What's the issue with current browser engines?
What's the issue with current browser engines?
I've seen in various threads that the current browser engines aren't good, such as gecko and blink. The question is why? Why do we need a new one, and what's stopping a new one being made? Is it just the fact that they're a lot of work to make?
One of the main issues is the lack of competition. There are now only 3 main browser engines, Blink, Gecko and WebKit. Blink (which poses Chrome and Edge) is by far the largest, and has a the enormous marketing might of Google (and Microsoft to a lesser extent) behind it. WebKit runs Safari, which only runs on Apple platforms and arguably only has the market share it does is because Apple doesn’t allow other browser engines to run on iPhones and iPads. Gecko, the engine of Firefox, continues to slide into irrelevance (which pains me to say as a long time Firefox user).
We are in real danger of the web being trapped in a browser monoculture again, like the dark dark times of Internet Explorer’s dominance. This led to a period of stagnation in web technology Microsoft at the time put little effort into developing IE. Allowing Blink/Chrome to do the same will likely be just as damaging, albeit in different ways - particularly for privacy on the web.
For the good of the web no one company should ever be in the position to dictate web standards, which is why we need a healthy and competitive marketplace of web browsers and browser engines. The problem is that web standards have now become so complex developing an indecent browser engine is now a monumental task. Opera gave up on Presto, once the poster child for browser innovation. Microsoft, a company with far more resources, gave up on Trident. Mozilla was developing a new generation browser engine called Servo, but gave up on the project also.
people around me are starting to realize that firefox is the go-to browser nowadays. my dad has actually been using it since he bought his current computer. and i've switched back from opera gx due to concerns of me being in a walled garden of advertisement.
Plus Opera and the predatory loan thing doesn't help.
Servo seems that now is under the Linux foundation projects and it also recently being reactivated so its an active project.
The big issue is that by adding more and more features, a browser has become an operating system and so complex that you can't hope to make a new one from scratch.
The last "new" browser engine (that wasn't built by a corporation) was KHTML which was
stolenharvested first by Apple for Webkit and subsequently by Google for Blink. KHTML then rotted without support.The most recent attempt was to build Servo in Rust. Mozilla "ran out of money" (they depend on Google for their existence), and it's already rotting.
mozilla dropped it but servo has been picked back up by the linux foundation. its under active development again.
Constantly moving targets don’t help. CSS, HTML, and JavaScript add new features way too quickly. Between supporting new shiny stuff and crusty old stuff there’s so much bloat to keep up with. It’s taking huge efforts to design and tune these rendering engines, so there are only a handful of efforts now backed by major players (Google, Apple, Microsoft, and Mozilla). Even Microsoft threw in the towel on their own engine and started using Google’s Chrome because it was cheaper and easier.
Theres nothing stopping a new engine from being made. People wanna say its too hard or whatever but there are budding projects people are working on from the ground up. See Servo, Ladybird and Flow for non-Chrome/Firefox browsers. It's absolutely possible no matter what these naysayers say.
While it is true that anyone could. A modern browser engine is complex AF and about on par with an OS.
I really want there to be more competition. I’m happy to see more people talking about how bad google is and switching back to FF
What are the biggest reasons for the complexity? What would we be giving up if browsers were simpler?
I remember back in the 90s when it was mostly text and hyperlinks (and animated gifs). Now, we have a lot of nice features of course, javascript and what not, but which of these features are the heaviest for browser complexity?
I want you to be right. Do any of these have production-ready browsers I can try? Nobody’s saying it’s impossible for someone to start a repo and push some initial commits. But is there even an MVP?
Ladybird got like at least 250k in funding last year by Shopify. The project refuses to release binaries so you'd have to build it yourself. So far the team has completed a JS engine from scratch and hit a 100/100 on the acid3 test.
Servo's earlier in development but its got some preliminary stuff. They have a few full time devs. The project is written in rust so its getting a lot of hobbyist support too. Their main project is getting CSS compatibility.
Flow is a closed-source browser by some company. Their browser is in beta right now but as evident by this paragraph I don't know much about its project.
Idk how to check commit numbers on mobile but I'd say all 3 are in the thousands minimum.
No one seems to answer the question « why do we need a new one? ». Can someone give his/her thoughts on this ? I’m curious. Are the current engine not good enough?
It's not technical and more political.
Google is a surveillance company with a huge impact on web standards thanks to its insane market share. Apple on its side uses its forced monopoly to prevent websites from competing with its app store. Gecko depends on Mozilla that depends on google for 90% of its revenue, and is today a good "look we're not a monopoly" excuse for google.
I don't think people want a new engine because the current ones don't work. They want a new engine because they want the web to be truly open.
Try resizing your browser window. Youll notice that the content reflows very slowly, and that youll often see giant ugly black bars. This is a pretty bad user experience for people who resize their windows often. That alone is not enough to warrant a rewrite of anything though, so thats why it hasn't been fixed yet.