Opensource

Key features

• Fully open model: open weights + open data + full training details including all data and training recipes
• Massively Multilingual: 1811 natively supported languages
• Compliant Apertus is trained while respecting opt-out consent of data owners (even retrospectivey), and avoiding memorization of training data

Bear Blog, a minimalist blogging platform focused on privacy and speed, has shifted from an MIT open source license to a source-available model in September 2025[^4]. According to creator Herman Martinus, the change restricts commercial exploitation while maintaining code accessibility for non-commercial use and security audits.

The new license prohibits for-profit hosting of Bear Blog or derivative services, while preserving the platform's commitment to "no-nonsense, super fast blogging" with no tracking or ads[^4]. This move mirrors similar licensing changes by companies like Elastic, which adopted a source-available model to protect against commercial exploitation while keeping code visible[^5].

"The original MIT license was selected without deep forethought, primarily to make the code easily auditable," explained Martinus on his blog[^4]. The shift aims to ensure Bear Blog's sustainability through its hosted version's modest subscriptions while preventing "open-source rug pulls" by larger corporations.

[^4]: WebProNews - Bear Blog Shifts to Source-Available License for Indie Protection [^5]: Elastic - License Change Clarification

cross-posted from: https://feditown.com/post/1813060

A baba yaga license!

Lobsters discussion where I found this

Putting on a hobby lawyering hat:
This line No mortal hand may twist it 'round. might technically not bind corporations, as they are different from natural people, i.e. mortals, in some jurisdictions.

Source code and details: https://git.disroot.org/Phoenix

The Book of Secret Knowledge is a popular GitHub repository that serves as a comprehensive collection of technical resources, including cheatsheets, manuals, lists, tools, and hacks[^1].

Key aspects of this repository include:

• DevOps and system administration resources, including 284 test questions for Linux system administrators[^8]
• DNS tools and diagnostics collections[^6]
• Command line tools and one-liners[^10]
• Cloud-native materials and training resources

The repository has gained significant community traction, accumulating over 100,000 stars and 12,000 forks on GitHub[^2]. It's particularly valued by DevOps engineers and system administrators as a free, open-source learning resource that provides alternatives to paid courses[^2].

[^1]: GitHub - trimstray/the-book-of-secret-knowledge [^2]: LinkedIn - Abhishek Veeramalla's Post [^6]: Hacker News - The Book of Secret Knowledge [^8]: Medium - 7 GitHub Repositories That Help You Become A Better DevOps Engineer [^10]: Zorin Forum - Linux commandline cheat sheet

Managarr v0.6.0 has been released with some fun new features!

Managarr is a terminal-based application for managing all your Servarr instances from one place. It provides a user-friendly interface to interact with your media libraries, making it easier to manage your downloads, monitor your series and movies, and perform various actions directly from the terminal.

It sports two modes: a TUI mode (Text User Interface) and a CLI mode (Command Line Interface).

TUI mode gives you an interactive User Interface right inside your terminal window, allowing you to navigate through your Sonarr and Radarr libraries, view details about your series and movies, and perform actions like adding or removing items, all through keyboard shortcuts.

CLI mode lets you execute commands directly from the terminal to manage your Servarr instances without needing to open the TUI. This is great for quick tasks or for integrating with scripts and automation tools.

The biggest change: Managarr now has themes!

The UI has been completely overhauled to support themes! You can now customize the look and feel of Managarr to suit your preferences. Choose from a variety of themes to change the color scheme and overall aesthetic of the application.

Here's just a few examples:

Default

Dracula

Watermelon Dark

You can also customize the themes to your heart's content! Check out the themes documentation for more details on how to create and apply your own themes.

Features

• Added support for alternative Vim-like navigation keybindings (hjkl movements) Discussion #34
• Added support for terminal-like backspace operations (Ctrl-h instead of Backspace)
• You can now specify the number of downloads to fetch from the CLI: managarr <sonarr/radarr> list downloads --count 1234
• You can now toggle movie monitoring from the CLI without needing to use the edit subcommand: managarr radarr toggle-movie-monitoring --movie-id 1234 #43
• You can also now toggle series monitoring from the CLI without needing to use the edit subcommand: managarr sonarr toggle-series-monitoring --series-id 1234 #43
• You can now also toggle movie/series monitoring directly from the Library view for each Servarr with the m key. No need to open the Edit [Series/Movie] modal anymore to simply toggle monitoring for an item! #43
• Users can now skip up/down tables 20 items at a time using Ctrl-d and Ctrl-u keys (mirroring the same functionality in the Helix editor). Alternatively, the standard PgUp and PgDown keys are supported for the same operation. This is particularly useful for large libraries with many items #45
• The total disk usage for any given series is now displayed in the Series Library view to mirror Radarr functionality #44
• All keybindings and help tips have been refactored into a unified, dynamic menu that displays the available keybindings for the current view. This is accessible by pressing ? in any view, and it will display the keybindings relevant to that view. #32
• Users can now add any number of custom headers to each Servarr's configuration, enabling support for OAuth and other custom authentication schemes for Servarr access #47

Fixes

• Fixed a bug that caused the Collection Details modal to vanish when attempting to add a new film to a collection
• Fixed a bug that caused the Radarr library to be rendered, then the Collections table to be rendered over it (merging the two), and then showing a popup which made for ugly and confusing UI
• Wrapped Season.statistics with Option to prevent a panic if the season doesn't have any statistics (edge-case, only happens with outdated Sonarr data) #35
• Corrected a bug that caused double key presses on Windows machines #40 (Thanks @cwesleys!)
• Defaulted to empty tags to improve fault tolerance within the Sonarr and Radarr UIs. This is in response to #42, #48. It seems like this may be a bug in Sonarr where a series can have an associated tag ID but that tag Id doesn't exist in the list of tags, but I still can't quite track it down.
• Fixed an issue that caused some panics to occur when video codecs are undefined in file metadata #38
• More than 10 downloads will be listed in the Downloads tabs for both Radarr and Sonarr
• Fixed a bug where Sonarr would have empty values on season releases for seeders/leechers instead of '0'
• Fixed a bug where some Radarr films don't have studios associated with them, so the studio field is now nullable, preventing crashes when loading the Radarr library

Security Fixes

• Upgraded to the most recent version of Tokio to mitigate CWE-664 Improper Control of a Resource Through its Lifetime
• Updated to the most recent patch of OpenSSL to mitigate CWE-416 Use-After-Free

Minor Changes

• Due to the new support for Vim-like navigation keybindings, the system logs are now opened using L instead of l
• Refactored the network module to be more idiomatic Rust and to improve maintainability

Documentation

• Update README.md to remove the cheeky Try Before You Buy heading since some users reported it as misleading; i.e. they thought it meant Managarr cost money. Managarr is and always will be, free

As always, thank you to everyone who reported an issue or requested a feature! You all make it a LOT easier to keep up with breaking API and add new features. If you have any feedback or suggestions, please don't hesitate to open an issue or discussion on the GitHub repository.

Jambi’s mission is to transcribe audio to your clipboard, as quickly and accurately as possible, while staying privacy-focused and open-source.

Jambi aims to help computer users with disabilities, such as vision or physical impairments, by providing real-time transcription of their speech. It’s also a great tool for anyone who wants to transcribe audio quickly and easily.

This is the alpha release and the project is still in early development. Currently looking for feedback and contributors. If you are a developer, you can contribute to the project by submitting pull requests or reporting issues.

If you like the project, please show your support by leaving a star. Thanks! https://github.com/guttermonk/jambi

OC text by developer @Transhumanist@lemmy.ml

Hi everyone!

It’s me again)

For those who haven’t heard of it yet - VOID is a local-first, open-source “second brain” app. Think of it as combining the flexibility of Obsidian with the structured organization of Notion.

As the title says - VOID just reached 100 stars on GitHub!

Huge thanks for all your support - it truly motivates me to work even harder and bring the project to release as soon as possible.

But this post isn’t only about “showing off.” Over the past week I’ve mainly been polishing the editor’s behavior, and now I’m finally happy with it. I think it’s time to share the roadmap for the next year(I think?):

• Note linking, global indexing of todos and tags, and embedded content (videos, images, audio, etc.)

• Migration from Excalidraw canvas to a custom JSON Canvas implementation

• Databases (similar to Notion)

• Sync server improvements and a self-hosted Docker container

• Plugin API (Lua for the backend and any frontend framework you can imagine)

• Full redesign of the entire app (yes, we finally have a designer!)

• And of course… optimization, optimization, optimization

We also now have a Lemmy community where I’ll post small devlogs, thoughts, and some behind-the-scenes updates about VOID. If you’re interested in the project, feel free to join there — or hop into our GitHub Discussions.

GitHub
And one more thing - the first beta is coming very soon!
Stay tuned!

https://conference.libreoffice.org/2025/

I’m a firefighter who’s also a software engineer and am working on a training app. In it, I have it generating text dispatches of various scenes for us to discuss on rainy slow training days. Such as “Respond to 123 Main st, for a report of a smell of smoke” etc.

I already use google maps to generate a random address and show the map / street view. With the maps api I can domain lock the key…

But with their Text to Speech api I cannot. Seems silly. But I get it.

Are there any alternatives? I would be ok spinning up a middle server to also rework the audio to generate radio static, etc, but first pass I am looking for non-robotic (ie not browser based) TTS.

Thoughts?