Your digital life isn’t yours: The hidden battle for software freedom  fsf.org

I am very sympathetic to free software. (And I regret not using more software of this kind.) On the Free Software Foundation blog, Jason Self reinforces the importance of the four freedoms of FOSS in the face of machine learning — which, in this context, is confused with what is commonly referred to as “artificial intelligence.” He defines it as follows:

[…] software that doesn’t just follow instructions, but learns and makes autonomous decisions. It’s a powerful new kind of code, and it has become the most profound black box ever created.

His post uses AI as a threat to revisit the foundations of the movement. This is always a good thing and, from time to time (as in this case), reveals stories unknown to the public (or to me, at least). It is because of one of these — the creation of the concept of free software — that I brought this link here:

At MIT’s Artificial Intelligence Lab, a programmer named Richard Stallman grew frustrated with a new Xerox laser printer that frequently jammed. His solution was simple: modify the program to automatically notify users on the network about the jam, saving everyone time and frustration. The problem was that he wasn’t allowed to; the source code of the program was a secret. Though a programmer at another university had the code, he was bound by a non-disclosure agreement and refused to share it. This wasn’t just an inconvenience; it was an ethical crisis in miniature. A practical problem had become impossible to solve, not for technical reasons, and most definitely not because it was better this way. A barrier was intentionally placed to deny users control over the software they used.

This moment of frustration ignited the spark for the free software movement.

The next time my printer jams, I will handle the situation with a little more enthusiasm. Stretching it a bit, it has a sacred quality, as it reproduces the moment of the creation of free software. Amen!

Home Assistant Voice Preview Edition — the open alternative to Alexa and Siri for controlling smart homes

by James Pond

When it comes to smart speakers, Amazon has Alexa, Apple has the HomePod, and Google has Nest. If you want something private — that runs locally — to control your home, there weren’t many alternatives.

Or there weren’t until now. To fill that gap, Nabu Casa, the sponsor of the Home Assistant open source project, released the Home Assistant Voice Preview Edition.

I bought six of these to replace six HomePods I had scattered around the house. After using them for a while, the question is: can you trust this for everyday use, or is it better to wait for a release without “preview” in the name?

(more…)

How should open source software projects handle AI‑generated code?

The excellent KeePassXC, an offline, open‑source password manager, is at the center of a controversy over the use of AI‑generated code after the project’s collaboration policy and README added this paragraph:

Generative AI is fast becoming a first-party feature in most development environments, including GitHub itself. If the majority of a code submission is made using Generative AI (e.g., agent-based or vibe coding) then we will document that in the pull request. All code submissions go through a rigorous review process regardless of the development workflow or submitter.

Users and critics backlash was so intense that on Sunday (9the) one of the project maintainers, Janek Bevendorff, published a post on the official blog detailing their stance on AI‑generated code.

(more…)

Tech companies are finding out everything is political

Framework, which makes and sells modular, repairable computers, is facing a small uprising on its official forum after announcing sponsorships of the Hyprland and Omarchy projects — a Linux window manager and a pseudo‑distro based on Arch, respectively.

(more…)

Command Palette is better than the Start menu

Microsoft PowerToys logo.

I understand that tradition and the power of branding carry a lot of weight, which explains the uproar surrounding any change involving the Start menu in Windows.

What I don’t understand is why Microsoft is sidelining a tool that seems so cool, like the Command Palette, the newest addition to PowerToys, a set of (open source!) utilities from Microsoft for Windows. It was released in version 0.90 at the end of March.

The Command Palette is a “launcher,” similar to Spotlight on macOS. Press Win + Alt + Space to invoke it and type what you want. (You can change the keyboard shortcut in the app settings.)

At first glance, it’s not much different from pressing the Win key and typing the name of an app or file. The Command Palette does that too. But it does so much more:

  • Execute commands (using the > command).
  • Switch between open windows.
  • Perform calculations.
  • Access websites or conduct web searches.
  • Run system commands.

Another cool feature is that it is extensible. The Command Palette itself has an “extension creator” based on a form. Those familiar with coding can create with more precision. Not tech-savvy? You can search for and install extensions.

Here on the other side, in macOS, I never use the closest thing Apple offers to the Start menu, the Launchpad. (Or is it the Dock?) I always use Spotlight instead, and unless when I occasionally forget the name of a rarely used app, it’s the fastest way to open any app.

Is my behavior weird, or uncommon? Do people really open the Start menu (or the Launchpad), find the app icon they want, and click it with the mouse?

Anyway, if you’re using Windows: Command Palette. That’s the way to go. Oh, and the app is free.

End of 10: Replace Windows 10 with a Linux distro

Support for Windows 10 will end on October 14th, 2025, just a few months away. A group involved with Linux distros has launched the website End of 10 to assist those who want to switch from Windows to Linux, instead of following Microsoft’s guidance to discard a functional computer and buy a new one with Windows 11. End of 10 gathers instructions and information about locations and events where volunteers install a Linux distro on computers for those who are not familiar with the process.

GNU nano’s two instances of naming by analogy

The folks in the free software community have a knack for coming up with clever names for their creations. Just look at GNU (GNU’s Not Unix) and Wine (Wine Is Not an Emulator) as prime examples.

On Mastodon, Simon Tatham shared the story of nano and its double-meaning naming.

The GNU nano text editor is named by analogy, after an earlier (non-Free) editor with a very similar UI, called pico. The name puns on SI prefixes: “like pico, but a bit bigger.”

pico was derived from the email client Pine: it’s the built-in editor Pine used for composing emails, pulled out and turned into a standalone tool. Short for PIne COmposer, as far as I know.

And Pine was also named by analogy, after an earlier email client called Elm.

So nano has two instances of “name a program by analogy to a previous one” in the history of how it got its name. (Not counting the step in between where pine gave rise to pico, because that wasn’t by analogy.)

Can anyone think of a longer chain than that, involving three or more generations of naming-by-analogy? Or is nano the record holder?

In the replies, they also mentioned Micro, another editor that aims to be a bit more feature-rich than GNU nano.

Fedora Linux 42

The stable version of Fedora Linux 42 is now available, featuring Gnome 48 in the Workstation edition and a new edition based on KDE Plasma (6.3.4), which has been promoted to the same status as Workstation in this cycle. Despite sharing the same status, the naming logic is different; the team is aware of the confusion and states that “we’ll get that figured out eventually.”

Anaconda, the Fedora installer, has received a significant update that makes the automatic disk partitioner smarter, adds the option to reinstall the system, and improves handling of dual boot. For now, it’s on by default only in Fedora Workstation (the edition with Gnome).

Oh, and a last-minute bug slipped through:

[…] just booting the Live media adds an unexpected entry to the UEFI boot loader even when Fedora Linux 42 is not installed to the local system.

The issue is purely cosmetic, but it’s good to be aware of it in advance. Here are the instructions on how to remove the entry.

Pinta 3.0

Pinta's logo: a brush inclined next to a tube of paint.

Pinta 3.0 is out. The new version of the lightweight image editor is built on the GTK 4 and Libadwaita, a much-welcomed modernization of the app.

While this alone brings a host of given improvements to Pinta, it’s not the only new feature. There are visible changes (new icons, menu, color picker, and smart layers) as well as under-the-hood enhancements (dynamic adjustments for different screen sizes and orientations, better gesture support, increased speed, and hopefully fewer crashes).

Support for add-ins, which was temporarily removed in the 2.x series, is back. For now, only two have been ported, but the developers say that “more are likely to be ported to the new release and future releases.”

Pinta’s origins trace back to being a multiplatform Windows’ Paint.NET alternative, meaning it aims to be a simple yet capable image editor; the missing link between Paint and Photoshop. The code is open source and the app is available on Linux, macOS (now with support for Apple Silicon), OpenBSD, and Windows.

WordPress is switching from three major releases a year to just one. This new cadence kicks in for 2025 — with WordPress 6.8, scheduled for next Tuesday (April 15th), set to be the only release for the year.

The change was announced during an online meeting with core contributors on March 27th, as reported by The Repository. Some of the contributors who attended the meeting shared their frustration, feeling that Matt Mullenweg, the project’s lead, had already made up his mind to slow the release cadence ahead of the discussion.

Earlier this month, Automattic let go of 16% of its workforce — about 280 folks spread across 90 countries.

It’s worth noting that the cascade of bad news for WordPress seems to have started when Matt impulsively got himself and Automattic tangled in a legal battle with WP Engine over… well, who knows what exactly.

WordPress’s uncertain future and the promise of ClassicPress

Almost a decade ago, I launched a tech blog in Brazil called Manual do Usuário (“User’s Guide” in Portuguese). Since its inception, it is published with WordPress, one of the oldest CMS — a content management system — and by far the most popular on the web today: it’s estimated that 40% of active websites use it nowadays. WordPress is open source, works well, there’s almost nothing to complain about.

In December 2018, Automattic, the company behind WordPress, released version 5.0 with big fanfare and a radical change: Gutenberg, a new, very visual post editor based on content blocks instead of text.

Gutenberg changes the writing process a lot. If before I was presented with a text area with some formatting buttons at the top when writing some post — a kind of simplified Word —, now it was possible to manipulate the whole appearance of the content using these blocks.

This was not a very well received change. To this day, the Classic Editor plugin, which restores the Word-style editor used until WordPress 4.9, is one of the most popular on the platform, with +5 million active installations and a five-star (top) rating.

Automattic doubled down on Gutenberg in early 2022 by bringing to WordPress 6.0 a thing called Full Site Editor: now, in addition to posts, someone could design the entire site with blocks/Gutenberg. WordPress moved even further away from being a mere blog or text-based publishing tool to become… I don’t know, anything other than that.

With Gutenberg, Automattic — which, it should be mentioned, runs a commercial operation based on WordPress, WordPress.com — decided to pick a fight with DIY and more modern rivals, notably Squarespace and Wix. Not by chance: these have achieved great recognition and a lot of users (and money) in recent years, because they are easier to handle for non-programmers.

And it is indeed easier to make a custom site with Gutenberg, but at what cost? For me (a person who can’t code, but can deal with simple HTML and CSS, by the way), the biggest hurdles are the added complexity when writing anything with blocks and the “dirty” code Gutenberg generates when displaying the site to visitors. (I care a lot about this “invisible” part of the site. I’m not the only one).

WordPress’ new direction alienates a significant portion of its user base. At the very least, those 5 million who use Classic Editor by this day. Maybe we aren’t the most profitable users, but we’re a crowd that, in many cases, has relied on this tool for a very long time to earn our living or just to maintain sites that are doing just fine without Gutenberg, thank you. This is my case: Manual do Usuário has been around for almost a decade.

At the moment, WordPress meets the needs of a site like mine because it is still possible to neutralize much of the excesses that Gutenberg brings to the system using a lot of workarounds in functions.php. Until when? I don’t know.

All WordPress development is dictated by Gutenberg, both within Automattic and in the ecosystem, by third-party developers of plugins, themes, and solutions. This creates apprehension in those who don’t get along with the blocks and would rather do without them. WordPress community support has always been stellar, but it started to fade into something sparse for those out of the blocks train.

The Classic Editor, for example, was supposed to be discontinued at the end of 2021. It got an extra year of support due to its popularity. At the end of 2022, will it be abandoned? I don’t know.

Even a simple site like Manual do Usuário has several dependencies with the chosen CMS. After all, it’s a huge archive that was published on the features, limitations and possibilities of WordPress. Migrating to another tool is always an option, not infrequently a traumatic one that leaves after-effects.

That’s why I’ve been looking fondly at ClassicPress. In 2019, shortly after WordPress 5.0 was released, a group of developers decided to stay in version 4.9, forking the main WordPress into something new. ClassicPress was born.

In three years, however, progress has been slow. Making matters worse, the bureaucratic part and the internal dramas of ClassicPress’ project continue to distract everyone from what matters, from writing code.

At the end of June, the two developers leading the ClassicPress Initiative, the non-profit company responsible for the project, left under heavy criticism. A new group took over with the mission to regain enthusiasm and move the project forward.

It’s not an easy job. Automattic’s structure (and money) are on another scale of magnitude. ClassicPress Initiative is still counting the pennies to pay operating expenses. On exit, the former directors said there was USD 352 left in the company’s bank account.

Even in this not-so-promising scenario, it would be great if ClassicPress thrived. The new management has opened a crowdfunding initiative to cover expenses. Manual do Usuário, in my capacity, has become an early supporter.

It is not yet time to migrate my site to ClassicPress, however. The project is too raw for my needs and current dependencies, and ClassicPress new board still has to figure out fundamental issues, such as deciding to maintain compatibility with WordPress plugins or going for a complete break.

One day, if things go well, I’ll migrate. My fear, however, is that that day will come before rough edges are polished, when WordPress becomes something incompatible with Manual do Usuário, with what it was at the beginning until the fateful version 5.0 at the end of 2018.