__  __      _        _____ _ _ _
|  \/  | ___| |_ __ _|  ___(_) | |_ ___ _ __
| |\/| |/ _ \ __/ _` | |_  | | | __/ _ \ '__|
| |  | |  __/ || (_| |  _| | | | ||  __/ |
|_|  |_|\___|\__\__,_|_|   |_|_|\__\___|_|
community weblog

A browser is born

Ladybird: A new cross-platform browser project - "The Ladybird browser came to life on July 4th [2022]"* and now today Announcing the Ladybird Browser Initiative via Simon Willison: "Andreas Kling's Ladybird is a really exciting project: a from-scratch implementation of a web browser, initially built as part of the Serenity OS project, which aims to provide a completely independent, open source and fully standards compliant browser." (previously)
posted by kliuless on Jul 01, 2024 at 11:10 PM

---------------------------

I hope they succeed. Am not feeling impressed with the ability of Mozilla to build Firefox into a genuine competitor to Chrome (even though I love FF and am using it to post right now).

Had a go and managed to build Ladybird from source a couple of weeks ago. Did great loading the wikipedia home page but was unusable crashy as I tried various other sites.

Interesting to see that Shopify are the biggest sponsor. Perhaps they see the value in not letting Google/Alphabet totally control the platform that is the only conduit to Shopify.
posted by i_am_joe's_spleen at 2:54 AM

---------------------------

As a Firefox user it seems that browsers are like mobile operating systems - there's only room for two big alternatives and the rest will sadly forever be niche.
posted by GallonOfAlan at 3:25 AM

---------------------------

Not sure that's true GallonOfAlan, that seems to be more the result of the way the market has shaken out. Opera hung around for quite some time, and other browsers, using WebKit, proliferate, usably, even now: my go-to example of Vivaldi.

I just found out about Ladybird a couple of weeks ago, and I consider that the best way to think of it is not as an alternative browser, because we actually do have a lot of those. In addition to Vivaldi there's Safari and Edge (and Brave for cryptobros). The problem is that they all use WebKit, in one form or another, for their engine. But the only alternative to Webkit among full-featured web browsers at the moment is Firefox's Gecko. This aims to change that.

Mind you, part of the issue with all these browsers is features being added to WebKit. Just yesterday there was a post about a browser-based After Effects workalike that only works on Chrome-family browsers because they implement AudioData, showSaveFilePicker and VideoEncoder, and Firefox doesn't. It starts to feel like the old days when Microsoft would put things into IE so everyone would prefer it to Navigator. I wonder if Ladybird will eventually chase after feature parity there?
posted by JHarris at 3:47 AM

---------------------------

Worth noting at this point, developer Andreas Kling is maybe not the most thoughtful person when it comes to gender...
posted by bigendian at 4:29 AM

---------------------------

In the list of operating systems they have "Windows (WSL) ". So, not really *Windows* but kinda-sorta Windows, if you have Windows Subsystem Linux installed (and presumably up-to-date).

Also saw no mention of running browser extensions, which for people like me are essential.

I wish them luck, and perhaps in time it will be more widely usable, but I think bootstrapping into that position is hard without real Windows support.
posted by Ayn Marx at 4:52 AM

---------------------------

The only long term path to being fully standards (whether or not they are "official") compliant is to take a dependency on chromium. Everything else is a dead end. I use Edge every day on all my devices and it is no different from chrome apart from not sending my browsing data to Google.
posted by simra at 5:43 AM

---------------------------

After reading on MeTa all the pros and cons of rewriting from scratch, this sounds particularly interesting (and potentially futile). But the need for interoperability, of tech in general but especially on the web, seems to doom this from the start.

I remember stringing cat5 around the office in 1997 or so, trying to get file sharing working between various macs and pcs and letting everyone share a connection to an ISDN line. I'd be willing to pay for a web browser that didn't do nefarious things, but it's hard to compete against free.
posted by rikschell at 6:47 AM

---------------------------

The only long term path to being full standards compliant is to take a dependency on chromium.

That's the point of Ladybird, to change that, whether they succeed or fail. I'm a Firefox/Firefox fork daily driver and I have never come across a website that doesn't work unless it has specifically chosen not to work, so it is possible. Though as bigendian pointed out, the developer is weird so I'm not optimistic about it.
posted by lianove3 at 6:47 AM

---------------------------

Gender neutral language within build instructions #6814

CONTRIBUTOR: It's a minor nitpick, but I think it's important; assuming the user and/or developer of the operating system is male isn't exactly the best.

ANDREAS KLING: This project is not an appropriate arena to advertise your personal politics. *immediately closes issue*

Oh fuck that. Gender neutral language is the minimal-effort baseline of professionalism and if Kling's immediate response is the unforced error of pulling the reactionary-politics-unconvincingly-veiled-as-apolitical card, we all know the community he'll eventually create is going to be full of assholes that goes nowhere because it drives away all the non-assholes.

Unless this project gets millions in funding from billionaires, he's chosen to die on the hill of "will only use male pronouns in the build instructions."
posted by AlSweigart at 6:51 AM

---------------------------

fuck this guy, fuck his project
posted by seanmpuckett at 6:53 AM

---------------------------

Things You Should Never Do, Part I from Joel Spolsky's Joel on Software blog:

There's a subtle reason that programmers always want to throw away the code and start over. The reason is that they think the old code is a mess. And here is the interesting observation: they are probably wrong.

Given that the blog post is about the Netscape browser rewrite leading to it losing the browser wars, it's worth a read.
posted by AlSweigart at 6:55 AM

---------------------------

Gender neutral language within build instructions #6814

My man submitted a patch to use inclusive language and the maintainer, rather than click a single button to accept it, choose to respond with "This project is not an appropriate arena to advertise your personal politics."

The maintainer has spent extra effort to reject a signal that female contributors are welcome. That itself is a strong signal that women are in fact unwelcome. And he has the chuzpah to claim that it isn't the place for personal politics.
posted by cotterpin at 7:26 AM

---------------------------

> Andreas Kling is maybe not the most thoughtful person when it comes to gender...

i'd suggest hen: "Much like the rise of the singular 'they' as an inclusive pronoun in English, Swedish has in recent years seen a veritable explosion in the use of the gender-neutral pronoun hen. The word was inspired by the Finnish pronoun hän, which refers to anyone at all, because Finnish is a genderless language... Like most languages with a gender system in nouns or pronouns, Swedish also has to deal with the issue of masculinity as the epicene gender (the gender that is meant to be inclusive and speak for everyone)"
posted by kliuless at 7:43 AM

---------------------------

Things You Should Never Do, Part I from Joel Spolsky's Joel on Software blog:

Ladybird isn't a rewrite, though, it's just a... write. And even if it were, constantly linking the same cautionary tale is becoming less and persuasive.

The core idea of a standards-compliant browser built completely independently from Chrome is good and necessary.

My man submitted a patch to use inclusive language and the maintainer, rather than click a single button to accept it, choose to respond with "This project is not an appropriate arena to advertise your personal politics."

Another such patch was submitted by a different author only to met with grumbles by the maintainers. Those potential contributors did not submit anything after that.

A new browser will be a community project; it would be good for the authors to recognize that and build that in from the beginning. I have not interacted with the Servo project, but it may be a more promising start in that regard.
posted by a faded photo of their beloved at 7:49 AM

---------------------------

And even if it were, constantly linking the same cautionary tale is becoming less and persuasive.

"Nobody goes there anymore; it's too crowded."

We all repeat "don't roll your own crypto" a lot, too. Some things have enduring popularity for a reason.
posted by AlSweigart at 8:12 AM

---------------------------

At the risk of a derail, "don't roll your own crypto" and "Netscape's rewrite went poorly" are very different. One is a guideline and one is an anecdote that suggests a guideline.

If you ask someone left-of-center why they think Trump is a liar, they can pull up (by this point) literally more than a thousand examples to illustrate their point. If you ask someone right-of-center why they think Obama is a liar, it's very likely that they'll give you just one- the same one every time: "you can keep your [health insurance] plan". They don't seem to notice that they always use that one example.

If a guideline is appropriate, one might expect there to be cases in which the guideline was not applied and bad results followed. For the guideline "don't roll your own crypto", there indeed have been many attempts to act contra the guideline that resulted in pwnage. For the guideline "don't attempt a rewrite", there probably *have* been many attempts that theoretically could be used to support this notion! However, the internet has settled on just this one, and for whatever reason it has no self-awareness of how much that weakens its case.

I'm not even saying that rewrites are good! I'm saying that if we settle on a conventional wisdom around rewrites, we should try to be more systematic about it than relying on one good writer's telling of one compelling anecdote. It's been 24 years since that essay was written. It's time for a follow-up chronicling more failed rewrites, or ideally a more systematic comparison of attempts at evolutionary change versus rewrites.
posted by a faded photo of their beloved at 9:41 AM

---------------------------

I've been increasingly concerned about Mozilla. jwz is vociferously against it, and I couldn't see why, even with all the flubs, but the more I see things happening, the more Mozilla acts like a corp and continually sucks at the teat of Google, and just follows Google's lead we have a real problem.

With that in mind I GMOFB (got my own fucking blog) and wrote a really long thing here about this.

The main takeaway:
1) We need alternative renderers, no we cannot just rely on "everyone being chromium" (or even webkit) this is a bad attitude to take. Monoculture is bad, it is letting one giant corporation with vested interests in one thing ("the market and how it can make money") to determine a lot of things, and Google has shown they are not friendly to open standards, and have to be bullied into relenting on the most egregious sins they've tried pushign on us.

2) We need a REAL standards body with teeth, not the corporate schlockfest known as W3C (see Mozilla's capitulation to DRM and now AI (though not part of the standards (yet, I think), you can be assured the 2 big boiz will push it in soon enough)

3) The standards body needs to make a rendering core that is, at the very least, an MVP, and even better, a fully functioning product that can be integrated free of charge.

The only other option right now in terms of rendering (and it's not nearly a mature product on its own) is Servo (parts of which are used in FF, but the complete renderer isn't complete yet, and won't be replacing gecko).

I was looking forward to Ladybird, and had heard of Kling through Serenity OS, but thanks to the post upthread I think I'm tapping out of that. I am so sick and tired of reactionaries playing the "I'm not gonna play politics" politics. And the toxic environment it leads to (see "hyprland" for something similar in Linux Wayland side of things).

No matter how many alternative browsers I've tried (I didn't try Opera, cuz my understanding is they're now Chinese owned, no longer Norwegian, and frankly, it doesn't have its own renderer so there's no point).... It seems the gecko/FF styled browsers are still the ones I "Kling" to. Basilisk and Pale Moon use a fork of Gecko called Goanna. I wanted to like Vivaldi (even though it's Blink based, IIRC) but after how many years they STILL don't have a proper OPML import. Sorry if you can't even get that going right you're not worth my time.

The hard part is that browsers are so complex nowadays, pulling a project off like this is hard work. I will say, I think making Servo in Rust makes it harder to get contributors (I know it's a sexy language for people to dive into, but it still seems like something that's a bit tough to work in for a lot of non-pro devs).

We need more than just skins on one rendering engine. I don't think browser wars are good, so I understand WHY people like a single engine to simplify devs life (nobody wants early 00s wars).

Not sure there's anything I said in my link that is more than the above, but the link is longer, and maybe you'll find whatever random shit I said there interesting/pertinent. Either way, this is my general stance I'll die on this hill.
posted by symbioid at 9:48 AM

---------------------------

In other milkshake duck dog whistles, awesomekling likes the post-Musk twitter:

Some folks have asked why I'm not active on Mastodon.
Simple answer: I like it here. X is full of positive energy and "let's build!"
Meanwhile, Mastodon is full of negative energy and "stuff sucks and it's someone else's fault"
To protect my own mental health, I choose to hang out in predominantly optimistic social spaces.
posted by autopilot at 9:49 AM

---------------------------

I'd previously been interested in the SerenityOS project, but that PR comment is really disappointing. I'll still give it a shot and explore avenues for shifting the project culture, but if it's one of those "meritocratic" "open source" projects with a benevolent dictator who also happens to subscribe to the idiot reactionary politics of the median programmer, my hopes aren't high.
posted by jy4m at 9:54 AM

---------------------------

I don't want to derail either. Let me just say that if you're writing/rewriting software from scratch, the odds are against you. The programmer mantra is, "We do these things not because they are easy, but because we thought they would be easy."

Starting from scratch, even for big projects like a browser, isn't necessarily a bad idea or the wrong move. And a new browser engine would be a great thing (I share the concerns about Mozilla) but I don't think this is it. The hardest stuff in tech is never the tech, and if Kling's mentality is stuck in the online cesspool culture of the 90s (and 2000s (and 2010s)) then I think he has even less chance of success (and is more deserving of failure) than the crypto-scummy Brave browser.
posted by AlSweigart at 10:22 AM

---------------------------

We definitely need a non-corporate, privacy respecting browser. Sadly, I can't offer my support to someone who thinks considerations of basic inclusion are "political" - it takes so little effort to just use gender neutral terms in documentation, it's a no-brainer win for the project, and increasingly even corporations are using less shitty terms (denylist instead of blacklist, for example) so you can't even make the argument that this is someone's personal preference and not accepted best practice anymore. That Kling is choosing this hill to die on really doesn't speak well of him, and if he has poor judgement in this area, he probably does in other areas as well.
posted by signsofrain at 10:23 AM

---------------------------

Also: fuck Kling. In case it's not clear - I was only responding to the question posed about having an alt engine, not endorsing Ladybird as the engine to use.

After learning more about this - I obviously don't think this is the project to take over. I was actually subbed to his blog and a recent post was "I'm back after a mental health break". Shame he didn't learn while in the "break" mode and just seems to double down instead of grow and adapt.
posted by symbioid at 11:29 AM

---------------------------

It's not clear to me that we need multiple renderers (ask any web dev how many renderers they want to code for). One common renderer that isn't controlled by a single for-profit entity would be plenty.

This effort seems like the old XKCD n+1 standards joke but unlikely to even move the needle from n to n+1.
posted by simra at 12:23 PM

---------------------------

If the Milkshake Duck weren't obvious enough already, here's the bottom of the project's CONTRIBUTING page:

On ideologically motivated changes

This is a purely technical project. As such, it is not an appropriate arena to advertise your personal politics or religious beliefs. Any changes that appear ideologically motivated will be rejected.


(via this post).

I don't understand why you're gonna put "lady" in the name of your project if you don't want any of them to feel welcome contributing.
posted by fedward at 1:15 PM

---------------------------

AlSweigart > Unless this project gets millions in funding from billionaires, he's chosen to die on the hill of "will only use male pronouns in the build instructions."

Well he got his first million.
posted by egypturnash at 1:34 PM

---------------------------

Calling the browser Ladybird and then refusing to use gender neutral language in your docs is... uhh, an interesting strategy.

Although the monoculture of chrome-based browsers does seem like a bad thing, I'm not so sure how bad it is, really. So long as chromium/webkit is open source, forks can be made (and are made!) if there are aspects of Google's Chrome product that people hate. And that's really all you need, I think. There's always an escape hatch.
posted by dis_integration at 2:43 PM

---------------------------

Lol, "purely technical." I wonder if this kind of person actually believes it's possible for a project to be "pure" and free of ideology, or if they're being cute and know exactly what this "please, no distasteful ideology in my beautiful technical project" stuff means. I assume the latter but who knows.
posted by Suedeltica at 3:37 PM

---------------------------

Many years late, but it seems like the wording was changed within the past hour or two. Link to GitHub commit.
posted by mundo at 4:42 PM

---------------------------

Oh, they don't use semantic linefeeds? Bah.

(Yes, I have wonky feelings about Markdown, of course I do)
posted by Pronoiac at 5:03 PM

---------------------------

Ideology is banned? Quick, hide the open source license!
posted by polytope subirb enby-of-piano-dice at 5:52 PM

---------------------------

"Well he got his first million."

LOL of course, and it's a Lunduke link of all people LOL. How very... apropros and telling.
At least the change was made (for now? wonder if the github dude was like "Look Kling... Here's the deal.")
posted by symbioid at 9:51 PM

---------------------------

Meanwhile, Servo, the new browser engine that was started and then killed at Mozilla, has been revived and is active, for what that's worth - hard to say whether it still has a lead.
posted by misuba at 10:54 PM

---------------------------

having given the matters at hand serious consideration over the course of a couple decades i have come to the conclusion that the only valid path forward is to (more or less) retvrn to gopher.

companies, and their non-profit cutouts (see: mozilla) have no valid role in the design and construction of worldwide information networks, and never have. in 1998 larry page and sergey brin convincingly argued that there must be a search engine in the academic realm that does not in any way fund itself through advertising, and that search engines that don't meet those criteria will inevitably and invariably be bad. those two men have made two and a half decades of mistakes since they wrote that 1998 paper, but their original idea was good — but insufficient.

if we want to have a useful global information interchange network, the entire network must be run by and for universities, governments, and ngos. full stop. the profit motive must be excised root and branch, and the disastrous mistake of 1993 — the decision to allow the for-profits in — must not be repeated. the network must be tightly focused on teaching, research, and civic participation, must support community formation toward those three ends, and must exclude everything else, because once the businesses get in they undermine the network's potential for teaching, research, civic participation, and community formation.
posted by bombastic lowercase pronouncements at 1:32 AM

---------------------------

retvrn to gopher.

there's a new protocol, gemini that's kind of a gopher for the modern day
posted by dis_integration at 6:09 AM

---------------------------