While I was fighting an unexpected cold (smart people catch colds in summer), I was disappointed by two stupidities regarding the Debian Project. One was some developers’ reaction to the results of the recent vote on LLM usage in Debian, and another concerned censorship that I found about a year too late.

Debian versus AI

I mentioned en passant the second vote within the Debian Project for 2026, General Resolution: LLM usage in Debian, through some links in an older post, specifically linking to posts by Aigarius, Iustin Pop, Raphaël Hertzog, and Anarcat.

Meanwhile, the vote results have been announced, the short version being that the winner was option 5: “Responsible Use of Generative AI.”

The long version is that voters didn’t choose one option among nine in a simple first-past-the-post vote; instead, they used a specific type of ranked voting described in Debian’s Constitution at “A.5. Vote Counting.” It uses the Condorcet method adjusted with the cloneproof Schwartz sequential dropping, aka the Schulze method, to select the candidate so that the Condorcet winner is indeed the majority-preferred winner once all cyclic ties are considered and all indirect victories are computed (e.g., if Alice beats Bob, and Bob beats Charlie, then Alice beats Charlie).

Jochen Voss briefly explained it, but this voting system is so complex that I couldn’t be bothered to try to understand it. For instance, I don’t know how it differs from ranked-choice voting, which the Marquis de Condorcet also discussed for the first time. What I appreciate about all ranked-voting systems isn’t that they try to select the person or option most people like when most people’s number one choice can’t win because there are more than two candidates, but that they select the candidate or option who is least hated across all voters!

That makes Debian the most intelligently organized and the most democratic organization on planet Earth. Unfortunately, Debian’s developers are only human, therefore rather stupid, especially as they are typically software developers by education. Nolite mittere margaritas ante porcos.

Specifically, right after the results were announced, Lucas Nussbaum, a former Debian Project Leader, expressed his regret that “the winning option does not acknowledge all the legitimate concerns around AI: environmental impact, copyright and licensing uncertainty, the health of Free Software communities, and the strain of aggressive scraping on the open web.”

He also produced this incomprehensible graph (I told you I never tried to understand Debian’s magic voting system!) to prove that “the community is split 64%/36% between the winning option and the most popular «ban» option (Debian is created by humans). Or 61%/39% with the most popular «discourage» option.” Whatever.

But at least one developer already left the project in protest, namely Antoine Le Gonidec:

Initially sent to debian-private, shared here because I want the consequences of Debian fully embracing fascist-backed technology to be known.

I do not expect any other Debian Developer to follow my example, that the most pro-LLM of all options won the last vote tells me all I need to know about the people I used to work with.

Do not CC me if you answer to that e-mail, I am no longer interested in anything coming from Debian.

I can not support the current decision of Debian about LLM use, and am no longer willing to be seen as a part of Debian under these new rules. Pretending to have a “neutral stance” when faced with fascism is not neutrality, it’s active collaboration.

Fuck that, I’m out, I don’t really care if it’s through “emeritus” status or erasure from databases. You can replace me with some agentic AI anyway, so do not even try to pretend you’re going to miss me.

I don’t think I have ever been so disappointed by a community I used to trust.

Antoine Le Gonidec (via nm.debian.org)

For details and to comment, visit https://nm.debian.org/process/1571/

Some people should be in a mental facility. This crétin seems to have missed the memo: even Linus Torvalds in person and the Linux kernel teams are using LLMs!

Censoring fortune cookies

As I was using Debian, with /usr/games/fortune added to .bashrc, what struck me was that I was never offered any sensitive or offensive fortune cookie! WTF?

At first, I thought I didn’t install all the fortunes, and it was true: I only had the minimal set. In my defense, the way the various packages are named is completely retarded:

  • fortune: virtual package provided by fortune-mod
  • fortune-mod: provides fortune cookies on demand
    🟦 recommends: fortunes-min
    🟢 suggests: fortunes
  • fortunes: Data files containing fortune cookies
    🔴 depends: fortunes-min
    🟦 recommends: fortune-mod
  • fortunes-min: Data files containing selected fortune cookies
    🟦 recommends: fortune-mod
  • fortunes-off: Data files containing offensive fortune cookies
    🔴 depends: fortune
    🟦 recommends: fortune-mod

I’ve never seen something so fucked-up for such a trivial case!

  • The actual program is in fortune-mod, but it includes zero fortune cookies, and it has no dependencies. It recommends a minimal set, and it suggests the full set, but it only works because, by default, apt installs the recommended packages as if they were hard dependencies; therefore, fortunes-min is automatically pulled in. But relying on recommended being treated as dependencies is not how things should be done!
  • As I only installed fortune, this resolved to fortune-mod, which only pulled fortunes-min. 🤦‍♂️
  • I should have installed fortunes (the large set), which would have pulled fortunes-min (the small set) as a hard dependency, and the program itself, fortune-mod, as a soft dependency! (Really, how retarded can some people be?)
  • On a system that only installs the hard dependencies, because those are the real dependencies (in red):
    • installing the program (fortune-mod) results in a binary with no data!
    • installing the large set (fortune) results in a complete set (including fortunes-min), but no binary to display them!
    • installing the small set (fortunes-min) results in a small set with no program to display them.
  • Only fortunes-off was “futureproof” (i.e., meant to survive a change of apt’s defaults): it required fortune, which was provided by fortune-mod (the binary). Inexplicably, it also recommended (cough) the same fortune-mod!

So they have that Condorcet-cum-Schwartz-aka-Schulze shit, but they create such illogical dependencies for no sensible reason! In contrast:

  • Red Hat’s dnf also uses hard dependencies and soft dependencies, but Fedora’s fortune-mod package includes all the fortunes offered by their project! However, they include only the original English-language ones and strip out the offensive ones, which you need to get from the peterww/fortune-mod-offensive COPR.
  • In the good old times of Slackware, all you had to install was bsdgames.tgz, which included four sets of fortunes: fortunes, fortunes-o (the offensive ones, for some reason ROT-13 sort-of-coded), startrek, and zippy (who the fuck was zippy?). OK, today Slackware includes in bsd-games several other data files: farber, fortunes2, fortunes2-o, limerick, limerick-o, linuxcookie, and unamerican-o. Some 20,600 fortune cookies in all, compared to less than 4,700 originally!

Well, except that the really bad news is that the last time Debian offered fortunes-off was in Debian 11 Bullseye (oldoldstable)! 😡

We’re no longer living in a free world.

Fortunately, you can still download the Debian 11 build of fortunes-off (say, fortunes-off_1.99.1-7.1_all.deb) and install it in any version of Debian! I looked into fortune-mod_1.99.1.orig.tar.gz , and I found 21,744 fortune cookies that you’re not supposed to see!

Occurrences of some words (full words, not substrings) or collocations: die: 230, death: 255, dying: 55, suicide: 26, suicidal: 1, fuck: 133, cunt: 70, cum: 5, jerk off: 2, my dick: 6, her pussy: 1, eat pussy: 3, suck my cock: 1, suck my dick: 1.

Well, how could this censorship happen? LWN, August 4, 2025: Debian grapples with offensive packages, again:

The packages that are being removed from trixie—fortunes-it-off and fortunes-scn-off—contain the Italian and Sicilian (respectively) offensive fortunes. “Offensive” here is not an editorial judgment call—the packages are officially designated as offensive in the package names and are maintained separately to ensure that users do not install them accidentally. These packages are not pulled in by default when a user installs fortune-mod; a person would have to install them separately. Even if the -off packages are installed, fortune only chooses from the database of offensive fortunes if one uses the -o option. Debian’s man page for fortune includes this guidance for that option:

Please, please, please request a potentially offensive fortune if and only if you believe, deep in your heart, that you are willing to be offended. (And that you’ll just quit using -o rather than give us grief about it, okay?)

Debian split the offensive fortunes in English off into a separate fortunes-off package in 1997. That package was dropped entirely ahead of the Debian 12 (“bookworm”) release, following a lengthy discussion on the debian-project mailing list. That discussion was started by Debian community team member Andrew M.A. Cater after the team had received a query about the fortunes-off package. Someone had asked if the package contained any Nazi quotes, which “might make it illegal to host the content on mirrors in at least Germany or Austria“. He said that he thought any such quotes had been purged from the original BSD package that served as a basis for Debian’s fortune packages.

However, he raised the question of whether Debian should remove fortunes-off since it contains quotes that “probably don’t fit in with our Debian values or general societal values 25 years on“. Ultimately, after much debate, it was removed.

But that was only the English version of the package. The fortunes-it-off package was still included in bookworm when it shipped in June, 2023. Salvo Tomaselli took over as maintainer of fortunes-it and fortunes-it-off in November 2023, and introduced the fortunes-scn and fortunes-scn-off packages that month as well. Note that the scn binary packages are built from the same source package (fortunes-it). The scn binary package was added on November 30, 2023.

On July 12, Cater filed bugs against fortunes-scn-off and fortunes-it-off. Cater’s bug reports simply said that the English fortunes had been removed, and the subjects for the bugs stated that the packages had “no place in Debian“.

For fuck’s sake!

🗒️ Note to self: I should add /usr/games/fortune -o to .bashrc. Minus-oh.

The 130 comments to that LWN article deserve a cursory reading. Because the entire situation is grotesque.

I’ll end this post here, for two reasons:

  1. I’m sweating from this cold I’m still having.
  2. This piece of shit of Firefox 154 is doing this to me:

Censoring, censoring, censoring. Removing, removing, removing. Maybe this is the universal fix. If I only had 8 GB of RAM and no swap, Firefox couldn’t have used 10 GB, right?

Either way, there aren’t enough references to suicide in fortunes-off. Although a nuclear war would be much more effective.