The text in the bubbles may be too small to read on mobile devices, which is why I've added the same text at the bottom of the image. I'm aware of the issue, so I can remove the text for desktop browsers.
That's a good point. Sometimes LLM generates incorrect GIT messages. I can add the whole source file including diff to the prompt, but it increases the token size. So I will find a good way to add more context to the prompt.
I and many others agree. But the pesky problem of needing to pay the people who want to make a difference sadly still persists.
I'd work on idealistic causes starting tomorrow if I didn't live in a world that mandates me to spend most of my energy on commercial work just so I can eat and have where to sleep.
Oh, the idea of my comment is to put it clearly and unambiguously that the problem is open, and the most commonly discussed "solutions" are complete bullshit, so we need to find a way to solve it.
And we won't find any solution within the main view of letting the market solve it. The market doesn't solve this kind of problem.
I never understood why we don't have more indie browsers. Individual developers love implementing their own take on anything: from the most simple projects to crazy ones like booting Linux from JavaScript or entire operating systems. Why are those talented individuals not implementing their own browsers?
I thought Ekioh Flow was promising because it’s multithreaded. Believe it or not, Blinks KHTML’s roots are single threaded, so that’s apparently a big deal. Then I read their blog post where they proudly claimed to finally get Google Docs to work….yikes.
They are a small team of I think 20 people, but it goes to show how complex and exhausting writing a modern browser truly is.
In terms of talented individuals, I actually emailed Fabrice Bellard on the topic because he wrote QuickJS along with everything else he’s known for. Even he admits writing a browser is a very complicated task because a lot of information is not correctly specified. He did however write QuickEmacs 20 years ago which supported html and css rendering.
It’s sad, even when W3C and other committees try their best to keep the web strictly standard, it truly isn’t. And like I said, it hasn’t been enough of an issue to raise any concern outside of bubbles like here.
I think the masses need to learn about the importance of keeping fundamental software free and standard as a whole before they can learn why browsers and OS’s should be. But it’s tough because the average user doesn’t know what a library is. A lot of people think that even in app development, the programmer does it all from scratch.
> I never understood why we don't have more indie browsers. Individual developers love implementing their own take on anything: from the most simple projects to crazy ones like booting Linux from JavaScript or entire operating systems. Why are those talented individuals not implementing their own browsers?
My understanding is it's been a long time since a browser was a cool little thing a single person could hack up in a reasonable amount of time. Now web standards are such that it's like some massive line-of-business app with a million requirements that all have to be implemented (and be fast, and be memory efficient) or something will break/someone will complain.
A browser these days is a mini-OS running inside an OS. A browser that tried to be just an HTML/JS rendering engine with a basic UI would fail just like FF is failing, despite getting glowing reviews from HN/reddit.
Because a browser is very complex and does many things simultaneously. Not going to list them here. We all know the many things a browser has to do right. It’s a bundling of many separate components.
Well, there is the web browser engine that the SerenityOS folks are making for their OS and IIRC their goal for it is to be advanced enough to be able to visit their Discord using it.
I mean I love serenity OS and its community. A perfect mix of passion and talent. But let's get real here, implementing a usable browser implementing all the specs at half the speed of chrome is an undertaking that would requires a huge amount of work.
Don't get me wrong I would love for this happen, just don't count on it.
And please, when SerenityBrowser have 80% market share in 10 years, don't hesitate to parade that comment all over HN or whatever cool website we'll be on.
You might already know though, a lot of chrome's Blink is enrooted in the KHTML project. So I at least feel optimistic that open source has a chance to fight back, I just don't know how or when....
I don't know why people are downvoting you you're right. The firefox engine has been struggling to keep up with chrome. We'd just be losing a worse browser engine which, while sad, isn't really that terrible. If we need a new engine someone can fork Chrome or write a new engine from scratch if chrome is a lost cause. The firefox foundation has stopped innovating and it isn't worth trying to keep them alive for a worse browser
Firefox is a great browser engine and has a hell of a lot less bugs than Chrome.
That's not the problem. The problem is that Firefox is funded by their direct competitor, who incidentally also is a monopoly with a bottomless marking budget.
