More important question is - how this nitter instance is still working!!
shh…it’s a spyware and adware!
What Firefox provides here:
A connector to LLM providers.
Accelerators (context menu options).
From a coding perspective, this should ideally be a very lightweight functionality.
This feature is very analogous to options to add a search engine, and also to provide accelerators via context menu.
While it can be done via third-party or Official Mozilla add-ons, but (to me) it still makes sense to have it part of the product.
If you’re using a VPN at the OS or browser level, just like any other traffic, your query to the LLM service will be routed via the VPN. That VPN could be any VPN of your choice - Firefox VPN, Mullvad, or Proton etc.
The only problem is that most LLMs require a profile/login to work with. In such cases, using a VPN will be useless, as the LLM server will know who you are.
It’s just a plain integration with 3rd-party or self-hosted LLM service.
I’m not sure if Mozilla will make money from this feature in any way.
Have you read anything about it anywhere?
It’s just an integration with LLM services and not AI baked-in the browser code. You can even self-host any such service (Ollama) and integrate Firefox with it. That will make sure your query is not leaving your network.
deleted by creator
Rustdesk controversy
The whole discussion on that pull request is extremely sketchy, IMO.
Tuba is now added to Gnome Circle. That’s a good news.
It’s about their FakeSpot subsidiary.
https://support.mozilla.org/en-US/kb/review-checker-review-quality#w_protect-your-privacy
Protect your privacy Firefox is committed to empowering you with information about review reliability while respecting your privacy. We use Oblivious HTTP (OHTTP) for Review Checker. When Review Checker is turned on, we use information about the products you visit on Amazon, Best Buy and Walmart to analyze the reviews, but by using OHTTP we ensure Mozilla cannot link you or your device to the products you have viewed. OHTTP uses encryption and a third party intermediary server to offer a technical guarantee that this is the case: all Mozilla learns from this network request is that someone, somewhere, looked at a given product.
Also not what I said.
Source: 2022 Hey look, years ago. And your other page was 2018.
Mozilla started selling private data to advertising companies in 2023
(Assuming this is about Pocket) Is it too much to expect from you to know the difference between aggregated non-PII data vs PII data?
Yes, like publishing a new article every day just to prove their commitment to end-users’ privacy.
Incremental updates to articles, hosted literally on home page, with details of newer privacy features is so old school.
Got it. Thanks for the clarification.
Source: 2022
Incorrect, that’s actually from 2022 B.C.
And your other page was 2018
Correct, the snap of article from 2018 looks exactly identical to 2024 instance with ZERO modifications. Mozilla finally gave us on Privacy it seems, as no one bothered to update that page since 2018.
Wait a sec, they also haven’t updated this article as well since 2020. https://www.mozilla.org/en-US/firefox/browsers/compare/chrome/
/s
Seems to have some severe bugs: https://github.com/wasi-master/13ft/issues
but Mozilla itself doesn’t want to broach the topic.
Again, a reminder that Mozilla plans to continue support for the Manifest Version 2 blocking WebRequest API (this API powers, for example, uBlock Origin) while simultaneously supporting Manifest Version 3.
Years ago, Mozilla would explicitly call ad blocking a privacy feature, and proclaim it explicitly.
Ahem! https://www.mozilla.org/en-US/firefox/features/ > https://www.mozilla.org/en-US/firefox/features/adblocker/
Cooking up conspiracy theory instead of research is easy, is not it?
I also noticed the same trend here and elsewhere as well.