Firefox To Be Neglected For A.I. Pivot?

[suzyne]

Is this article clickbait, or valid concerns?

https://lunduke.locals.com/post/4671162/mozilla-downplaying-firefox-moving-into-a-i

tl;dr there is no specific direction for Firefox in the 25 year roadmap of Mozilla, and that is worrying.

[moonbat]

Why is it worrying? It was inevitable given the gradual deliberate degradation of the browser over the years and the effect on its marketshare
Good riddance if Firefox goes away - it will no longer be there as a fig leaf for Google to hide their browser monopoly behind. At this point taking it out back and shooting it in the head would be a mercy killing. My guess is they will do what Microsoft has done vis a vis Edge - retire Gecko and turn it into a Chrome wrapper that it has been anyway edging towards for the last 12 or so years. Frees up having to maintain more than a skeleton of a dev team and that much more money to line the CEO's pockets

What this would mean for Pale Moon - I dunno. AFAIK the codebases have diverged enough but there were still links to their dev team for porting CVEs and maybe adding new web features (not easy, don't know whether possible, given they moved to using Rust)

[athenian200]

I think Firefox has been neglected for years, honestly. I don't know if the A.I. pivot has much to do with it, so much as they are basically just giving up in general.

As far as I can tell, Mozilla as an organization has been more focused on developing and promoting the Rust programming language than the Firefox browser, and basically just turned Firefox into a showcase for what their new language could do. If they do go for an A.I. direction, I imagine they will try to find some way of incorporating that into Firefox too. Just like all their other half-baked ventures, like Pocket and FirefoxOS.

But yeah, for UXP, the biggest downside of Mozilla giving up on Firefox is that it would be harder for us to demonstrate that our browser was safe by addressing Mozilla CVEs. Also, user agent sniffing would become more of a problem, because as of right now we can report a newer Firefox user agent to websites as soon as we implement required features... if Firefox goes away, that option goes out the window, and it would be much harder to convince a website that Pale Moon is Chromium than that it is Firefox. Finally, while Rust has largely kept us from taking what we need for CSS-related stuff, a lot of JS stuff is still somewhat possible to backport. That will all have to be done from scratch if Mozilla ever calls it quits.

In short, it would be one further step on the road to Chromium domination... you all saw how much faster things got bad after Microsoft abandoned Trident, think what it will mean when Blink and WebKit is all any web developer has to worry about.

[suzyne]

Why is it worrying?

I often read that having only one browser engine available is bad and more than one is good to make it just a tiny bit harder for Google to push through future browser directions. But I gather that you think Firefox has been an ineffectual charade for a long time.

[moonbat]

But I gather that you think Firefox has been an ineffectual charade for a long time.

At this point they are just controlled opposition for Google, given how heavily Mozilla depends on their search engine revenue financially. Mozilla simply adopts whatever Google adds to the spec without much pushback other than for the manifest V3 (we'll see how long they hold out). Add to that the steady removal of features and visual aping of Chrome since Firefox 4, and one wonders why they don't just hurry up and switch to Blink since that's the inevitable end game.

[gepus]

In short, it would be one further step on the road to Chromium domination...

This!

[back2themoon]

...for UXP, the biggest downside of Mozilla giving up on Firefox is that it would be harder for us to demonstrate that our browser was safe by addressing Mozilla CVEs... ...if Firefox goes away, that option goes out the window, and it would be much harder to convince a website that Pale Moon is Chromium than that it is Firefox.

I wonder if there is a formal procedure to become registered/recorded as a browser/browser engine in some sort of globally relevant, official organization - if there's such a thing. Thus, it'd be easier for Goanna to pick up the baton from Gecko when the time comes, and hopefully lead more websites into testing it.

I know, quite unlikely. It's probably all about market share and financial/political power?

[Basilisk-Dev]

I wonder if there is a formal procedure to become registered/recorded as a browser/browser engine in some sort of globally relevant, official organization - if there's such a thing. Thus, it'd be easier for Goanna to pick up the baton from Gecko when the time comes, and hopefully lead more websites into testing it.

I know, quite unlikely. It's probably all about market share and financial/political power?

There is https://tc39.es/ which is a group involved in developing new Javascript proposals and standards. They've allowed some smaller groups such as the Ladybird browser from SerenityOS to join.

I often read that having only one browser engine available is bad and more than one is good to make it just a tiny bit harder for Google to push through future browser directions. But I gather that you think Firefox has been an ineffectual charade for a long time.

Many people who say that only one browser engine being bad are the same people who say "Pale Moon is just an old insecure fork of Firefox". They don't truly believe what they preach. If they did they would be happy there is an alternative engine such as Goanna available.

Off-topic:
I was literally called a white supremacist at one point by one of these people solely because I said I supported the Pale Moon browser. In that conversation I did not say anything that would indicate any sort of racist views. The mental leaps some people will go through are pretty insane sometimes.

[Moonchild]

It's probably all about market share and financial/political power?

It is.
This is why there has been unrelenting warfare against true alternatives gaining marketshare foothold. As a result we're in a niche, and niche browsers will simply not be given time of day when it comes to testing. I'm more than happy to have sec issues reported to me independently to take care of, as well. it just... hasn't really happened all that much.
Without significant market share, we're also not interesting as a target for black hats that generally require a significant quantity of exposed users to be profitable. As a result, security firms are also not interested in analyzing our code or reporting sec issues.

To be fair though, we haven't actually had much in terms of CVEs be applicable to us. Even with the ongoing development in the platform to catch up to the ever-increasing obscenity of the broader web specifications, without the incessant code churn and not in the least the on-going issues with e10s tossing sec issue after sec issue after sec issue up for them but not us, it certainly looks like our base is quite mature, stable and secure.
Take this release cycle for example. There were 3 sec bugs I addressed, all of them were not directly exploitable and the patches were defense-in-depth. The ones that weren't applicable were regressions from code churn (4x from insecure implementations of "optimizations", most of which are likely never impacting real-world scenarios) or e10s (8x) which is the absolute bulk of what I audited over the past year. Multi-process is just super fragile and in many cases like playing whack-a-mole; it's surprising it works at all, let alone reliably.

[Moonchild]

Off-topic:

I was literally called a white supremacist at one point by one of these people solely because I said I supported the Pale Moon browser. In that conversation I did not say anything that would indicate any sort of racist views. The mental leaps some people will go through are pretty insane sometimes.

I think that particular leap has resulted from me withdrawing from Patreon because of the at-the-time very outspoken support by them for BLM and dumping a ton of money into it. If you recall, I withdrew not because of any particular group Patreon was supporting, but rather that the pretty sizeable margin taken by Patreon was going towards goals and groups that supporters of Pale Moon might not want to implicitly also fund; Patreon also made clear that their mission was that they were going to continue doing these kinds of things for activist groups, without giving creators a choice if they wanted to be part of that or not.
There were some people who immediately made a stink out of that saying I "hated BLM" and therefore had to be a racist and/or nazi. And go from there...

[Mæstro]

To answer the original question, I would deem this hearsay unless Mozilla announce anything on newfound priorities. I think it likelier that attention will leave AI for something else, just as it had left cryptocurrency for AI a year ago. All these fascinations are…

[andyprough]

I think it likelier that attention will leave AI for something else, just as it had left cryptocurrency for AI a year ago.

But you don't seem to understand Mozilla's genius move here. Once they get AI figured out they can start using it to dominate the global cell phone market with Firefox OS smartphones.

Mozilla will be worth trillions.

[back2themoon]

There is https://tc39.es/ which is a group involved in developing new Javascript proposals and standards. They've allowed some smaller groups such as the Ladybird browser from SerenityOS to join.

That's interesting, thanks.

I used to believe that Firefox disappearing would be bad for the Internet and good for PM, but not so sure of the latter now. If that were to happen, perhaps it would help the Pale Moon project to venture into uncommonly (for PM and Moonchild) attention-grabbing moves like the aforementioned TC39 group, a security audit, a full-blown uBO fork and whatever else you all can think of. Not for publicity per se, but to secure a place in the unknown Web of the not-too-distant future.

[Moonchild]

Nice idea, but we have to solve more web compatibility first. And time is a very rare and valuable commodity for me at the moment too. All nice and good if I had little else to do but to get deeply involved with TC39, but that doesn't even cover the entirety of the browser. It's just for JS. so id have to get involved with TC39, W3C, WhatWG, CSSWG, the CAB forum, etc etc. and i just don't have the time to do that. And i don't have the budget to hire a lobbyist either.

[andyprough]

Nice idea, but we have to solve more web compatibility first. And time is a very rare and valuable commodity for me at the moment too. All nice and good if I had little else to do but to get deeply involved with TC39, but that doesn't even cover the entirety of the browser. It's just for JS. so id have to get involved with TC39, W3C, WhatWG, CSSWG, the CAB forum, etc etc. and i just don't have the time to do that. And i don't have the budget to hire a lobbyist either.

However, it should be done. Not by you using up all your time, surely. But Goanna is really the only independently developed browser engine that can browse the modern web to any reasonable extent and that isn't subservient to Google. Netsurf can't do it, Dillo can't, Otter can't.

I can't really understand what Seamonkey is doing anymore. It kind of works, but is it really in active development, or is it winding down?

Anyway, all of those web consortiums and working groups should be paying close attention to Goanna. There should be a reasonable way to assign certain knowledgeable volunteers to interface with them on behalf of the project, or something of that nature.

[moonbat]

I can't really understand what Seamonkey is doing anymore. It kind of works, but is it really in active development, or is it winding down?

They are between a rock and a hard place now as the Firefox codebase diverges further and further from the common platform code that it was built on. I bet they now regret spurning Moonchild's offer to switch to UXP back when it was was forked. Also a pity, had they done so we would at least have had several more experienced developers working on it.