way to disable addons removal - setting url addresses as 127.0.0.1 bc Microzilla had blocked many addons (adblockers included) for unknown reasons and when that was implemented with a browser update, I turned out with a browser stripped off of half of its addons. Even an addon I’ve been using for years (previously known as Speed Dial, now known as GroupSpeedDial) was removed and marked as “dangerous” (a complete bullsh*t, ofc, there’s nothing dangerous about it).
