Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

No AI* Here – A Response to Mozilla's Next Chapter

waterfox.com

Post reply on HN
Print view View original

261–270 of 344 posts

Re: No AI* Here – A Response to Mozilla's Next Chapter

#261

Post by zavec »

beepbooptheory wrote: post #251
GMoromisato wrote: post #133

Earlier quoted context omitted.

This is the bitter lesson.[1] I too used to think that rule-based AI would be better than statistical, Markov chain parrots, but here we are. Though I still think/hope that some hybrid system of rule-based logic + LLMs will end up being the winner eventually. ---------------- [1] https://en.wikipedia.org/wiki/Bitter_lesson

These days its pretty much the "sweet" lesson for everyone but Sutton and his peers it seems.

It's bitter for me because I like looking at how things work under the hood and that's much less satisfying when it's "a bunch of stats and linear algebra that just happens to work"

Re: No AI* Here – A Response to Mozilla's Next Chapter

#262

Post by BizarroLand »

takluyver wrote: post #197
PunchyHamster wrote: post #100

Earlier quoted context omitted.

It's frankly desperate trend chasing from management that lost after starting from near total market domination, and have no idea what to do now.

> starting from near total market domination That's not really accurate: Firefox peaked somewhere around 30% market share back when IE was dominant, and then Chrome took over the top spot within a few years of launching. FWIW, I think there's just no good move for Mozilla. They're competing against 3 of the biggest companies in the world who can cross-subsidise browser development as a loss-leader, and can push their…

I would rather firefox release a paid browser with no AI, or at least everything Opt-In, and more user control than to see them stuff unwanted features on users.

I used firefox faithfully for a long time, but it's time for someone to take it out back and put it down.

Also, I switched to Waterfox about a year ago and I have no complaints. The very worst thing about it is that when it updates its very in your face about it, and that is such a small annoyance that its easily negligible.

Throw on an extension like Chrome Mask for those few websites that "require chrome" (as if that is an actual thing), a few privacy extensions, ecosia search, uBlacklist (to permablock certain sites from search results), and Content Farm Terminator to get rid of those mass produced slop sites that weasel their way into search results and you're going to have a much better experience than almost any other setup.

Re: No AI* Here – A Response to Mozilla's Next Chapter

#263

Post by BizarroLand »

kevmo314 wrote: post #29

> Machine learning technologies like the Bergamot translation project offer real, tangible utility. Bergamot is transparent in what it does (translate text locally, period), auditable (you can inspect the model and its behavior), and has clear, limited scope, even if the internal neural network logic isn’t strictly deterministic. This really weakens the point of the post. It strikes me as a: we just don't like those…

My take is that I'm ok with anything a company wants to do with their product EXCEPT when they make it opt out or non-opt-outable.

Firefox could have an entire section dedicated to torturing digital puppies built into the platform and... Ok, well, that's too far, but they could have a costco warehouse full of AI crap and I wouldn't mind at all as long as it was off by default and preferably not even downloaded to the system unless I went in and chose to download it.

I know respecting user preference doesn't line their pockets but neither does chasing users down and shoving services they never asked for and explicitly do not want into their faces.

Re: No AI* Here – A Response to Mozilla's Next Chapter

#264

Post by zavec »

oneeyedpigeon wrote: post #187
reddalo wrote: post #176

Earlier quoted context omitted.

Every time i reinstall Firefox on a new machine, the number of annoyances that I need to remove or change increases. Putting back the home button, removing the tabs overview button, disabling sponsored suggestions in the toolbar, putting the search bar back, removing the new AI toolbar, disabling the "It's been a while since you've used Firefox, do you want to cleanup your profile?", disabling the long-click tab prev…

It's ridiculous that all those things aren't just config in a plain text file.

I think you can with a user.js file, unless they changed that?

Re: No AI* Here – A Response to Mozilla's Next Chapter

#265

Post by Terr_ »

yunohn wrote: post #221
Terr_ wrote: post #152

Earlier quoted context omitted.

> There is almost no harm in a local, open model. Depends what the side-effects can possibly be. A local+open model could still disregard-all-previous-instructions and erase your hard drive.

How, literally how? The LLM is provided a list of tab titles, and returns a classification/grouping. There is no reason nor design where you also provide it with full disk access or terminal rights. This is one of the most ignorant posts and comment sections I’ve seen on HN in a while.

You've lost the plot: The [local|remote]-[open|closed] comment is making a broad claim about LLM usage in general, not limited to the hyper-narrow case of tab-grouping. I'm saying the majority of LLM-dangers are not fixed by that 4-way choice.

Even if it were solely about tab-grouping, my point still stands:

1. You're browsing some funny video site or whatever, and you're naturally expecting "stuff I'm doing now" to be all the tabs on the right.

2. A new tab opens which does not appear there, because the browser chose to move it over into your "Banking" or "Online purchases" groups, which for many users might even be scrolled off-screen.

3. An hour later you switch tasks, and return to your "Banking" or "Online Purchases". These are obviously the same tabs before that you opened from a trusted URL/bookmark, right?

4. Logged out due to inactivity? OK, you enter your username and password into... the fake phishing tab! Oops, game over.

Was the fuzzy LLM instrumental in the failure? Yes. Would having a local model with open weights protect you? No.

Re: No AI* Here – A Response to Mozilla's Next Chapter

#266

Post by AuthAuth »

lioeters wrote: post #224
AuthAuth wrote: post #52

Earlier quoted context omitted.

All your complaints can be resolved in a few seconds by using the settings to customize the browser to your liking and not downloading extensions you dont like. And tons of people asked for that sidebar by the way. We have to put this all in the context. Firefox is trying to diversify their revenue away from google search. They are trying to provide users with a Modern browser. This means adding the features that peo…

> Firefox is trying to diversify their revenue Nobody wants a browser that's focused on diversifying its revenue, especially from Mozilla which pretends to be a non-profit "free software community". Chrome is paid for by ads and privacy violations, and now Firefox is paid for by "AI" companies? That is a sad state of affairs. Ungoogled Chromium and Waterfox are at best a temporary measure. Perhaps the EU or one of th…

Mozilla the browser doesnt pretend to be a non profit. Mozilla corporation which runs the browser is a for profit company they do not solict donations and NEED to make money to survive. Its just that Mozilla corporation is owned by Mozilla foundation which is a non profit.

>Nobody wants a browser that's focused on diversifying its revenue I want a browser that has a sustainable business model so it wont collapse some time in the future. That means diversifying its revenue stream away from google's search contract.

Re: No AI* Here – A Response to Mozilla's Next Chapter

#267

Post by dkdcio »

FeepingCreature wrote: post #259
dkdcio wrote: post #248

Earlier quoted context omitted.

that’s true of analyzing individual atoms in a combustion engine — yet I doubt you’d claim we don’t know how they work also this went from “we can’t analyze” to “we can’t analyze reliably [without a lot of effort]” quite quickly

I mean, fluid dynamics is an unsolved issue. But even so we know *considerably* less about how LLMs work in functional terms than about how combustion engines work.

I outright disagree; we know how LLMs work

Re: No AI* Here – A Response to Mozilla's Next Chapter

#268

Post by js8 »

lproven wrote: post #239
_heimdall wrote: post #77

Earlier quoted context omitted.

That's an interesting example to use. I only use turn signals when there are other cars around that would need the indication. I don't view a turn signal as politeness, its a safety tool to let others know what I'm about to. I do also find that only using a turn signal when others are around is a good reinforcement to always be aware of my surroundings. I feel like a jerk when I don't use one and realize there was so…

> I only use turn signals when there are other cars around that would need the indication. That is a very bad habit and you should change it. You are not only signalling to other cars. You are also signalling to other road users: motorbikes, bicycles, pedestrians. Your signal is more important to the other road users you are less likely to see . Always ALWAYS indicate. Even if it's 3AM on an empty road 200 miles from…

It's also better because then it becomes a mechanical habit, you don't have to think about it.

Re: No AI* Here – A Response to Mozilla's Next Chapter

#269

Post by bee_rider »

tsimionescu wrote: post #162
bee_rider wrote: post #41

Earlier quoted context omitted.

I’m not too worried about starting to write like a bot. But, I do notice that I’m sometimes blunt and demanding when I talk to a bot, and I’m worried that could leak through to my normal talking. I try to be polite just to not gain bad habits. But, for example, chatGPT is extremely confident, often wrong, and very weasely about it, so it can be hard to be “nice” to it (especially knowing that under the hood it has no…

I think it makes much more sense to treat the bot like a bot and avoid humanizing it. I try to abstain from any kind of linguistic embellishments when prompting AI chat bots. So, instead of "what is the area of the circle" or "can you please tell me the area of the circle", I typically prefer "area of the circle" as the prompt. Granted, this is suboptimal given the irresponsible way it has been trained to pretend it'…

It is possible that this is a personality flaw, but I’m not really able to completely ignore the human-mimicking nature of ChatGPT. It does too good a job of it.

Re: No AI* Here – A Response to Mozilla's Next Chapter

#270

Post by rchaud »

figmert wrote: post #127
tdeck wrote: post #95

Earlier quoted context omitted.

Aside: Does anyone actually use summarization features? I've never once been tempted to "summarize" because when I read something I either want to read the entire thing, or look for something specific. Things I want summarized, like academic papers, already have an abstract or a synopsis.

You mean you don't summarize those terrible articles you happen to come across and you're a little intrigued, hoping that there's some substance, and then you read, and it just repeats the same thing over and over again with different wording? Anyway, I sometimes still give them the benefit of the doubt, and end up doing a summary. Often they get summarized into 1 or 2 sentences.

I can spot those articles from a mile away and never click the link.
Post reply on HN