Best of PrivacyJune 2024

  1. 1
    Article
    Avatar of mozillaMozilla·2y

    Transform Firefox into the ultimate minimalist browser

    Learn how to transform Firefox into a minimalist browser that helps you stay focused and enhance productivity. Steps include setting a blank homepage, customizing the toolbar, using keyboard shortcuts, and utilizing Reader View to declutter web pages. Suitable for anyone looking to create a distraction-free browsing experience.

  2. 2
    Article
    Avatar of mlnewsMachine Learning News·2y

    Perplexica: The Open-Source Solution Replicating Billion Dollar Perplexity for AI Search Tools

    Perplexica is an efficient, transparent, and open-source search tool that solves the problems of inadequate search relevance and privacy issues in traditional and proprietary AI-powered search engines.

  3. 3
    Article
    Avatar of itsfossIt's Foss·2y

    Using LM Studio to Run LLMs Easily, Locally and Privately

    LM Studio allows you to run Large Language Models (LLMs) directly on your computer, ensuring privacy and control over your data. The user-friendly interface simplifies downloading and using models, such as LLaMA 2, without lengthy terminal commands. Ideal for users with powerful systems, the application provides an alternative to cloud-based AI solutions by leveraging local hardware for AI tasks.

  4. 4
    Article
    Avatar of omgubomg! ubuntu!·2y

    Proton Pass for Linux Released (Encrypted Password Manager)

    Proton Pass for Linux and macOS has been released, expanding Proton's support for major desktop operating systems. The new app integrates with the system's authentication mechanisms and APIs, allowing for unlocking with a Linux user account, fingerprint reader, or Linux PAM-supported module. Proton Pass is an encrypted password manager with a built-in 2-factor authentication system, support for passkeys, password sharing, and email masking. It can import passwords from other managers like Bitwarden, 1Password, and KeePass. There is a free version with some limitations and a paid plan that unlocks additional features and access to Proton Monitor, which scans for data leaks on the 'dark web'.

  5. 5
    Article
    Avatar of hnHacker News·2y

    CutOffFacebook.md

    Follow these steps to reduce the impact of AI-targeted ads on Facebook, improving your psychosecurity. This involves using the browser console to automate the process of unsubscribing from advertisers. Remember not to interact with the browser during the process.

  6. 6
    Article
    Avatar of braveBrave·2y

    Bring Your Own Model (BYOM): using Brave Leo with your own LLMs

    Brave's new 'Bring Your Own Model' (BYOM) feature for their AI assistant, Leo, allows users to run local or remote AI models directly within the browser, ensuring data privacy and customization. This feature enables users to configure their models to meet specific needs while safeguarding data on-device. BYOM is available on Brave Nightly and will soon launch fully. Users can integrate popular frameworks like Ollama to run performant local models effortlessly.

  7. 7
    Article
    Avatar of freecodecampfreeCodeCamp·2y

    How to Improve Your Digital Security and Privacy – Best Practices for Developers

    Developers can improve their digital security and privacy by following five best practices: using a password manager and strong, unique credentials; choosing a security-focused browser and possibly a VPN; understanding the importance of encryption (including HTTPS and end-to-end encryption); being mindful of how digital transactions can leave traces; and selecting secure devices that are regularly updated. These steps, based on extensive research and practice, can help mitigate risks like data breaches and unauthorized access.

  8. 8
    Article
    Avatar of dailydoseofdsDaily Dose of Data Science | Avi Chawla | Substack·2y

    Introduction to Federated Learning

    Federated learning is an innovative machine learning technique that enables the training of models using private data on users' devices, without aggregating data centrally. This approach addresses privacy concerns while still allowing valuable data to be utilized. The technique distributes computations to the user's device, reducing the need for central server resources. However, it poses several challenges, such as data skewness and model aggregation. As more users value privacy, federated learning is receiving increased attention for its potential in preserving data confidentiality.