Firefox uses on-device downloaded-on-demand ML models for privacy-preserving translation.
Uncategorized
220
Posts
107
Posters
1.1k
Views
-
Firefox uses on-device downloaded-on-demand ML models for privacy-preserving translation.
They're not LLMs. They're trained on open data.
Should translation be disabled if the AI 'kill switch' is active?
@firefoxwebdevs I’d say keep the translation thing and simply lose all the other LLM/GenAI/chatbot stuff altogether*. I think this is an excellent marketing opportunity. There are plenty of people highly skeptical of “AI”. This is a big market! You could be the next Brother, winning by refraining from shooting yourself in the foot (https://www.theverge.com/23642073/best-printer-2023-brother-laser-wi-fi-its-fine). And you’ll be ahead of the curve when The Bubble pops.
I’m not kidding.(*they can still be opt in plugins)
-
@firefoxwebdevs The frame of this question is risible.
I am begging you to just make a web browser.
Make it the best browser for the open web. Make it a browser that empowers individuals. Make it a browser that defends users against threats.
Do not make a search engine. Do not make a translation engine. Do not make a webpage summariser. Do not make a front-end for an LLM. Do not make a client-side LLM.
Just. Make. A. Web. Browser.
Please.
@m0rpk @firefoxwebdevs quite honestly, you're off the mark, **a lot**.
A browser with a built-in translator is a door opener for the open web for so many people that don't read English well enough to benefit from the dominant corpus of technological, cultural and scientific websites.
Firefox could indeed remove that functionality and instead of letting people translate websites on their phone make them use the google translate app that directly. Congrats on how you've advocated for the open web. -
@firefoxwebdevs I don't care. Local translation in FF is on the level of free early 2000s web translators. So maybe just remove it and add it again, when it's production ready
@flxtr @firefoxwebdevs as someone who used these in the early 2000s: no, it's not. It's not as good as DeepL, but it's worlds ahead of machine translation in the 2000s.
-
@joepie91 yeah, I agree with all that, but even tech folks are asking for a way to 'get rid of AI'. I'm pretty certain if we tried to redefine what they're asking for, it would be received poorly.
@firefoxwebdevs @joepie91 i'm a "tech folk". Just give us a version of firefox with zero AI. Translation can either be an extension or not there. We ask of you to supply a base for broSing the web, the rest is what the community delivers.
We won't ask you to integrate ad blockers, but we have them.
We won't ask you to integrate quick procy switchers, but we have them.Stop the feature creep and go back to the roots, make a very good browser with extension support and let people make the rest.
-
@angelfeast @twifkak No, I don't think so. It says this (with a takedown compliance process posted afterward)...
License
These data are released under this licensing scheme: PD
We do not own any of the text from which these data has been extracted.
We license the actual packaging of these parallel data under the Creative Commons CC0 license ("no rights reserved"). -
Firefox uses on-device downloaded-on-demand ML models for privacy-preserving translation.
They're not LLMs. They're trained on open data.
Should translation be disabled if the AI 'kill switch' is active?
@firefoxwebdevs you came up with the "killswitch" as if it was opt-in (it's *clearly* opt-out!), you put translate and llm-stuff into one box, *you* are the ones engaging in worst faith. why don't you go ahead and ask us why we're punching ourselves?
-
@wes @firefoxwebdevs Sure. But can we agree that it does not represent a core functionality of a web browser?
Like "this meeting could've been an email," but "this feature could've been an add-on."
A web browser should load web pages, allow you to interact with them, and offer add-on support for functionality that doesn't match the definition of "web browser." It's all pretty straight-forward if you're not a marketer, whose brains are all broken.
@liquor_american @wes @firefoxwebdevs This is super reductive. There is not some canonical definition of "web browser".
-
@zzt @firefoxwebdevs OK, now make the same argument for the spell-checker, sync, and the set of CAs, etc. etc. supplied with the browser. Its as if y'all were trained by Microsoft PR to take the arguments Mozilla used against tying IE to Windows and extend them ad-absurd-um to features in Mozilla's own browser ("just turn it around back in their faces" said the Armani suit).
Meanwhile, Red Hat is quietly undermining any legal basis for copyleft and leaning into the idea that gratis products (Fedora) shouldn't have robust & transparent system update tools. Oh and the umpteen other for-profit controlled (opposite of Mozilla) FOSS projects that get plugged in these spaces pretty much constantly. Linux Foundation being controlled by Microsoft and Google...? crickets chirping.
This is what makes me tired of IT and geek culture. Its become like everything else, just kneejerk crap with zero reflection and sense of proportion. As I hinted above, it morphs into this shadow of corporate PR. Consider, if people spent their time criticizing actual badness in Firefox, like ad tracking and DoH, that would be inconvenient for certain interests from Brave on up to Apple and Google. I think the style and quality of venting we usually see about Mozilla serves those interests, much of it probably fed by sock puppets.
"Meanwhile, Red Hat is quietly undermining any legal basis for copyleft and leaning into the idea that gratis products (Fedora) shouldn't have robust & transparent system update tools."
it's a bit off topic, but would you mind elaborating more about the system update tools? i'm out of the loop on that, and it sounds concerning
-
@firefoxwebdevs you came up with the "killswitch" as if it was opt-in (it's *clearly* opt-out!), you put translate and llm-stuff into one box, *you* are the ones engaging in worst faith. why don't you go ahead and ask us why we're punching ourselves?
@malte there will be granular options for this stuff. The question is about the non-granular "kill switch".
-
@firefoxwebdevs come on man.
@dante seems like a valid question to me. I mean it's literally a different tool than prompted genAI, and the definition of "AI" keeps shifting.
-
Firefox uses on-device downloaded-on-demand ML models for privacy-preserving translation.
They're not LLMs. They're trained on open data.
Should translation be disabled if the AI 'kill switch' is active?
I chose “No”. I find the translation feature very useful and greatly appreciate that is is local.
I do however think the local translate functionality should have an enable/disable switch right next to the AI enable/disable switch along with a brief and expanded description of functionality and locality of the feature.
-
Firefox uses on-device downloaded-on-demand ML models for privacy-preserving translation.
They're not LLMs. They're trained on open data.
Should translation be disabled if the AI 'kill switch' is active?
@firefoxwebdevs nobody wants any of it.
The machine learning, the LLM, the translations, the "open model", the "open data" the open model uses, the kill switch, none.
I've use firefox since 2005 maybe 2006, even spent about 8 hours compiling it on my laptop to have it.
I've keept using your software despite its many flaws because you provided something.
You putting AI is taking the one advantage that justified using firefox over the alternatives.Just to name a few flaws:
* In the last 20 years I've not managed to make firefox keep 2 dictionaries installed and working as expected over time.* I'm having to download firefox focus from your FTP because you don't seem to want to have it available outside of the play store.
Why don't you maintain a f-droid repo for it?* Where is USB midi support? Where is most of the USB support?
So much stuff we have to use chrome for.* Why do you hate people restoring tabs?
Why move the lines to restore the tabs and delete said tabs next to each other?The AI push in your software makes none of the effort of using firefox worth it.
-
@firefoxwebdevs Here's a concrete example of what I mean, that should be pretty consistent with the Firefox UI design:
@joepie91 I think a lot of people in the replies would consider this sneaky. It's a tricky UX problem. But yes, granular control needs to be part of the solution, along with a kill switch.
-
@malte there will be granular options for this stuff. The question is about the non-granular "kill switch".
@firefoxwebdevs you can't cherry-pick yourself out of your general bad faith engagement.
-
@firefoxwebdevs Perhaps it would be a good idea to do occasional one-time surveys of Firefox users - like when they start the browser up after an update. That way you get to hear directly from the people who are using it. Lots of folk on the socials have strong opinions but aren't necessarily using FF as a daily driver.
If I was writing the questions they might include things like...
- Should FF enable new AI features by default? [y/n]
- Would you like to be able to see at a glance which AI features are enabled? [y/n]
- Are there any particular features (AI or not!) that you feel FF is missing, and which you would actually use on a regular basis?
On that last one, I would maybe have some check boxes for things that tend to come up again and again like native RSS reader, FTP, Gemini (protocol!), WebUSB, WebSerial, UXP etc.
@m I agree the folks I'm polling here do not represent the average user, but in this case I'm specifically interested in the thoughts of those who really dislike 'AI', and I think I've reached them 😀
-
@liquor_american @wes @firefoxwebdevs This is super reductive. There is not some canonical definition of "web browser".
@tedmielczarek @wes @firefoxwebdevs Yes, this is what the marketers keep trying to convince us of.
-
Firefox uses on-device downloaded-on-demand ML models for privacy-preserving translation.
They're not LLMs. They're trained on open data.
Should translation be disabled if the AI 'kill switch' is active?
@firefoxwebdevs As worded, and if we can trust Mozilla, then the acceptable answer should be No for these reasons: ML is not AI, and on-device means nothing is sent out of the device. In exchange you get free translation. Win.
BUT… there’s the trust issue now.
And what we REALLY need is not an AI kill switch but more of a “data transfer/phone-home kill switch”, almost like a firewall, where we know the browser is not taking any data and sending it to a device we don’t control ourselves.
-
@made @firefoxwebdevs There's already lots of work for on-device ML: https://searchfox.org/firefox-main/search?q=toolkit%2Fcomponents%2Fml
Integrating models into a finalized product with the wide spectrum of end-user devices is tricky though, so it has to be done with care.
@gregtatum @firefoxwebdevs great to hear! I can imagine! Thanks for the link ☺️
-
@firefoxwebdevs As worded, and if we can trust Mozilla, then the acceptable answer should be No for these reasons: ML is not AI, and on-device means nothing is sent out of the device. In exchange you get free translation. Win.
BUT… there’s the trust issue now.
And what we REALLY need is not an AI kill switch but more of a “data transfer/phone-home kill switch”, almost like a firewall, where we know the browser is not taking any data and sending it to a device we don’t control ourselves.
@mdavis folks want to disable 'AI' for more reasons than privacy. Privacy is important of course, but folks are also concerned about the training data, and energy used for the training.
-
@firefoxwebdevs not trying to split hairs here but how are the ML models doing translation when they are not LLMs? Maybe they are not as huge as ChatGPT but they are transformers probably with all that entails.
(A Killswitch should of course kill all ML/AI functionality and people could then reactivate certain specific features of they want to, it's really not that hard. Just cause you consider a feature"better" than others does not override consent practices.)
@tante it’s a SLM :)
Gli ultimi otto messaggi ricevuti dalla Federazione
-
@levelbot good job everyone
-
@thedarktangent The problem may be related to the #authorized_fetch parameter being enabled.
Here is the GitHub discussion:
https://github.com/LemmyNet/lemmy/issues/4451
Here is the parameter specification in the Mastodon documentation:
https://docs.joinmastodon.org/admin/config/#authorized_fetch
@signorina37 @grabbi_it @informapirata@defcon.social @support @TCMBC @informapirata@mastodon.uno
-
Hmmm.... I'll just put him in Andorra, since that's right smack in the middle
-
Se esistesse una comunità di #T1, ben vengano i confronti.
Foto puramente dimostrativa, clickbait e nu*ity.
-
horny levels are 74% and steady
(74%) ■■■■■■■□□□
-
@thedarktangent This is interesting.
Every time a defcon.social user tries to interact with feddit.it, this error appears:2026-02-26T09:23:02.075679Z WARN lemmy_server::root_span_builder: CouldntFindObject: Failed to parse object https://defcon.social/@signorina37/116135473370481795 with content {"error":"Request not signed"}: data did not match any variant of untagged enum SearchableKinds
by @signorina37
or
2026-02-26T07:16:22.978630Z WARN lemmy_server::root_span_builder: Unknown: Failed to parse object https://defcon.social/users/grabbi_it/statuses/116133671937824462 with content {"error":"Request not signed"}: data did not match any variant of untagged enum PageOrNote
by @grabbi_it
We are currently receiving errors from the following instances, but they don't have much in common: they are running different versions or even forks.
https://bsd.network
https://cathode.church
https://defcon.social
https://kolektiva.social
https://mastodon.art
https://social.lol
https://universeodon.com
https://wehavecookies.social@informapirata@defcon.social @support @TCMBC @informapirata@mastodon.uno
-
@eniko I voted for “other” because although I have never personally roasted a whole chicken, we do have “girarrosto” shops here that basically sell just that
-
@NIGHTEN the right one (second)
