at this point I think I just have a sense of resignation that mozilla is probably going to end up "accidentally" training an LLM on my passwords and credit card numbers 🙃
someone will then get in your mentions aghast you were bothered by this, as (1) well, that's just what every company does and (2) if an ethical open source company doesn't get in there and train an LLM on your credit card numbers, then doesn't that just mean that the big evil corporations will be in charge of training LLMs on your credit card numbers?
btw, does anyone have like a wiki (or google doc or something) that ranks browsers in order of least-to-most harmful/hostile to users?
bc this keeps happening (and it's why i migrated from chrome to firefox a couple/few years back) and it seems like it deserves its own wikipedia article if not an entire website
@JamesWidman yeah uhhh i feel like it's less a ranking and more of like a fire emblem style weapons triangle where each browser defeats each other in different areas
all three browsers have like-third-party-cookies-but-worse systems where they gather data on where their users browse and send them directly to advertisers, but google is the only one that gets a form of consent from the user before enabling it. based on this i'd argue if there were a ranking google would be at the top :(
My favorite part is "the crypto stuff" masquerades as tipping, which works like this:
You like what Tom Scott is doing. Brave sees you watching Tom Scott videos, suggests you tip him. So Brave sells you some crypto (Brave takes a cut, naturally), and then you send him a tip (Brave takes a cut, naturally.)
But Tom Scott isn't aware of this scheme, and eventually discovers that Brave has been holding all the tips they've talked his fans into giving him "in escrow." In order to get the tips his fans have given "him" he just has to sign up for a Brave partner account, and then can withdraw the tips. (Brave takes a cut, naturally.)
Except now Tom Scott has a useless Brave branded crypto. He can convert it into dollars (and Brave takes a cut, naturally.)
Debian is also shipping ESR (IE the 128 line from last July) as just its normal Firefox. This is actually better than stable Firefox because it doesn't have "ad measurement". "Old Firefox" is only a complaint if Waterfox is failing to take the bugfixes and security updates from the ESR channel.
@nyquildotorg @glassresistor @nat @JamesWidman @nullagent Honestly, the best case scenario might be to have someone or someones *reject* the next ESR and maintain bugfixes for Firefox 128 indefinitely. Just make 2024 ESR a hard fork. No more features, ever. Society has progressed past the need for features
is it an up to date ESR tho. Like I could imagine them failing to republish ESR point releases. 128 ESR has got some point releases right? I am posting this from bed and cannot check lol
...i've accepted that we can't have a total ordering, but it definitely sounds like we need some wiki/site that lays out the current options, the fire-emblem-style weapons triangle, and the state of the shortcomings of every browser that isn't actively trying to sap & impurify all of our precious bodily fluids
@JamesWidman yeah uhhh i feel like it's less a ranking and more of like a fire emblem style weapons triangle where each browser defeats each other in different areas
all three browsers have like-third-party-cookies-but-worse systems where they gather data on where their users browse and send them directly to advertisers, but google is the only one that gets a form of consent from the user before enabling it. based on this i'd argue if there were a ranking google would be at the top :(
fundamentally, what we're missing is an organization with a stable, long-term source of funding that *isn't* based on advertising.
Development probably needs to be funded by the government(s) of one or more countries that stand a good chance of not turning fash in the foreseeable future.
Every time Mozilla steps on a rake regarding privacy or security, people come rushing out telling everyone to switch to Firefox forks. And I always feel bad about it, because if your main concern is privacy or security, there is no scenario where you would be better off on a Firefox fork
@JamesWidman @nyquildotorg @glassresistor @nat @nullagent I still don't take this altogether seriously. The argument here is less that you shouldn't *use* a Firefox fork, but rather that Firefox forks will be unsustainable if Mozilla goes out of business. But nothing I do will influence whether Mozilla goes out of business— it will happen whether I use Firefox, a Firefox fork, or Chrome. Mozilla will try very hard to go out of business even if I try very hard to support them.
@JamesWidman @nyquildotorg @glassresistor @nat @nullagent I'm also not sure it's supported. People in the 90s didn't stop and go "wait— how will we pay for this without a corporate sponsor?" before writing open source software. What if us thinking we need the Mozilla Organization, specifically is learned helplessness? Lots of open source projects today exist without becoming advertising companies. Linux hasn't turned into an advertising company.
I think the big thing that differentiates Firefox from any other open source project is it was a very commercial project that was essentially abandoned when it was found not to be economically viable to continue to work on, and well-meaning open-source loving employees convinced the owner to let them open source it. And then it was like "ok... now what?"
where "now what?" ended up being "continue to run it like a business"
Mozilla folks claim that there were legal requirements that mandated having a commercial entity (which appears to be a similar situation with WordPress. The company that handles WordCamp is a wholly owned subsidiary that is not a non-profit specifically because of some kind of tax reasons behind donations or something), but I've never been convinced that the browser needed to be handled under that entity.
@nyquildotorg @JamesWidman @glassresistor @nat @nullagent It does legitimately appear trying to operate as a nonprofit in the US is very hard due to weird scope rules. However, I don't think the for-profit entity is the problem here. I think we are seeing what the people I know in social justice work call "the nonprofit industrial complex". A board that starts operating for the purpose of keeping the board employed and paid. (Not accusing anyone of anything specific here!)
@mcc @nyquildotorg @glassresistor @nat @nullagent right, linux isn't an advertising company, but Linux does get some funding, probably because each of the "platinum members" of the linux foundation maintain a lot of critical stuff running on linux in-house, so they view the $500,000 membership fee as a tech support fee.
In that case, they can probably see a clear connection between "spend $500k per year" and "receive maintenance of large & critical business systems".
@JamesWidman @nyquildotorg @glassresistor @nat @nullagent So, there are companies that benefit from the existence of Linux. And what we've seen since 2000 is that it is easier to get that kind of funding for a architectural/server component than it is to get it from a pure GUI component. And also arguably Linux is a simpler project than a web browser. But that's not the same as saying a non-architecture, non-malware GUI application is entirely unfundable, not even at the browser scale.
Maybe something like the linux foundation for browser development will be motivated after some browser-facilitated security breaches...? But it would be nice if we didn't have to wait that long.
In the 6+ years that I've been bitching on the fediverse every single time Mozilla steps on the rakes they keep inexplicably throwing down, I've had a very large number of people assure me that they will never, ever pay money for a browser.
So that "for free" thing is a bigger obstacle than a lot of people expect.
@mcc @nyquildotorg @JamesWidman oh! I should do that too. I paid money for Blender and didn't use it for several years and that ended up being really positive.
I also wonder how many resources are actually needed to fork and maintain a browser engine starting from one of the bases we have now.
Obviously creating a new engine+UI from scratch that can work with the present state of web applications is extremely complicated, compared to customizing the UI around an existing one.
I suppose the facts that Opera and Microsoft threw out their old codebases and all the Chromium based browsers are still tracking upstream Chromium says something about that
But what would it take for a project like LibreWolf or WaterFox to become self-sustaining such that they would be insulated a bit from upstream hijinks?
(Apparently Pale Moon is still around. I just looked and there's a recent release, though I have no idea what the current state of the community or software is.)
@kelson @mcc @JamesWidman @nyquildotorg @glassresistor @nat @nullagent in my experience, being road hauled by an upstream vendor with hundreds of contributors with different priorities than you is several full time jobs worth of frustrating work if you're doing a good job maintaining your fork.
@aeva@gamedev.place IMO Vivaldi is the most trustworthy of the Chromium lot, unless you go for fully-Ungoogled Chromium (though plain old Chromium is ok on Linux where there are actual package managers for it instead of it just being the dev branch for Chrome). I don't trust Brave or Opera. Arc is interesting, but more as an experiment, and it's not available for Linux. I like Falkon too, which is KDE's replacement for Konqueror, but it's rudimentary compared to the rest.
On the Gecko side, I've only just started trying out current alternatives. LibreWolf seems to be a good privacy-respecting fork of Firefox, and Zen's interesting -- sort of reimplementing Arc's experimental UI on top of Firefox.
mcc
in reply to aeva • • •James Widman
in reply to mcc • • •btw, does anyone have like a wiki (or google doc or something) that ranks browsers in order of least-to-most harmful/hostile to users?
bc this keeps happening (and it's why i migrated from chrome to firefox a couple/few years back) and it seems like it deserves its own wikipedia article if not an entire website
aeva
in reply to James Widman • • •mcc
in reply to James Widman • • •@JamesWidman yeah uhhh i feel like it's less a ranking and more of like a fire emblem style weapons triangle where each browser defeats each other in different areas
all three browsers have like-third-party-cookies-but-worse systems where they gather data on where their users browse and send them directly to advertisers, but google is the only one that gets a form of consent from the user before enabling it. based on this i'd argue if there were a ranking google would be at the top :(
Nat Oleander
in reply to James Widman • • •aeva
in reply to Nat Oleander • • •aeva
in reply to aeva • • •Jer Warren
in reply to aeva • • •My favorite part is "the crypto stuff" masquerades as tipping, which works like this:
You like what Tom Scott is doing. Brave sees you watching Tom Scott videos, suggests you tip him. So Brave sells you some crypto (Brave takes a cut, naturally), and then you send him a tip (Brave takes a cut, naturally.)
But Tom Scott isn't aware of this scheme, and eventually discovers that Brave has been holding all the tips they've talked his fans into giving him "in escrow." In order to get the tips his fans have given "him" he just has to sign up for a Brave partner account, and then can withdraw the tips. (Brave takes a cut, naturally.)
Except now Tom Scott has a useless Brave branded crypto. He can convert it into dollars (and Brave takes a cut, naturally.)
aeva
in reply to Jer Warren • • •meta physical deflationist
in reply to aeva • • •Jer Warren
in reply to meta physical deflationist • • •I don't know anything about floorp, but the common issue folks have eith waterfox is being based on pretty old firefox releases.
The current waterfox release seems to be based on Firefox 128 from last july.
mcc
in reply to Jer Warren • • •@nyquildotorg @glassresistor @nat @JamesWidman @nullagent is that an old release or is it a current ESR release?
Debian is also shipping ESR (IE the 128 line from last July) as just its normal Firefox. This is actually better than stable Firefox because it doesn't have "ad measurement". "Old Firefox" is only a complaint if Waterfox is failing to take the bugfixes and security updates from the ESR channel.
mcc
in reply to mcc • • •Jer Warren
in reply to mcc • • •mcc
in reply to Jer Warren • • •James Widman
in reply to Jer Warren • • •@nyquildotorg @glassresistor @nat @mcc @nullagent
between this and:
mastodon.social/@mcc/114079708…
...i've accepted that we can't have a total ordering, but it definitely sounds like we need some wiki/site that lays out the current options, the fire-emblem-style weapons triangle, and the state of the shortcomings of every browser that isn't actively trying to sap & impurify all of our precious bodily fluids
mcc
2025-02-28 04:50:13
James Widman
in reply to James Widman • • •@nyquildotorg @glassresistor @nat @mcc @nullagent everyone who suggested a firefox-based thing (e.g. floorp) should read this:
vmst.io/@jalefkowit/1140828235…
fundamentally, what we're missing is an organization with a stable, long-term source of funding that *isn't* based on advertising.
Development probably needs to be funded by the government(s) of one or more countries that stand a good chance of not turning fash in the foreseeable future.
Jason Lefkowitz
2025-02-28 18:02:19
mcc
in reply to James Widman • • •mcc
in reply to mcc • • •Jer Warren
in reply to mcc • • •I think the big thing that differentiates Firefox from any other open source project is it was a very commercial project that was essentially abandoned when it was found not to be economically viable to continue to work on, and well-meaning open-source loving employees convinced the owner to let them open source it. And then it was like "ok... now what?"
where "now what?" ended up being "continue to run it like a business"
Jer Warren
in reply to Jer Warren • • •mcc
in reply to Jer Warren • • •mcc
in reply to mcc • • •James Widman
in reply to mcc • • •@mcc @nyquildotorg @glassresistor @nat @nullagent right, linux isn't an advertising company, but Linux does get some funding, probably because each of the "platinum members" of the linux foundation maintain a lot of critical stuff running on linux in-house, so they view the $500,000 membership fee as a tech support fee.
In that case, they can probably see a clear connection between "spend $500k per year" and "receive maintenance of large & critical business systems".
mcc
in reply to James Widman • • •Joe Groff
in reply to mcc • • •mcc
in reply to Joe Groff • • •silverwizard
in reply to mcc • •Joe Groff likes this.
ROTOPE~1
in reply to silverwizard • • •silverwizard likes this.
James Widman
in reply to James Widman • • •@mcc @nyquildotorg @glassresistor @nat @nullagent
like linux, browsers are also critical infrastructure, but so far everyone has been getting them "for free".
Maybe something like the linux foundation for browser development will be motivated after some browser-facilitated security breaches...? But it would be nice if we didn't have to wait that long.
Jer Warren
in reply to James Widman • • •In the 6+ years that I've been bitching on the fediverse every single time Mozilla steps on the rakes they keep inexplicably throwing down, I've had a very large number of people assure me that they will never, ever pay money for a browser.
So that "for free" thing is a bigger obstacle than a lot of people expect.
mcc
in reply to Jer Warren • • •aeva
in reply to mcc • • •Kelson
in reply to mcc • • •I also wonder how many resources are actually needed to fork and maintain a browser engine starting from one of the bases we have now.
Obviously creating a new engine+UI from scratch that can work with the present state of web applications is extremely complicated, compared to customizing the UI around an existing one.
I suppose the facts that Opera and Microsoft threw out their old codebases and all the Chromium based browsers are still tracking upstream Chromium says something about that
But what would it take for a project like LibreWolf or WaterFox to become self-sustaining such that they would be insulated a bit from upstream hijinks?
(Apparently Pale Moon is still around. I just looked and there's a recent release, though I have no idea what the current state of the community or software is.)
aeva
in reply to Kelson • • •Kelson
in reply to James Widman • • •@aeva@gamedev.place IMO Vivaldi is the most trustworthy of the Chromium lot, unless you go for fully-Ungoogled Chromium (though plain old Chromium is ok on Linux where there are actual package managers for it instead of it just being the dev branch for Chrome). I don't trust Brave or Opera. Arc is interesting, but more as an experiment, and it's not available for Linux. I like Falkon too, which is KDE's replacement for Konqueror, but it's rudimentary compared to the rest.
On the Gecko side, I've only just started trying out current alternatives. LibreWolf seems to be a good privacy-respecting fork of Firefox, and Zen's interesting -- sort of reimplementing Arc's experimental UI on top of Firefox.
Edit: I should link to the reviews I've been writing as I go: hyperborea.org/reviews/tag/web…
Web Browser at Kelson Reviews Stuff
Kelson Reviews Stuffaeva
in reply to Kelson • • •aeva (@aeva@mastodon.gamedev.place)
Gamedev Mastodon