I do not understand the urge to start from scratch instead of forking an existing, mature codebase. This is typically a rookie instinct, but they aren’t rookie so there’s perhaps an alternative motive of some sort.
Because there are only like 3 browser engines: Chrome’s Blink, Firefox’s Gecko and Apple‘s WebKit. And while they are all open source, KHTML, the last independent browser engine got discontinued last year and hasn’t been actively developed since 2016.
There’s need in the space for an unaffiliated engine. Google’s share is far too high for a healthy market (roughly 75%), WebKit never got big outside of Safari (although there are a few like Gnome Web, there’s no up to date WebKit based browser on Windows) and Gecko has its own problems (like lack of HEVC support).
So, in my book, this is exciting news. Sure it‘ll take a while to mature and it is up against software giants but it‘s something because Mozilla doesn’t seem to have a working strategy to fight against Google‘s monopoly and Apple doesn’t have to.
Also Gecko’s development is led by people thinking that it being usable outside of Firefox\Thunderbird is a bad thing. There was a time when Gnome’s browser was based on Gecko, not WebKit. And in general it’s influenced by bad practices.
SerenityOS is an amazing project, of course. To do so much work for something completely disconnected from the wider FOSS ecosystem, and with such results.
So it’s cool that they’ve decided to split off the browser as its own project.
That’s always struck me as odd, but I’m also very much an outsider looking in. A “gecko electron” does sound intriguing though.
Could they not add HEVC support? Or is there some technical limitation that meant starting from zero was a good idea?
HEVC is almost entirely down the the licensing. This section of the wikipedia page details it pretty well.
The tl;dr is that the LA group wanted to hike the fees significantly, and that combined with a fear of locking in led to the mozilla group not to support HEVC.
And it’s annoying at times. Some of my security cameras are HEVC only at full resolution, which means I cannot view them in Firefox.
They could, probably. My guess is, that it’s either a limitation of resources, the issue of licensing fees or Google‘s significant financial influence on Mozilla forcing them to make a worse browser than they potentially could. Similar to how Firefox does not support HDR (although, to my knowledge, there’s no licensing involved there).
The biggest problem most people have with Mozilla is said influence by Google, making them not truly independent.
They get most of their money from google for the “default search engine deal” make of that what you want. For me personally it doesn’t sound fully independent.
No webkit browser on Android either. If there was gnome web for Android id switch in a heartbeat
Does anyone know why there are barely any WebKit based browsers? WebKit is open source and at least Safari works really well. Is it hard to work with? Do people just hate Apple that much? Is there some limitation?
Because software monocultures are bad. The vast majority of browsers are Chromium based. Since Google de-facto decides what gets in Chromium, sooner or later the downstream forks are forced to adopt their changes. Manifest V3 is a great example of this. You can only backport for so long, especially when upstream is being adversarial to your changes. We need an unaffiliated engine that corrects the mistakes we made with KHTML/Webkit.
Why are open source software monocultures bad? The vast majority of non-Windows OSes are Linux based. Teams who don’t like certain decisions of the mainline Linux team maintain their forks with the needed changes.
Manifest V3 is a great example of this. You can only backport for so long, especially when upstream is being adversarial to your changes. We need an unaffiliated engine that corrects the mistakes we made with KHTML/Webkit.
And we could get a functional one today by forking Chromium and never accepting a single upstream patch thereafter. I find it really hard to believe that starting a browser engine from scratch would require less labor. This is why I’m looking for an alternative motive. Someone mentioned licensing.
Perhaps some folks just want to do more work to write a new browser engine. After all Linus did just that, instead of forking the BSD kernel.
There is currently no implementation of web standards that is under a more permissive license than LGPL or MPL. I think that is a gap worth filling and if I recall that is what Ladybird is doing.
I guess Chromium isn’t fully BSD. This could be the reason. Although I’d think reimplementing the non-BSD bits in Chromium would be less work than reimplementing all the bits, including the BSD ones.
Why is that a gap worth filling? There is no benefit to users as long as its free of a EULA they don’t have to care either way. For those wanting to produce open source software based on same they already have all the rights they could need. The only party clamoring for permissively licensed software are companies intending to close off the source and sell other people’s work.
I understand why they would want to do that I don’t understand why anyone would feel the need to work for free for something someone else closes off.
There are some cases where it’s just not possible to release the source code, even if you wanted to.
For example, if you’re developing a Nintendo switch game, you aren’t allowed to release any code that uses Nintendo’s sdk, so that means you also can’t use any copyleft libraries.
Maybe MPL-licensed libraries would be ok though. Idk, I’m not a lawyer.
I can’t understand how people can continue relying on chrome and derivatives like electron, CEF etc. and not see it as a problem.
It’s easy to understand when you think most comments are similar to yours and don’t provide any insight as to why this might be a problem.
Maybe you could update your post and share your knowledge and experience with others, so that there are less people in the world who don’t see the problem.
When trying to render a relatively simple page consisting few thousands of text lines in a table, any current browser will cause mouse cursor to lag for some time, then you’ll discover it consumes at least 2 GB ~ 4 GB of RAM. YouTube lags like I have 2 cores instead of 16. Any electron app is either clunky or too clunky, also either hungry or too hungry.
I’m sorry but I don’t have time to look up other cases.
Ladybird was born from SerenityOS, which is a hobbyist unix-like (or POSIX compliant?) OS that simply aimed to do things “from the ground up”. It just happened that they needed to make a browser, and the response was to make one from scratch.
From there it seemed to have brought a lot of attention organically to the point where it can stand on its own, but originally it was never intended to be a “third browser engine” from its inception.
