Eleanor SaittaOct 4, 2024

It really sucks to be in a world where you have to choose between two vendors for the tools you must use to access most of your life, both of which have clearly stated they are actively working against both your best interests and the best interests of society. Of course, there's an alternative — various third parties, many of which are actively malicious and none of which are remotely reasonable choices when it comes to security, feature parity, or anything else.

Today this is about browsers, but the fact that I have to specify is its own problem.

Show threadJacketOct 4, 2024
@dymaxion What is the duopoly for browsers?
Show threadEleanor SaittaOct 4, 2024
@jacket
Chrome and Firefox. Safari is still digging its way out of a decade of underinvestment, and all the third party browsers a) use either the firefox or chrome engine, and b) with the exception of Microsoft, don't really have the development and security lift to keep up with their upstream and ensure their fork isn't causing problems.
Show threadJacket
@dymaxion The other alternatives are ports of those or uses the engine they use. If we look at web engines there is only 4 that are actively still being developed. WebKit (apple), Blink (google), Gecko (Mozilla) and Goanna (M. C. Straver). The LST one is a for of Gecko. It's a bit of a old school Gecko. So we can say that there is only 3 engine for all the browsers, kind of. There is an interesting project that is being developed at the moment called #ladybirdbrowser. It's a new browser made from scratch, with it's own engine. Something we didn't see in decades.
Oct 4, 2024 at 3:37pmWeb
Show threadEleanor SaittaOct 4, 2024

@jacket
Yes, I've been following browser development for 30 years and browser security for 20.

A browser isn't a rendering engine, and forking a browser, let alone a rendering engine, still leaves an amazing amount of work to do, in particular from a security perspective. While the alternatives may sometimes deliver interesting user experiences, I wouldn't ever recommend that someone use them in anger.

Show threadMiloWinterBurnOct 5, 2024
@dymaxion @jacket so now that we're here, what's your opinion on ladybird? Do they eventually have a shot?
Show threadEleanor SaittaOct 5, 2024
@MiloWinterBurn
Maybe in a five years if they get a hell of a lot of money? They're going to be looking at thousand of engineers to get to a level of feature and security parity that makes them as serious alternative without using other people's code, even if they're smart about architecture, language, and legacy support. If they start integrating other people's code, they'll also inherit its security issues, which means fewer devs but much more integration, patching, and hardening work. I don't see a funding model that's going to get them there, though.
@jacket
Show threadMiloWinterBurnOct 5, 2024

@dymaxion and I'm guessing this is the kind of coding where genAI can't contribute a lot?

I was thinking about funding models as well - was briefly triggered to go on one of my fun "let's build a startup in my head" joyrides again (I've done dozens, up to nuclear containerships). There is a demand, and there is money. But it's a long way there.

Show threadJacketOct 5, 2024
@MiloWinterBurn @dymaxion Ladybird need to go fast to get up to date with the features that exists on other browsers I think. But it is going fast. I checked the repo and there is no day passing without a bunch of PR. The development looks solid. So, I don't know. Maybe? It's a fun experiment whatever happens to it.
Show threadMiloWinterBurnOct 5, 2024
@jacket @dymaxion so how does their progress compare to servo? Servo seems to have more funding, but it seems like it's not nearly enough to get somewhere
Show threadJacketOct 5, 2024
@MiloWinterBurn @dymaxion I just follow ladybird by curiosity, and maybe to contribute at some point but I don't know much about servo. It looks interesting. I just know the challenge of creating a competitive browser nowadays is huge.
Show threadJacketOct 5, 2024
@dymaxion You're right. Do you think the problem is the bar being higher for a browser to be decent today? I feel like I could easily make a competitive browser for the 90's along pretty fast. I would probably need a big team to compete with today's browsers.
Show threadEleanor SaittaOct 5, 2024
@jacket
Well, the security bar cannot be lowered, and that leaves a lot of irreducible complexity. Things like service workers are pretty important to how people actually use web apps now. You could support fewer media formats and ditch DRM, of course, but that also breaks a lot of existing content. Yes, if you redesigned the web ecosystem from scratch you could eliminate a lot of complexity without losing functionality, but that's also not going to happen. Yes, some of that complexity exists as a competitive moat by browser vendors, and it's working.