We Make Fedora

Urgent : vérification du bien-être, appelez maintenant pour l’équipe féminine de football iranienne

2026-03-08T12:00:00Z

Comme cela a été largement médiatisé dans le monde entier, l'équipe féminine de football iranienne est actuellement en tournée en Australie .

Leur dernier match a lieu aujourd'hui, Journée internationale des femmes, après quoi elles seront placées dans un avion pour retourner en Iran .

Lors de leur premier match, ils ont refusé de chanter l'hymne national.

Lors de leur deuxième match, ils ont salué et chanté l'hymne national.

De nombreux experts ont constaté que ce changement de comportement était dû à un contrôle coercitif.

L'État du Queensland a été le premier État australien à légiférer sur le contrôle coercitif. La police a l'obligation de faire respecter la loi. La finale féminine se déroule au Queensland.

L'équipe de sécurité venue d'Iran loge à proximité des femmes dans le même hôtel. Selon la loi du Queensland, il s'agit d'une relation domestique. Les réservations d'hôtel et les accompagnateurs sont organisés pour les femmes. Elles ne peuvent se déplacer seules. Par conséquent, il s'agit d'une forme d'interdépendance domestique et la loi s'applique à chacune d'elles.

Lorsqu'elle est alertée d'une situation familiale, la police a l'obligation d'intervenir. À tout le moins, elle est tenue de vérifier le bien-être des personnes concernées. Elle n'interviendra que si ces personnes le lui demandent.

La télévision d'État iranienne a clairement affirmé que ces femmes étaient des « traîtresses de guerre ». Dans la plupart des cas de répression, les preuves de telles menaces sont dissimulées au public. Dans ce cas précis, les femmes qui s'expriment depuis l'étranger sont menacées de poursuites pénales pour trahison.

En 2013, le gouvernement travailliste australien de l'époque a publié une vidéo humiliant des demandeuses d'asile iraniennes .

J'ai adhéré au Parti travailliste australien (ALP) en 1997 et, lorsque j'ai vu la vidéo humiliant des femmes iraniennes , j'ai immédiatement démissionné. Ma lettre de démission a fuité et a été publiée par Crikey . Vous trouverez ci-dessous une copie intégrale du courriel.

Mon courriel de démission est soudainement devenu une preuve cruciale que le gouvernement australien cherche à dissuader les femmes iraniennes de demander l' asile politique . De ce fait, il est d'autant plus urgent que toute personne concernée contacte la police locale et lui demande de vérifier le bien-être de ces femmes.

Si vous pensez qu'une femme dans le Queensland est en danger de mort immédiat, composez le 000. Ce numéro fonctionne uniquement en Australie et ne doit être utilisé qu'en cas d'urgence.

Pour toute autre demande de vérification de bien-être, vous pouvez vous rendre dans un poste de police local ou appeler la police du Queensland au +61730556206 (à partir du site Web ).

Lorsqu'on s'adresse à la police, il est important de souligner les points suivants :

Les femmes se trouvent actuellement sous la juridiction de la police du Queensland.
Ils seront contraints de quitter le territoire dans les prochaines heures.
Ces femmes sont soumises à un arrangement domestique dans le cadre de leur réservation de voyage de groupe.
Ces femmes sont accompagnées partout où elles vont par un groupe d'hommes.
Leur refus de chanter l'hymne national lors du premier match est un signe qu'ils expriment silencieusement leur mécontentement et demandent de l'aide.
Leur changement de comportement, avec le salut national et le chant de l'hymne lors de leur deuxième match, indique qu'ils étaient soumis à des pressions pour les contraindre à retirer leur plainte, qu'ils avaient formulée par le silence.
Ces femmes ont fait l'objet d'une menace directe, diffusée sur la télévision d'État iranienne, laissant entendre qu'elles seraient persécutées pour trahison.
La vidéo produite par le gouvernement australien en juillet 2013 vise à exercer un contrôle coercitif, en dissuadant les femmes iraniennes d'exercer leurs droits en Australie, sous la menace d'humiliation publique.
Lors de la Coupe du monde féminine de la FIFA organisée en Australie, les autorités fédérales n'ont pas réagi au comportement du président espagnol du football, Luis Rubiales , malgré le fait que ses agissements aient été vus par tous à la télévision nationale. Il a pu prendre un avion et quitter l'Australie sans qu'aucune enquête ni poursuite ne soit engagée.
Il est essentiel que la police du Queensland ouvre un dossier officiel sur le risque encouru par chaque femme du groupe.
Il est essentiel que la police du Queensland effectue un contrôle de bien-être auprès de chaque femme du groupe.

Il n'y a pas si longtemps, en Suisse , j'ai rencontré le patron d'une des fédérations sportives iraniennes lors des Jeux olympiques d'hiver de la jeunesse :

Voici le courriel de démission lorsque le gouvernement du Parti travailliste australien (ALP) a humilié des femmes iraniennes avec une vidéo en juillet 2013 :

Sujet : Politique du Parti travailliste australien, fierté et horreurs du viol
Date : mardi 23 juillet 2013 23:44:17 +0200 (CEST)
De : Daniel Pocock
Réponse à : Travailliste victorien
À : [tous les membres du Parti travailliste victorien]

Il n'y a pas si longtemps, la branche étatique a pris l'initiative d'imprimer
« Fier membre du syndicat » sur les cartes de membre. Quand vous avez vu la vidéo dégradante
Le témoignage d'une réfugiée pauvre et vulnérable a été diffusé dans le monde entier sur
Lundi, la fierté était-elle l'émotion que vous ressentiez dans votre cœur ? Si oui, vous
Inutile de lire la suite.

Je suis membre du Parti travailliste australien (ALP) depuis 15 ans. Je n'ai jamais eu autant honte.
que maintenant, quand j'ai vu les informations concernant notre gouvernement travailliste qui met les réfugiés à l'écart
remis en détention aux mains de ceux qui les avaient violés.
C'est bien pire que les accusations portées contre l'Église catholique.

Le Parti travailliste australien (ALP) est au pouvoir depuis six ans. Il est absurde de penser…
Leurs programmes d'espionnage conjoints américano-australiens peuvent surveiller les déplacements des suspects.
à l'autre bout du monde, au Moyen-Orient, mais ils n'en ont pas conscience.
lorsque des viols ont lieu sous la garde et la protection de notre propre nation
programme de réfugiés. Si votre député n'était pas personnellement au courant de ces choses,
Alors, quel genre de travail fait-il exactement ? Aucun programme de détention n'est prévu.
L'histoire de l'humanité s'est jamais terminée de façon moins tragique.
que ce que nous voyons actuellement.

Si votre député est (ou était) membre du cabinet, la seule question est de savoir comment
Que savaient-ils vraiment et depuis combien de temps nous trompent-ils ?
Le reste du pays n'en a pas parlé. Ils ont fait obstruction aux médias et aux mineurs
Les partis politiques sont tenus responsables de la surveillance de ces camps de concentration afin d'empêcher leur examen approfondi.
chacun d'entre nous dans l'obscurité.

Plusieurs de ces individus sans scrupules prévoient de prendre leur retraite avec une généreuse pension.
La pension sera versée dans quelques semaines seulement, tandis que ces malheureux humains
Les êtres qui se sont placés sous leur protection en portent encore les stigmates.
de l'automutilation et du viol pour le restant de leurs jours. Quel contraste.
Les valeurs syndicales transparaissent.

Lorsque la campagne électorale fédérale débutera officiellement, qui sera en lice ?
Je me sentirai vraiment à l'aise en portant un t-shirt ALP.
Vous vous demandez à quoi ressembleront les prochaines révélations d'abus ?
Devons-nous attendre que M. Assange ou M. Snowden contestent la situation
les excès de notre gouvernement mettent leur propre vie en danger ou font
Avons-nous le courage d'en parler, chacun d'entre nous ?

J'ai commencé à inclure certaines de mes propres observations sur l'Australie
J'ai récemment constaté des problèmes sur mon blog. Vos commentaires sont les bienvenus.
Là-bas ou via votre plateforme sociale préférée. Il est temps de dire à votre
Le député et le monde entier ont compris que les êtres humains ne sont pas un sujet de négociations, de débats,
échanges et viols.

http://danielpocock.com/if-assange-is-a-rapist-what-kind-of-monster-runs-australia

Quand des personnes sont victimes de viol et de maltraitance, ce n'est pas le moment d'être patient ou complaisant.
ou apathiques. Ce n'est pas le moment pour une autre excuse lâche concernant les pendus.
Le parlement ou l'absence de contrôle absolu au Sénat. Si vous êtes venu
Adrian Bayley violant une autre femme dans une ruelle, la manière dont il l'a sauvagement fait
Jill Meagher, pourriez-vous simplement passer votre chemin ? C'est exactement ce que ce parlement attend de nous.
a fait pour les personnes dans le besoin. Si vous êtes vraiment un fier membre du Parti travailliste, vous ne le feriez pas.
pouvoir vivre avec cette honte.

---------------------------------------------

L'employé qui a dénoncé ces viols fait actuellement l'objet de poursuites judiciaires.
pour avoir enfreint une clause de confidentialité de son contrat. Est-ce que le travail
faire le nécessaire pour garantir que les lanceurs d'alerte qui dénoncent les crimes puissent
Vous vous sentez à l'abri des persécutions ?

---------------------------------------------

Vous recevez ce courriel car vous figurez sur une liste de diffusion gérée et diffusée par la section victorienne du Parti travailliste australien (ALP). L'en-tête de réponse est configuré pour que vous puissiez répondre.
Contactez-les directement si vous souhaitez que votre adresse soit retirée de la liste.

Si vous pensez qu'une femme dans le Queensland est en danger de mort immédiat, composez le 000. Ce numéro fonctionne uniquement en Australie et ne doit être utilisé qu'en cas d'urgence.

Pour toute autre demande de vérification de bien-être, vous pouvez vous rendre dans un poste de police local ou appeler la police du Queensland au +61730556206 (à partir du site Web ).

Lors de votre entretien avec la police, il est important de souligner les points énumérés ci-dessus.

Pour en savoir plus sur l'asile politique , cliquez ici .

Urgent: welfare check, call now Iran womenâ€™s football team

2026-03-08T12:00:00Z

As publicised all around the world, the women's football team from Iran is currently touring Australia.

Their final match takes place today, International Women's Day and then they will be placed on a plane back to Iran.

At their first match, they refused to sing the national anthem.

At their second match, they saluted and sung the national anthem.

Experts have widely observed this change in behaviour was due to coercive control.

The state of Queensland was the first state in Australia to introduce laws on coercive control. The police have an obligation to uphold the law. The women's final match is in Queensland.

The security team who came from Iran are living in close proximity to the women in the same hotel. This is a domestic relationship for the purpose of Queensland law. The hotel bookings and travel companions are all organised for the women. They can not go anywhere on their own. Therefore, it is a form of domestic interdependency and the law applies to each woman in this case.

When alerted about a domestic situation, the police have an obligation to act. At the very least, police are obliged to make a welfare check. They will not do so unless people ask them.

State TV in Iran has clearly stated the women are "wartime traitors". In most cases of coercive control, the evidence of such threats is hidden from public view. In this case, there is a threat that women who speak up while outside Iran will be subject to criminal persecution under the laws for treason.

In 2013, Australia's then Labor government published a video humiliating female asylum seekers from Iran.

I joined the ALP in 1997 and when I saw the video humiliating women from Iran I resigned immediately. My resignation was leaked and published by Crikey. A full copy of the email is below.

My resignation email has suddenly become a vital piece of evidence that the Australian government wants to deter women from Iran from making claims for political asylum. On that basis, it is even more urgent for anybody concerned with the matter to contact local police and ask them to make a welfare check on these women.

If you believe there is an urgent and immediate threat to the life of any woman in Queensland then the number to dial is 000. This number only works in Australia and it should only be used for emergencies.

For any other request for a welfare check, you can visit a local police station or call Queensland police on +61730556206 (from the web site).

When speaking to the police, it is important to emphasise these points:

The women are currently in Queensland police jurisdiction
They will be forced to leave the jurisdiction within a few hours
These women are subject to a domestic arrangement by way of their group travel booking.
These women are accompanied by a group of men wherever they go.
Their refusal to sing the national anthem at the first match is a sign they are silently making a complaint and asking for help
Their change of behaviour, saluting and singing the national anthem at their second game, is a sign they were under coercive control to withdraw the complaint they made by way of silence
These women were subject to a direct threat, which was broadcast on Iran's state television, implying the women will be persecuted for treason.
The video produced by the Australian government in July 2013 is intended to have the impact of coercive control, by deterring Iranian women from exercising their rights in Australia, with the threat of public humiliation.
When the FIFA women's world cup was held in Australia, federal authorities failed to act on the behaviour of Spanish football boss Luis Rubiales, despite the fact his behaviour was seen by everybody on national TV. He was able to board a plane and leave Australia without any investigation or prosecution.
It is essential for Queensland police to open an official file on the risk to each woman in the group
It is essential for Queensland police to make a welfare check on each woman in the group

Not so long ago, in Switzerland I met the boss of one of the Iranian sporting federations during youth Winter Olympics:

Here is the email resignation when the ALP government humiliated Iranian women with a video in July 2013:

Subject: ALP Policy, Pride and the horrors of Rape
Date: Tue, 23 Jul 2013 23:44:17 +0200 (CEST)
From: Daniel Pocock 
Reply-To: Victorian Labor 
To: [ all Victorian ALP members ]

Not long ago the state branch took it upon themselves to print
"Proud Labor Member" on the membership cards.  When you saw the degrading video
of a poor, vulnerable, female refugee broadcast around the world on
Monday was pride the emotion that you felt in your heart?  If it was, you
need not bother reading any further.

I've been a member of the ALP for 15 years.  I've never been more ashamed of
that than today when I saw the news about our ALP Government putting refugees
put back into detention in the hands of people who had raped them.
This is far worse than the allegations against the Catholic Church.

The ALP has been in Government for 6 years now.  It is absurd to think
their joint US/Australia spy programs can monitor the movements of suspects
half way around the world in the middle east, but they are not aware
when rapes take place under the care and protection of our own national
refugee program.  If your MP didn't personally know about these things,
then just what kind of a job is he doing?  No program of detention in
the history of mankind has ever ended up in anything less tragic
than what we are seeing now.

If your MP is (or was) a cabinet member, the only question is how
much did they know and for how long have they been deceiving us and the
rest of the nation about it.  They obstructed the media and the minor
parties from scrutinising these concentration camps to try and keep
each and every one of us in the dark.

Several of these shameless people are planning to retire on a hefty
pension in just a few weeks from now while those unfortunate human
beings who came under their protection are suffering with the scars
of self-mutilation and rape for the rest of their lives.  What a contrast.
Labor values shining through.

When the federal election campaign officially kicks off, who is
really going to feel comfortable going out there in an ALP T-shirt
wondering what the next revelations of abuse are going to look like?
Do we have to wait for Mr Assange or Mr Snowden to challenge
the excesses of our Government at risk to their own lives or do
we have the guts to speak up about it, each and every one of us?

I've started including some of my own observations about Australia's
failings on my blog recently.  You are more than welcome to comment
there or through your preferred social platform.  It's time to tell your
MP and the world that human beings are not a subject for deals, debates,
swaps and rapes.

http://danielpocock.com/if-assange-is-a-rapist-what-kind-of-monster-runs-australia

When people are being raped and abused, it's no time to be patient, complacent
or apathetic.  It's no time for another spineless excuse about the hung
parliament or the lack of outright control in the Senate.  If you came
across Adrian Bayley raping another woman in a laneway the way he savaged
Jill Meagher, would you just walk by?  That is exactly what this parliament
has done for people in need.  If you are really a Proud Labor Member you wouldn't
be able to live with this shame.

---------------------------------------------

The worker who blew the whistle on these rapes is currently facing litigation
for breaching a confidentiality clause in his contract.  Is Labor
doing enough to ensure whiste-blowers who speak out about crime can
feel safe from persecution?

---------------------------------------------

You are receiving this email as you are on a list maintained and circulated by the Victorian branch of the ALP.  I've set the reply header so you can reply
directly to them if you want your address removed from the list.

For any other request for a welfare check, you can visit a local police station or call Queensland police on +61730556206 (from the web site).

When speaking to the police, it is important to emphasise the list of points given above.

misc fedora bits first week of march 2026

2026-03-07T17:27:53Z

Here we are in the first week of March 2026 already. This was a pretty quiet week for me, partly due to the Fedora 44 Beta freeze and partly I think due to people traveling/being away. In any case it was welcome to me to have a chance to work on some planned work instead of day to day or fighting fires.

GPU machine

This week I finally got our gpu machine all setup, which has been a very long road. Last year we thought it would be very handy to have a machine that has desktop GPUs in it that we could use to test / build / explore things that could use those. We didn't want a server with fancy datacenter gpus, we wanted things that Fedora users might have. This of course is tricky, since that entails a desktop like machine in a datacenter.

After some looking around, we found the Dell Precision 7960 Rack, which is a rackmount machine, but sort of a desktop too.

We got a loaner to test things out with, and finally decided to buy it and use it. There have been so many little delays with this thing ( wrong network card, need a new one. Time of people involved to setup the testing. Drac license was wrong and I couldn't install it, and more).

But finally this week it's up. We will see how useful it becomes and what new exciting things it opens up.

Fedora 44 Beta is GO for release next tuesday

We had our Fedora 44 Beta go/nogo meeting on thursday and amazinly we were go for release on tuesday. The second beta candidate had no accepted blockers. I'm always a bit surprised when things go so smoothly, but I will take it!

Secure boot signing

I also made some more progress on my secure boot signing setup, but then i hit a blocker. I was able to sign grub and kernel for aarch64, but it doesn't actually boot. (I have my lenovo slim7x and also another aarch64 box that supports secure boot to test with). Hopefully we can get to the bottom of that soon so we can switch things on. I really hope we can have it running before Fedora 44 final freeze.

This also has been a long road.

Just missed it

I'm getting a solar system with batteries and home backup installed late this month, and I'm really looking forward to it.

Unfortunately, my electic coop just informed me that there is going to be a 4 hour power outage on monday for maint work. If it had only been next month, I could have just ignored it. Oh well, one more time for the generator! :)

comments? additions? reactions?

As always, comment on mastodon: https://fosstodon.org/@nirik/116189196834124254

Ideal Laptop

2026-03-07T08:15:44Z

Do you remember the most important characteristics you should look for in a good laptop? In the following order:

1⃣ A high-resolution, high-density display: 3K or 4K, far beyond HD or Full HD
2⃣ A battery that lasts all day
3⃣ Fast storage (SSD)
4⃣ Light, thin, and elegant

It’s not the CPU.
It’s not AI.
It’s not having huge storage capacity.
It’s not a large physical size.
It’s not having more than 8 GB of memory (memory is not storage).
It’s not having a stylus, tablet convertible, or having a detachable or articulated keyboard.
And it’s definitely not having a numeric keypad on the side.

Until last week, the best and most affordable laptop on the market with these characteristics was the $1100 MacBook Air. But now Apple has launched the MacBook Neo, which delivers all these qualities — display, battery, storage, lightness, and elegance for 45% less: $600.

❝A laptop for me is just for browsing the internet, email, editing documents, messaging, watching movies, and relaxing with games like Solitaire or Roblox.❞

Congratulations, you’re like 99.999% of humanity. The MacBook Neo delivers the best value for you.

In the Windows laptop universe, these truly important characteristics (display quality, lightness, etc.) are usually found only in the most expensive product lines. To justify the high price, their marketing shifts the focus to things that are largely irrelevant: unnecessarily powerful CPUs, unnecessarily large storage, unnecessarily large memory, tablet modes, styluses, and so on.

All unnecessary for 99.999% of humanity.

And even in those expensive lines, the battery rarely lasts more than two hours (the reason: inefficient CPU), let alone all day.

Don’t be misled when choosing your next laptop. Pay attention to the characteristics that really matter: display, battery, fast storage, lightness, and elegance. General rule is to stay away and avoid laptops that use Intel CPUs.

This also applies to the laptops that companies give to their employees.

Laptop ideal

2026-03-07T07:56:23Z

Lembra quais são as características mais importantes que você deve procurar num bom laptop? 1️⃣ tela de alta resolução e densidade, 3K ou 4K, bem mais que HD ou full HD; 2️⃣ bateria que dura o dia todo; 3️⃣ armazenamento rápido (SSD); 4️⃣ leve, fino e elegante.

Não é CPU. Não é IA. Não é armazenamento grande. Não é ter tamanho físico grande. Não é ter mais do que 8GB de memória (memória não é armazenamento). Não é ter caneta, virar tablet, teclado removível ou articulado. E definitivamente não é ter teclado numérico lateral.

Até semana passada, o melhor e mais barato laptop do mercado com essas características era o MacBook Air de $1100. Mas agora a Apple lançou o MacBook Neo que entrega todas essas características de tela, bateria, armazenamento e leveza e elegância, 45% mais barato: $600.

“Laptop prá mim é só prá navegar na internet, e-mail, editar documentos, mandar mensagem, assistir filmes, me distrair jogando paciência e Roblox”. Parabéns, você é como 99,999% da humanidade. O MacBook Neo entrega o melhor custo-benefício para você.

No universo dos laptops Windows, essas características importantes (tela, leveza etc), só se encontram nas linhas mais caras dos fabricantes. E para justificar o alto preço, seu marketing procura mudar o foco para coisas completamente irrelevantes, como CPU desnecessariamente poderosa, armazenamento desnecessariamente alto, memória desnecessariamente grande, virar tablet, ter caneta etc. Desnecessário para 99,999% da humanidade. E mesmo assim, nem nessas linhas mais caras a bateria dura mais do que 2 horas (motivo: CPU ineficiente), quanto menos o dia todo.

Não seja ludibriado ao escolher seu próximo laptop. Preste atenção nas características que realmente importam: tela, bateria, armazenamento rápido, leveza e elegância.

Isso vale também para laptops que empresas dão a seus funcionários.

Bon Ap : Lieu de naissance des ANZAC, ancêtre de la guerre hybride, lié à un restaurant français

2026-03-06T23:30:00Z

Depuis le début de l'invasion de l'Iran , les reportages et les commentaires sur la guerre n'ont pas manqué. Je ne peux sûrement rien aborder sur mon blog qui n'ait déjà été écrit ailleurs.

En réfléchissant à ce que je ferais pour la journée de l'ANZAC cette année, j'ai ressorti quelques vieux dossiers concernant mon arrière-grand-père, le sergent Robert Ernest Edward Pocock , héros de l'ANZAC . J'ai décidé de parcourir les documents ligne par ligne pour voir s'il y avait quelque chose d'intéressant. Je n'ai pas eu besoin d'aller plus loin que le premier document, son acte de naissance, pour avoir une surprise.

J'ai vu l'adresse du lieu de naissance du sergent Pocock, le 193 Brunswick Street, à Fitzroy, sur son acte de naissance et j'ai décidé de la rechercher sur Internet pour voir ce qu'il en est advenu. En fait, c'est l'un des restaurants français les plus connus de Melbourne, le Bon Ap Petit Bistrot, à Fitzroy. J'ai réalisé un montage photo retraçant le parcours de mon arrière-grand-père, de Bon Ap à la restauration des soldats dans le nord de la France .

Alors que le sergent Pocock travaillait dans l'industrie des munitions, il est intéressant de noter que je me suis orienté vers des projets de cybersécurité modernes , en particulier ceux concernant l'ingénierie sociale .

Beaucoup trop de soldats ne sont jamais rentrés en Australie. Ils ont dû affronter l'ennemi, le froid et, pire encore, la maladie. On retrouve tous ces éléments dans le rapport concernant le sergent Robert Pocock (matricule 10768) du 3e train divisionnaire, 22e compagnie, dans le nord de la France et en Belgique .

En regardant vers le sud, le long de Brunswick Street, depuis la maison natale de Robert Pocock, on aperçoit la flèche de la cathédrale Saint-Patrick. Robert était baptiste. Son deuxième fils, Walter, se maria à l' église catholique . Son petit-fils, John, se maria dans la cathédrale même. L'arrière-arrière-petit-fils de Robert devint membre de la chorale de la cathédrale.

Le sergent Robert Pocock est au centre de la photo, avec l'insigne de grade de sergent sur sa manche.

Lisez l'histoire du sergent Robert Ernest Edward Pocock, 10768, 3e train divisionnaire, 22e compagnie, AIF sur ANZAC.site .

Bon Ap: ANZAC birthplace, Grandfather of hybrid warfare linked to French restaurant

2026-03-06T23:30:00Z

Since the invasion of Iran got under way, there has been no shortage of news reports and commentary about warfare. Surely there is nothing I can cover on my blog that hasn't been written somewhere else already?

While contemplating what to do for ANZAC day this year, I took out some of the old files about my great grandfather, the ANZAC hero Sgt Robert Ernest Edward Pocock. I decided to go through the documents line-by-line to see if there was anything interesting. I didn't have to go any further than the first document, his birth certificate, to get a surprise.

I saw the address of Sgt Pocock's birthplace, 193 Brunswick Street, Fitzroy, on his birth certificate and I decided to type it into a search engine to see what is there today. In fact, it is one of Melbourne's most well-known French restaurants, Bon Ap Petit Bistrot, Fitzroy. I put together a series of pictures telling the story of how my great-grand-father went from Bon Ap to feeding the diggers in the north of France.

While Sgt Pocock worked in the munitions industry, it is an interesting coincidence that I've gone into modern day cybersecurity projects, especially those concerning social engineering.

Far too many diggers didn't make it back to Australia. They had to battle the enemy, the cold and worst of all, disease. We see all of these things in the report about Sgt Robert Pocock (10768) of the 3rd Divisional Train, 22 Company in the north of France and Belgium.

Looking south down Brunswick Street from Robert Pocock's birthplace, we see the spire of St Patrick's Cathedral. Robert was a Baptist. His second son, Walter, married in the Catholic Church. His grandson, John, married in the Cathedral itself. Robert's great-great-grandson would become a member of the Cathedral Choir.

Sgt Robert Pocock is in the middle of the photo with the rank insignia of a sergeant on his sleeve.

Read the story of Sgt Robert Ernest Edward Pocock, 10768, 3rd Divisional Train, 22nd Company, AIF on ANZAC.site.

GNOME Foundation Update, 2026-03-06

2026-03-06T18:38:56Z

This post is the latest in my series of GNOME Foundation updates. I’m writing these in my capacity as Foundation President, where I’m busy managing a lot of what’s happening at the organisation at the moment. Each of these posts is a report on what happened over a particular period, and this post covers the current week as well as the previous one (23rd February to 6th March).

Audit time

I’ve mentioned the GNOME Foundation’s audit on numerous occassions previously. This is being conducted as a matter of routine, but it is our first full formal audit, so we have been learning a lot about what’s involved.

This week has been the audit fieldwork itself, which has been quite intense and a lot of work for everyone involved. The audit team consists of 5 people, most of whom are accountants of different grades. Our own finance team has been meeting with them three times a day since Tuesday, answering questions, doing walkthroughs of our systems, and providing additional documents as requested.

A big part of the audit is cross-referencing and checking documentation, and we have been busy responding to requests for information throughout the week. On last count, we have provided 140 documents to the auditors this week alone, on 20 different themes, including statements, receipts, contracts, invoices, sponsorship agreements, finance reports, and so on.

We’re expecting the draft audit report in about three weeks. Initial signs are good!

GUADEC 2026

Planning activity for GUADEC 2026 has continued over the past two weeks. That includes organising catering, audio visual facilities, a photographer, and sponsorship work.

Registration for the event is now open. The Call for Papers is also open and will close on 13 March – just one week away! If you would like to present this year, please submit an abstract!

If you would like travel sponsorship for GUADEC, there are two deadlines to submit a request: 15th March (for those who need to book travel early, such as if they need a visa) and 24th May (for those with less time pressure).

LAS 2026

This year’s Linux App Summit is happening in Berlin, on the 16th and 17th May, and is shaping up to be a great event. As usual we are co-organizing the event with KDE, and the call for proposals has just opened. If you’d like to present, you have until 23rd March to submit a paper.

The Travel Committee will be accepting travel applications for LAS attendees this year, so if you’d like to attend and need travel assistance, please submit a request no later than 13th April.

Infrastructure

On the infrastracture side, GNOME’s single sign on service has been integrated with blogs.gnome.org, which is great for security, as well as meaning that you won’t need to remember an extra password for our WordPress instance. Many thanks to miniOrange for providing us with support for their OAuth plugin for WordPress, which has allowed this to happen!

That’s it for my update this week. In addition to the highlights that I’ve mentioned, there are quite a number of other activities happening at the Foundation right now, particularly around new programs, some of which we’re not quite ready to talk about, but hope to provide updates on soon.

Final Reminder: Flock 2026 Financial Assistance Applications Close Sunday, March 8th

2026-03-06T10:15:00Z

Do you want to join us for our annual contributor conference? We want to see you there! However, we know that traveling to a global event is a big trip. It costs real money. To help out, the Flock Organizing Team offers Flock 2026 financial assistance. We want to make sure money does not stop our active contributors from attending.

This is your final reminder. You must submit your form by Sunday, March 8th, 2026. The organizing team starts looking at the data on Monday morning. Because of this fast timeline, we cannot accept any late forms. Sunday is a hard stop.

What does this funding actually cover? We can help you pay for your travel. This includes your airfare or train tickets. We can also help cover your hotel room at the main event venue. We have a limited budget. Because of this, we cannot fully fund every person who applies. Your peers on the organizing team review all the forms. They look at your community impact to make these tough choices.

Note for Flock 2026 speakers

Are you giving a talk this year? We are excited to hear from you! But please remember one important rule. Being an accepted speaker does not give you guaranteed funding. You still need to ask for help. All speakers must fill out the Flock 2026 financial assistance form if they need travel support.

Apply for Flock 2026 financial assistance

Applying is easy. Just follow these steps:

Open the Google Form application.
Fill in your travel details and your estimated costs.
Explain your recent work and your impact in the Fedora community.
Submit the form before the end of the day on Sunday, March 8th.

We want to bring as many builders and contributors together as possible. Please do not wait until the last minute. If you need support to join us, fill out the application today!

The post Final Reminder: Flock 2026 Financial Assistance Applications Close Sunday, March 8th appeared first on Fedora Community Blog.

Community Update – Week 10 2026

2026-03-06T10:00:00Z

This is a report created by CLE Team, which is a team containing community members working in various Fedora groups for example Infrastructure, Release Engineering, Quality etc. This team is also moving forward some initiatives inside Fedora project.

Week: 02 – 06 March 2026

Fedora Infrastructure

This team is taking care of day to day business regarding Fedora Infrastructure.
It’s responsible for services running in Fedora infrastructure.
Ticket tracker

Migrated remaining pagure.io repositories, there is one remaining, but needs to have private issues implemented
Provide DNS domain name for konflux (ticket)
Meeting with the IPA-tuura team to move forward with the Ipsilon replacement project

CentOS Infra including CentOS CI

This team is taking care of day to day business regarding CentOS Infrastructure and CentOS Stream Infrastructure.
It’s responsible for services running in CentOS Infratrusture and CentOS Stream.
CentOS ticket tracker
CentOS Stream ticket tracker

Release Engineering

This team is taking care of day to day business regarding Fedora releases.
It’s responsible for releases, retirement process of packages and package builds.
Ticket tracker

Still in Beta Freeze, we have had a couple of release candidate composes for Fedora 44 Beta.
GO/NO-GO call for the Beta Release is tentatively scheduled for Thursday, March 5th.

RISC-V

This is the summary of the work done regarding the RISC-V architecture in Fedora.

Continue to comb through the list of Fedora packages that need work — submit changes to Fedora / upstream, review patches from others, and submit builds as needed.
Build/test the RISC-V unified kernel.
Figured out logistics for 24-h remote access to SpacemiT ‘K3’ hardware (based on RVA230
Discussed with the RISC-V SIG and Fedora docs about migrating the Fedora RISC-V docs from the wiki to forge.
Figure out the prerequisites for F44 builds — toolchain dependencies are being sorted out.

AI

This is the summary of the work done regarding AI in Fedora.

Aurelien Bompard Worked on the AI-powered This Week in Fedora.
Claude helped Gordon Messmer develop an extension to glibc’s malloc API to solve a common problem with memory inflation, and produced some illustrations of the problem and solution. The write-up will be posted shortly. Draft at: https://codeberg.org/gordonmessmer/dev-blog/src/branch/main/memory-efficiency-with-arena-allocators.md
Quality Team workflow for AI code review in Forgejo — easily reused if you’re interested! https://forge.fedoraproject.org/quality/workflows/src/branch/main/.forgejo/workflows/ai-review.yml

QE

This team is taking care of quality of Fedora. Maintaining CI, organizing test days
and keeping an eye on overall quality of Fedora releases.

Fedora 44 RCs are now available and under heavy testing. The Go/NoGo is scheduled in a few days. Lots of blocker (and not blocker) bugs were discovered, discussed, voted on and resolved.
Kernel 6.19 test days are complete and Podman 5.8 test days are under way.
A new version of BlockerBugs bot was implemented and deployed to staging, which should allow us to migrate blocker voting repository from Pagure to Forge (ticket).
Our cloud automation now reports results to the production wiki for each new compose.
The refactoring of Fedora Easy Karma was merged to the main branch. All places with links were also updated wrt the recent transition to Forge, and the migration is now done for this repo.
Forge QA repos now has a reusable workflow for an AI review.
More bugs hitting OpenQA automated testing were resolved, some some particular ones are very hard to debug, e.g. an ibus hang or an xdg-desktop-portal crash.

Forgejo

This team is working on introduction of https://forge.fedoraproject.org to Fedora
and migration of repositories from pagure.io.

[Forgejo] New Organization and Teams Request: Fedora KDE [Followup]
[Forgejo] Participated in the Forge review, planning and retrospective meet with Rodney, Nils, Ryan and David [Board]
[Forgejo] New organisation for Fedora Code of Conduct committee + migration [Followup] [Namespace] [Followup]
[Forgejo] New repo for Security SIG [Followup] [Namespace] [Followup] [Followup]
[Forgejo] New Organization and Teams Request: Fedora Btrfs [Followup] [Namespace]
[Docs] Will be migrating Release Notes imminently + branching for F44, contributions welcome once the repo appears on Forgejo
Forgejo runners added for orgs mindshare and apps
Private Issues
- Unit tests for code ⇔ DB boundary (Issue & PrivateIssue: done, related types: ongoing)
- Verify, debug and fix DB migration

UX

This team is working on improving User experience. Providing artwork, user experience,
usability, and general design services to the Fedora project

Updated all references to Fedora Design ticket tracker to point to Forgejo [ticket], migration of tickets still wip, but significant progress made.

If you have any questions or feedback, please respond to this report or contact us on #admin:fedoraproject.org channel on matrix.

The post Community Update – Week 10 2026 appeared first on Fedora Community Blog.

Logsqueak: Rescuing Insights from the Logseq Journal

2026-03-04T11:30:43Z

I’ve been using Logseq for a year now and it’s become the backbone of my workflow. I have pages dedicated to specific topics, concepts, projects, meetings… all sorts of things.

During my day, when I want to note something down or write something out to think about it, the daily Logseq journal is the obvious place for it to go. It has been an invaluable habit to build. But there’s a catch: the journal can easily become a black hole. It ends up as a chaotic mix of meeting notes, fleeting thoughts, random ideas, task lists and the occasional moment of genuine insight.

Most of the time, I try to link journal items to the relevant pages. Sometimes I remember to update those pages in light of new information. But other times I forget, and those insights get buried in the timeline, only resurfacing if I explicitly search for them.

All of those things belong in the journal, but some of them also belong in permanent pages. I wanted a way to filter the signal from the noise and capture things that I can integrate into my pages, in a way that makes them traceable back to the journal, without leaving the keyboard.

Enter Logsqueak: a proof-of-concept experiment to see if a local AI model can act as an automated gardener for a Personal Knowledge Management (PKM) system.

How Logsqueak Works

It’s a Python-based terminal UI built with Textual, using RAG (Retrieval-Augmented Generation) via Ollama. Because PKM data is highly personal, my aim was to be able to build a tool that can run entirely on a local GPU, meaning your private journal entries never have to leave your machine. (Though you can certainly connect it to much larger cloud models if you prefer.)

The workflow is broken down into 3 phases:

1: Extraction (Signal vs. Noise)

In this phase, Logsqueak reads your Logseq journal and helps identify which items are ephemeral daily noise (e.g., “Morning standup at 9am”) and which are actual knowledge or insight worth keeping.

2: Refinement (Making it Evergreen)

Temporal context is stripped away, and additional context from parent bullet points is added in.

Original Journal Entry:
- Working on the new analytics dashboard
  - Finally figured out why the main chart was double-fetching data on load. The useEffect hook was missing the empty dependency array.
Logsqueak Refinement:
- To prevent double-fetching data on load in the analytics dashboard, ensure the useEffect hook for the main chart includes an empty dependency array.

3: Integration (Filing it Away)

In this final phase, the most semantically relevant pages in your Logseq graph are tracked down, and the best insertion point is identified. Logsqueak will suggest exactly which page and heading the new insight belongs under.

Logseq is built around powerful block properties, so crucially this is where the traceability happens. When an insight is integrated, Logsqueak adds an extracted-to:: property to the original journal block, linking it directly to the new block. The new block on the target page gets an id:: property linking back. This means you can always jump from your polished knowledge base straight back to the original journal entry to see the full context of what you were doing that day.

All writes are performed using a custom engine specifically built for Logseq’s Markdown format, ensuring your notes stay safe. Because this is a proof-of-concept, all writes are guarded by explicit user approval—Logsqueak won’t change your files without you saying “yes.”

Try it out!

Logsqueak requires Python 3.11+ and an AI assistant. You can use Ollama to run everything locally.

If you’re on Fedora, getting the prerequisites running is incredibly straightforward. Since Fedora Workstation ships with recent Python versions out of the box, you’re already halfway there. You just need to grab Ollama to run the models locally, set up a virtual environment, and you’re good to go:

# Assuming you've installed Ollama
git clone https://github.com/twaugh/logsqueak.git
cd logsqueak
./setup-dev.sh
source venv/bin/activate
logsqueak init

Taming the Knowledge Graph

This tool can help you turn a pile of daily logs into a structured, searchable knowledge base. Although it can’t yet create new pages from scratch or be given custom instructions about how best to integrate things into the graph, it’s already useful enough for me to use in my daily routine.

It’s very much a proof-of-concept though, and I’d love to get some feedback from other developers and knowledge management enthusiasts. You can check out the code on GitHub.

Building Logsqueak made me realise just how much time I spend thinking about note-taking friction. While Logsqueak handles my fast, keyboard-driven daily logging, I actually do a lot of my deep thinking away from the screen on a Ratta Supernote e-ink tablet.

I recently found myself trying to solve a similar “black hole” problem over there. The result is Slipstream: a Zettelkasten framework to let you build infinitely nested idea networks by hand.

If you happen to be an e-ink user who prefers a stylus to a keyboard when you need to disconnect and focus, you might find it an interesting contrast. As a bonus, because Slipstream has a structured convention, exporting those handwritten notes to plain text makes them perfectly readable for the exact kind of LLM processing Logsqueak relies on. It’s analogue thinking, ready for the AI age.

The post Logsqueak: Rescuing Insights from the Logseq Journal appeared first on PRINT HEAD.

Uniqlo T-Shirt Bash Script Easter Egg

2026-03-03T21:18:35Z

At Uniqlo flagship store in Ginza, Tokyo, there was this T-shirt with an encoded shell script.

Well, I had to decode it and see the result.

I took a photo with my iPhone and used its own OCR, which made a lot of confusion between 0 (zero), O (capital o) and 8, mixed 1 and l (small L), and yielded many chars as very similar glyphs but in advanced Unicode ranges, which are invalid for Base64 encoding. It took me some time to fix it all. The final corrected text is this:

#!/bin/bash
eval "$(base64 -d <<< 'IyEvYmluL2Jhc2gKCiMgQ29uZ3Jhd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 ')"

When base64-decoded, this bash script appears:

#!/bin/bash
# Congratulations! You found the easter egg! ❤
# おめでとうございます！隠されたサプライズを見つけました！❤
# Define the text to animate
text="♥PEACE♥FOR♥ALL♥PEACE♥FOR♥ALL♥PEACE♥FOR♥ALL♥PEACE♥FOR♥ALL♥PEACE♥FOR♥ALL♥"
# Get terminal dimensions
cols=$(tput cols)
lines=$(tput lines)
# Calculate the length of the text
text_length=${#text}
# Hide the cursor
tput civis
# Trap CTRL+C to show the cursor before exiting
trap "tput cnorm; exit" SIGINT
# Set frequency scaling factor
freq=0.2
# Infinite loop for continuous animation
for (( t=0; ; t+=1 )); do
    # Extract one character at a time
    char="${text:t % text_length:1}"
    # Calculate the angle in radians
    angle=$(echo "($t) * $freq" | bc -l)
    # Calculate the sine of the angle
    sine_value=$(echo "s($angle)" | bc -l)
    # Calculate x position using the sine value
    x=$(echo "($cols / 2) + ($cols / 4) * $sine_value" | bc -l)
    x=$(printf "%.0f" "$x")
    # Ensure x is within terminal bounds
    if (( x < 0 )); then x=0; fi
    if (( x >= cols )); then x=$((cols - 1)); fi
    # Calculate color gradient between 12 (cyan) and 208 (orange)
    color_start=12
    color_end=208
    color_range=$((color_end - color_start))
    color=$((color_start + (color_range * t / lines) % color_range))
    # Print the character with 256-color support
    echo -ne "\033[38;5;${color}m"$(tput cup $t $x)"$char\033[0m"
    # Line feed to move downward
    echo ""
done

The original encoded text, when executed in my Linux terminal gives a beatiful animation, similar to this:

After decoding and executing the script, I found other people doing the same:

But I swear I decoded it myself first, without any help.

Mauvaise foi : Hugo Roy savait que la FSFE se faisait passer pour la FSF devant le tribunal français, ses collègues ont été trompés.

2026-03-03T13:00:00Z

Les avocats ne sont généralement pas responsables des crimes de leurs clients. Si le client ment sous serment et que l'avocat n'est pas au courant du mensonge, ce dernier ne peut être tenu responsable.

Toutefois, dans les cas où l'avocat aide sciemment et volontairement son client à tromper le juge, les conséquences peuvent être extrêmement graves pour lui.

Par exemple, dans certaines affaires récentes, des avocats ont eu recours à l'intelligence artificielle pour rédiger des documents juridiques. Dans ces cas précis, il est évident pour tous que c'est l'avocat, et non le client, qui a pris un raccourci.

Dans le cas des marginaux de la FSFE , on constate que l'avocat français Hugo Roy a effectué un stage auprès d'eux en 2009. En 2011, il a intégré l'Assemblée générale (AG) de la FSFE. L'AG est la liste de diffusion privée où les initiés et les marginaux discutent et complotent contre le reste de la communauté du logiciel libre. Elle est comparable au groupe secret (et divulgué) de debian-private , où les débianistes dissidents se livrent à des actes tels que de fausses accusations de viol .

En mai 2017, la véritable FSF a réitéré sa demande aux marginaux de la FSFE de cesser d'utiliser le nom FSFE en raison de la confusion qu'il engendre.

La demande a été partagée avec tous les membres de la liste de diffusion de GA. Jonas Oberg , le directeur exécutif, nous a indiqué que la véritable FSF avait raison et que les membres de la FSFE n'avaient pas le droit d'utiliser un nom suggérant une affiliation avec la véritable FSF et le Dr Richard Stallman . De fait, ces membres de la FSFE se livrent à une usurpation d'identité , ou à une forme de fraude nigériane, chaque fois qu'ils utilisent le nom FSFE sur un document.

Voici la discussion de mai 2017. Tous les membres de l'Assemblée générale, y compris Hugo Roy , l'ont vue :

Commentaires de Jonas Oberg , le directeur général :

Objet : Re: La FSF nous demande de changer de nom II
Date : mar. 30 mai 2017 15:31:39 +0200
De : Jonas Oberg 
À : Bernhard Reiter 
CC : ga@fsfeurope.org

Salut Bernhard ,

Je suis globalement d'accord avec vous, mais j'aimerais vous demander une précision sur un point.
cette partie :

Je ne vois pas pourquoi. Nous devrions leur demander de mettre en place la coopération convenue.

Si j'examine honnêtement l'accord-cadre, je crois que c'est
formulé de manière plutôt favorable à la FSF, et beaucoup de ce que nous ferions
J'aimerais voir – par exemple, une prise de décision conjointe sur des questions importantes liées
au logiciel libre – cela ne figure pas réellement dans l'accord, mis à part une intention.
développer une telle voie dans un futur hypothétique.

Et je comprends parfaitement pourquoi la FSF estime que nous sommes en infraction avec la loi.
accord, [ ... extrait ... ]

[ ... extrait ... ]

L'accord-cadre, en l'état, n'est donc pas respecté.
N'importe quel côté. [ ... extrait ... ]

[ ... extrait ... ]

Sincèrement,

--
Jonas Öberg, directeur exécutif
Fondation européenne du logiciel libre | jonas@fsfe.org
Votre soutien nous permet de poursuivre notre travail (fsfe.org/join)

Quelques mois plus tard, lors de la discussion sur le conflit d'intérêts, Hugo Roy a confirmé qu'il était avocat, autorisé à exercer au barreau de Paris et employé par Baker McKenzie :

Objet : Re: conflit d'intérêts
Date : mar. 15 août 2017 08:27:29 +0000
De : Hugo Roy
À : ga@fsfeurope.org

Salut,

-----
En résumé : il faudrait plus de transparence sur ces questions au sein de l’Assemblée générale.
Du moins, et c'est ce que nous avons fait. Donc, pour ceux qui ne le font pas...
Sachez-le : je suis avocat et je conseille des clients qui se trouvent être de grandes entreprises.
Les fournisseurs de logiciels propriétaires également.
-----

Je ne vois rien de mal pour le moment dans la façon dont nous fonctionnons actuellement.
Je gère ça. Ce que je veux dire, c'est qu'aucun problème n'est apparu et que je
Je n'y ai pas vraiment réfléchi. D'ailleurs, (peut-être ai-je aussi analysé la discussion)
(Je vais vite, alors excusez-moi si je me trompe sur certains points, mais je ne pense pas que ce soit le cas.)
Quelqu'un ici a suggéré que toutes les informations divulguées conformément à une
La politique en matière de conflits d'intérêts serait nécessairement publique, voire même élaborée.
public par défaut.

Cependant, ces problèmes n'apparaissent pas toujours explicitement, il arrive parfois
plus pernicieux que cela (sans vouloir sous-entendre aucune malice)
ici).

Je ne serais *pas* à l'aise si je le découvrais par moi-même.
Coïncidence ou pas, qu'un membre de GA soit chez Microsoft, Red Hat ou Facebook ?
membre du personnel.

Je pense qu'il est important qu'au sein du groupe GA, nous ayons certains
visibilité sur ces sujets. Nous sommes un petit groupe et nous devrions connaître chaque élément.
autre. Et cela exige aussi une certaine transparence --- car il s'agit d'un
le fait que nous ne nous connaissons pas tous forcément bien, ne serait-ce que pour le
le fait que certains membres aient adhéré récemment.

Or, quelqu'un a fait remarquer à juste titre que pour ceux qui sont travailleurs indépendants,
Cela serait plus logique si la divulgation des plus gros clients était faite ; mais voilà.
difficilement possible (les gros clients ont tendance à exiger la confidentialité).

Il semblerait que cela me concerne. Je suis avocat et je travaille dans un grand cabinet.
cabinet ( Baker McKenzie ) --- cela ne vous donnera probablement pas d'informations intéressantes
informations à des fins de déclaration de conflit d'intérêts. Cependant, je ne suis pas à
liberté de divulguer mes clients ou ceux du cabinet — bien que je puisse dire que
Ils comprennent notamment de grands éditeurs de logiciels propriétaires, ce qui ne devrait guère être le cas.
Cela peut paraître surprenant ; après tout, je donne des conseils en droit informatique.

Meilleur,
Hugo

Peu après, Hugo Roy s'est activement impliqué dans la désignation d'un autre avocat français pour intenter une action en justice en France contre le ministère de la Défense. Hugo Roy a transmis des informations à cet avocat. Il savait que le nom de la Free Software Foundation Europe figurerait dans les documents judiciaires publics et que le public l'associerait à la véritable FSF et au Dr Richard Stallman . Il semble qu'il n'ait averti ni l'avocat chargé de la procédure, ni l'avocat qui a présenté l'affaire devant le tribunal, de cette usurpation d'identité et du litige en cours.

Objet : Décision d'intenter une action en justice contre l'armée française – Contrat Microsoft
Date : jeu. 26 oct. 2017 17:32:56 +0200
De : Hugo Roy 
À : team@lists.fsfe.org

Désolé, je suis pressé.

Avons-nous déjà pris une décision ?
--
Envoyé depuis mon mobile

Objet : Re: Tr: CONTRAT MICROSOFT (Ministre français de la Défense)
Date : Sam., 28 oct. 2017 02:41:49 +0561
De : Hugo Roy 
À : Alexis Fitzjean O Cobhthaigh , france@lists.fsfe.org, Matthias Kirschner , team@lists.fsfe.org


Salut tout le monde,

J'ai pris contact avec Nexedi et April aujourd'hui, pour m'assurer que le lancement
Cette action n'aurait pas d'effets néfastes sur eux. Tout semble
Très bien --- Alexis, n'hésite pas à donner ton avis.

Meilleur,
Hugo

↪ Matthias Kirschner / 27 octobre 2017 11:57 :
Bonjour Alexis, bonjour Hugo,
>
Comme je viens de le convenir avec Hugo : oui, allez-y. J’ai pleinement confiance.
Je vous en parle. On pourrait régler certains détails la semaine prochaine.
(Je serai de retour mercredi). Au cas où il y aurait quelque chose d'urgent d'ici là,
> veuillez m'appeler).
>
Meilleurs vœux,
> Matthias
>
> * Hugo Roy [2017-10-27 09:06 +0200] :
>> 1) Nous devrions déposer une demande auprès du CADA dès que possible pour tenter d'obtenir
>> les parties effacées du contrat.
>>
>> Il faut faire vite. C'est assez facile et je le ferai *bénévolement*.
Je ne suis pas sûr que ce soit très utile, mais ça vaut le coup d'essayer.
>> déposé dans les deux mois suivant la réception de la réponse du ministère
>> (Pourriez-vous confirmer la date ?)
>>
>> 2) Je pense que nous devrons déposer une plainte devant le tribunal administratif de
>> Paris (*avant le lundi 30 octobre*).
>
> [...]
>
Pour cette partie, je peux baisser mon tarif horaire à 125 $ avant taxes (150 $ taxes incluses).
>> inclus). Et, si nous portons plainte par précaution et que vous choisissez ensuite de
>> Retirez-le une semaine plus tard, je ne facturerai rien.
>
> --
Matthias Kirschner - Président - Free Software Foundation Europe
> Schönhauser Allee 6/7, 10119 Berlin, Allemagne | +49-30-27595290
> Enregistré auprès de l'Amtsgericht Hamburg, VR 17030 | (fsfe.org/join)
> Contact (fsfe.org/about/kirschner) - Blog (k7r.eu/blog.html)
>

Sujet : Tr : Re : Re : Affaire Microsoft
Date : dim. 5 nov. 2017 01:13:04 +0561
De : Hugo Roy 
À : team@lists.fsfe.org


Salut tout le monde,

Pour information, voici le mémoire déposé auprès du tribunal de Paris dans l'affaire Microsoft.
Placez tous les dossiers de cas dans le SVN interne.

Meilleur,
Hugo

Sujet : Re: Re: Affaire Microsoft
De : Alexis Fitzjean O Cobhthaigh 
Date : 31/10/2017, 12:02
À : Hugo Roy 
CC : france@lists.fsfe.org

Bonjour à tous,

La requête a bien été enregistrée, par le tribunal administratif de Paris, sous le n° 1716711.

Je vous mets en pièce jointe un exemplaire, au format PDF, de la requête telle qu'elle a été déposée hier soir.

Je reste naturellement à votre disposition si vous souhaitez que l'on échange, par écrit ou de vive-voix, sur ce dossier.

Alexis

Le 30 octobre 2017 à 22:07, Hugo Roy  a écrit :


    La requête a été déposée aujourd'hui. Alexis, tu peux nous envoyer le
    PDF officiel lorsqu'il est prêt ?

    Merci

    PS : garder france@ en copie autant que possible - désolé je n'ai pas eu eu
    le temps du faire aujourd'hui


    ---------- Message transféré ----------
    De : Alexis Fitzjean O Cobhthaigh 
    À : Frédéric Couchet 
    Cc : Hugo Roy , Étienne GONNU 
    Date : lundi 30 octobre 2017 10:00:35 +0100
    Objet : Re : Affaire Microsoft
    Bonjour à tous,

    Comme convenu lors de notre discussion vendredi avec Hugo (en copie), vous trouverez en pièce jointe le projet de requête sommaire que je m'apprête à déposer dans la journée pour la FSFE. J'ai fait au mieux dans un délai si réduit, étant précisé que l'argumentation pourra être complétée ultérieurement.

    Vos éventuelles remarques et suggestions sont bien entendues les bienvenues. J'envisage un dépôt dans l'après-midi, en ajoutant quelques modifications et corrections mineures.

    N'hésitez pas à m'appeler à mon cabinet au besoin pour en discuter.

    Bien à vous,

    Alexis FITZJEAN Ó COBHTHAIGH
    Avocat au Barreau de Paris / Avocat au Barreau de Paris
    Chargé d'enseignement à l'Université Paris II Panthéon-Assas
    c/o SCP MARLANGE-de la BURGADE
    5, rue Daunou 75002 PARIS
    afoc.avocat@gmail.com
    Tél : (+33)1.53.63.33.10 Fax : (+33)1.45.48.90.09


    PRIVILÉGIÉ ET CONFIDENTIEL/CONFIDENTIEL

    Ce message contient des informations privilégiées et confidentielles. Si vous n'êtes pas le destinataire désigné, veuillez le détruire sans le lire, le copier ni en divulguer le contenu à quiconque.
    Cet e-mail contient des informations confidentielles protégées par le secret professionnel. S'il ne vous est pas destiné, nous vous remercions de le détruire immédiatement, sans le copier, ni révéler ou transmettre son contenu à qui que ce soit. &ZeroWidthSpace

    Le 28 octobre 2017 à 16:08, Frédéric Couchet  a écrit :


        Bonjour à vous,

        >>>>> Étienne GONNU  écrit :

            > J'en profite également pour une autre précision: sur quelle liste,
            > ou avec qui, le sujet de l'action de la FSFE a t'il été évoqué via
            > courriel ? Je n'en trouve pas trace dans mon courrielleur.

        pareil, je n'ai aucune trace d'un quelconque échange sur ce sujet, mais
        nous avons pu louper quelque chose :)

        Librement,
        Fred.
        --
        Avril - Promouvoir et défendre le logiciel libre

        https://www.april.org

Objet : Fwd : 1716711 - FSFE vs Ministre des armées (Ministry of Defence) - Mémoire en défense (defence)
Date : jeu. 19 avr. 2018 17:15:07 +0200
De : Hugo Roy
À : team@lists.fsfe.org, legal@lists.fsfe.org
CC : france@lists.fsfe.org

Salut tout le monde,

Petit point sur l'affaire contre Microsoft (tous les documents sont dans le dossier)
SVN interne).

Nous avons reçu la réponse du gouvernement. Il semble qu'ils traitent
Cela fait l'objet d'une attention particulière. Alexis , notre avocate dans cette affaire, est
Nous préparons un projet qui sera partagé avec france@ lundi.
Partagez-le également avec April (et éventuellement d'autres personnes) pour obtenir leurs commentaires.

Vous trouverez ci-joint le courriel d'Alexis à ce sujet, au cas où cela vous intéresserait.
et également la réponse du gouvernement français.

Notre réponse au gouvernement doit être adressée au tribunal de Paris avant
Le 30 avril, ce qui nous laisse (fsfe) un peu moins d'une semaine pour examiner
Le brouillon d'Alexis .

Si vous avez des questions, n'hésitez pas !

Meilleur,
Hugo

Objet : 1716711 - FSFE vs Ministre des armées (Ministry of Defence) - Mémoire en défense (defence)
De : Alexis Fitzjean O Cobhthaigh
Date : 06/04/2018, 18:56
À : Matthias Kirschner , france@lists.fsfe.org, Hugo Roy

Bonjour à tous,

Veuillez trouver ci-joint la défense du Ministère. C'est formidable, car nous savons maintenant où nous allons. Sans surprise.

1) Le ministère soutient que la juridiction compétente n'est pas le tribunal administratif de Paris, mais celui de Melun (ville très proche de Paris). Cela n'a pas d'incidence majeure : même si le tribunal compétent est celui de Melun et non celui de Paris, le tribunal de Paris transmettra simplement l'affaire à celui de Melun (aucun risque d'incompétence ou d'irrecevabilité à ce sujet).

2) Sans surprise, le ministère soutient que FSFE n'a pas qualité pour agir car elle n'est pas concurrente de Microsoft. Rien d'étonnant.

3) Le ministère écrit, sur plusieurs pages, que ce contrat est parfaitement normal. Rien de surprenant. Le tribunal devra trancher. Or, d'ordinaire, le ministère ne consacre que six ou sept pages aux affaires simples portées devant le tribunal administratif. Cela confirme donc qu'il y a bien matière à contester, même si la tâche s'annonce ardue. Mon avis : ils sont vraiment mécontents de notre situation.

4) Plus étonnant encore, le ministère demande que le Tribunal condamne la FSFE à 3000 euros de frais de procédure, ce qui est beaucoup pour ce type d'affaire.

Pour l'équipe de France :

C'est intéressant qu'ils aient produit dès à présent car cela va nous permettre de répliquer utilement, plutôt que de produire une nouvelle mémoire "dans le vide".

En résumé, il y a 3 points sur lesquels répondre :
- l'intérêt à agir de la FSFE ;
- le moyen sur l'absence de publicité et de mise en concurrence ;
- le moyen sur l'abus de position dominante.

Ce qui est intéressant c'est que, même s'ils contestent la recevabilité de notre recours, ils ont quand même jugé utile de répondre significativement sur le fond et, surtout, d'ajouter à titre subsidiaire que, si le TA estime notre action recevable ET fondée, il devrait qu'il prenne en compte l'intérêt général pour moduler les effets de la décision qu'il rend. Cela montre bien qu'ils ne sont pas très à l'aise et, à tout le moins, qu'ils nous prennent au sérieux. S'ils étaient sûrs d'eux ils n'auraient pas pris la peine de produire une défense aussi substantielle.

Autre point : demander 3 000 euros de frais de procédure, de la part de l'Etat, devant le TA, contre une ONG, c'est très rare et cela montre bien que ce recours les agace et, peut-être que cela les inquiète aussi.

Enfin, le recours est signé par Claire Legras, qui est la directrice des affaires juridiques du ministère en personne. C'est très rare que ce soit la DAJ qui signe devant le TA, ce qui montre bien qu'ils nous prennent au sérieux.

A titre plus subsidiaire, s'agissant de la communication d'autres documents, le ministère mélange un peu tout. Sa réponse se place sur le terrain des règles applicables aux procédures CADA (les documents demandés doivent être désignés de manière suffisamment précise) et non celles, autonomes, qui régissent les pouvoirs d'instruction du juge administratif (les parties peuvent suggérer des documents mais c'est le juge qui, in fine, décide lequel il demande la production). Je prépare également une réponse sur ce point, en précisant que l'on souhaite la production des notes préparatoires sur lesquelles le ministère a décidé de ne pas utiliser à une procédure d'appel d'offres préalable.

Je vous remercie de me faire part de vos observations d'ici vendredi 13 mars prochain. De sorte que je peux ensuite avoir le temps de les intégrer et de vous envoyer un projet la semaine suivante, pour un dépôt qui doit intervenir, en toute hypothèse, avant le 30 avril (date de la clôture de l'instruction).

Salutations,
Alexis

Alexis FITZJEAN Ó COBHTHAIGH

Avocat au Barreau de Paris / Avocat au Barreau de Paris
Chargé d'enseignement à l'Université Paris II Panthéon-Assas
c/o SCP MARLANGE-de la BURGADE
5, rue Daunou 75002 PARIS
afoc.avocat@gmail.com
Tél : (+33)1.53.63.33.10 Fax : (+33)1.45.48.90.09

En 2019, Hugo Roy a été démis de ses fonctions au sein du groupe dissident FSFE . Ce groupe n'a publié aucune déclaration publique concernant sa démission. A-t-il démissionné suite aux attaques contre ma famille ou à cause de la plainte pour harcèlement sexuel déposée par Galia Mancheva ?

En 2020, un tribunal berlinois a rendu un jugement dans une affaire de harcèlement sexuel . Le nom de la « Free Software Foundation Europe » figure dans les documents juridiques. Une fois de plus, un observateur non averti consultant ces documents pourrait croire, à tort, qu'il s'agit de la véritable FSF et du Dr Richard Stallman .

Le jugement a finalement été rendu contre la Free Software Foundation Europe (FSFE) le jour de l'anniversaire d'Hitler .

Les gens examineront le jugement rendu contre la FSFE et auront l'impression qu'il s'agit d'un jugement contre la véritable FSF et le Dr Richard Stallman .

Par conséquent, nous pouvons conclure hors de tout doute raisonnable que le Tribunal administratif de Melun a été trompé en publiant un jugement sous le nom usurpé par le plaignant.

Lorsque Hugo Roy a mandaté un autre avocat pour présenter les documents au tribunal, sachant que le demandeur n'avait pas le droit d'utiliser ce nom, il savait que le tribunal et le public seraient induits en erreur. Par conséquent, s'il n'a pas averti le tribunal et les autres avocats de ce problème, s'il le leur a dissimulé, il a agi de mauvaise foi et a manqué aux obligations déontologiques des avocats dans leurs interactions avec le tribunal.

Veuillez visionner ma vidéo de campagne de financement participatif et en discuter avec votre communauté dès aujourd'hui .

Fedora Documentation translations not available from March 4th, 2026

2026-03-03T10:00:00Z

Fedora Documentation translations will be put on hold from March 4th as the Fedora Localization Team has started the process of migration from pagure.io to the Fedora Forge. From the date, translation projects of the documentation (with ‘fedora-docs-l10n’ in name) will be gradually locked on the Fedora translation platform. Translation automation of the Docs website will also be stopped in the Fedora infrastructure. Consequently, there will be no translation updates available in the language versions on the Fedora Documentation.

The migration involves all repositories which support and ensure the availability of translations of the Fedora Documentation. There is no possibility the migration can be performed ‘on the fly’ as changes in the repositories, related scripts and continuous integration with the translation platform cannot be dealt with independently. Therefore the translation process of the Fedora Documentation is kept on hold.

We regrettably ask the Fedora contributors, our translation community, to pull back from translating of the Fedora Documentation and wait till the translation automation of the documentation is resumed again.

The progress of migration can be followed in the localization tracker as issue #52.

The post Fedora Documentation translations not available from March 4th, 2026 appeared first on Fedora Community Blog.

Honey, I Shrunk the Model (Maybe): blk-archive vs AI Data

2026-03-03T00:00:00Z

Because “It Should Work” Isn’t Data

After reading about the billions spent on AI infrastructure, I kept wondering: how much of that storage is just… the same bytes over and over? So I decided to find out. As I’m involved with https://github.com/device-mapper-utils/blk-archive I thought it would be good to understand how much storage blk-archive can realistically save when pointed at AI-style datasets.

Let me be clear upfront: this isn’t a speed test or hardware benchmark. I’m only interested in one question: how many bytes go in, and how many come out?

LImiting What an Agent can do

2026-03-02T22:40:38Z

I do not work with AI tools. This is not advice from experience of working with AI. It is advice from working with access controls in general.

Any agent has responsibility and authority. Responsibility is what it is required to produce. Authority is the set of resources that you provide to that agent. This does not change if the agent is human or automation, and AI agents fall in to that later category.

The way to limit what an agent can do is to allow it access to nothing, and then see what it requests access to. If that resource is reasonable, provide access.

The best example I can point to for a workflow like this is SELinux. When a new program is added to Fedora or comparable OS, it requires and update to the SELinux policy to say what files it can read/write/execute.

To generate this policy, the developer runs the program on a scratch system in permissive mode, and tracks the aces where policy would deny the program access to a resource. The engineer can then look at the set of resources and build a new policy to allow access to those resources, and only those. If one of the requests is suspect, the SELinux policy team is unlikely to accept the updated policy.

You do not require an agent to self-limit access. We don’t trust humans to that, we certainly should not require automation to do that.

An agent should not be able to make any web request by default. No posts to github/gitlab, or Wells Fargo, or the NSA. Every URL, every Host should be denied until authorized.

However, yes-no access alone may not be sufficient. A request to read or write or make a web call may be perfectly innocuous. And a payment made for a small resource may be perfectly acceptable. But filling up a hard disk, or deleting files, or emptying a bank account are all issues of scale. Any resources that can be exhausted require limits. Quotas are hard, and delegation of quotas to other systems are even harder. But not impossible. I wrote about it a while back: https://adam.younglogic.com/2018/05/tracking-quota/

If your agent is supposed to write code, let it write code in a sandbox. Let the human delegating to the agent take the responsibility of promoting that code to a live system. Do not allow it to delete database schemas that you would not let a human delete.

misc fedora bits last week of feb 2026

2026-02-28T17:10:26Z

The year is rolling along, and here we are at the end of Feb.

Lots of small day to day items

There were a lot of small day to day investigations and incoming requests, along with a pretty large amount of pull requests for our ansible repo. Since we are in Beta freeze some of them will have to wait, but some we can test out in staging now. It's great to see people submitting fixes and enhancements.

There were also some small fun to debug issues this week, including:

The https://whatcanidoforfedora.org/ site was sometimes alerting that it's ssl cert was expired. Turns out this was caused by that domain having old ip's for 2 proxies that had moved datacenters. So, sometimes it hit those, timed out and the ssl check just assumed it was bad. So, it was DNS. :)
The fedorapeople.org web server started being very slow to respond. Turns out the scrapers were hitting the cgit interface there and downloading xz snapshots of every commit. This caused the server to have to try and compress things over and over again. So, for now I just disabled those links and increased resources on the webserver. scrapers continue to keep on giving.

Secure boot signing work

Much of my time this week has been spent working on our new secure boot signing workflow. This is really really overdue and something I was hoping to finish mid last year, but things kept coming up and it kept getting pushed back.

The new setup leverages our existing signing infrastructure (sigul) so there's no need for special build hardware anymore. It also removes some constraints in the existing setup allowing us to do something we have wanted for a long time, namely sign aarch64 boot loader artifacts for secure boot.

Kudos to Jermey Cline for all his work on the code to make this possible. This uses the siguldry-bridge, rust based server to talk to sigul, and hopfully before too long we can replace the sigul server side with the new rust based server too.

I got everything deployed, I am now able to sign things, but in testing on my aarch64 laptop, there's still some issue with grub2 that needs to be sorted out. Hopefully it's something not too difficult to track down and we can move to this new setup after beta freeze once and for all.

comments? additions? reactions?

As always, comment on mastodon: https://fosstodon.org/@nirik/116149442549416772

Installing TeX Live 2025 on Fedora: A Minimalist User-Space Guide

2026-02-28T10:40:00Z

This guide consolidates the steps required to establish a professional-grade LaTeX environment into a single, clean workflow. Following this chronological order ensures that directories exist, dependencies are met, and the environment is correctly configured before the software attempts to utilize them.

By installing TeX Live in user-space (within ~/.local/opt), we adhere to a minimalist philosophy. This approach avoids cluttering the system root, removes the need for sudo during package management via tlmgr, and ensures you are using the most recent upstream binaries rather than potentially outdated distribution packages.

Pre-Installation: System Dependencies

Fedora's "minimal" Perl and X11 profiles often lack specific libraries required by the TeX Live installer and its graphical management tools. While we aim for a lean system, these Perl modules are essential for the installer's checksum verification and the tlmgr interface.

Install these requirements using dnf:

sudo dnf install perl-File-Find perl-File-Copy perl-Digest-MD5 perl-Tk perl-Term-ANSIColor xdotool

Create the Directory Structure

We will create a "user-space system" directory. Utilizing ~/.local/opt keeps your home directory organized while mimicking the Linux Filesystem Hierarchy Standard (FHS).

mkdir -p ~/.local/bin

mkdir -p ~/.local/opt/texlive

Download and Launch the Installer

To avoid "checksum mismatch" errors common with unstable local mirrors, we will point the installer directly to the Comprehensive TeX Archive Network (CTAN) global repository.

cd ~/Downloads

wget http://mirror.ctan.org/systems/texlive/tlnet/install-tl-unx.tar.gz

tar -xzf install-tl-unx.tar.gz

cd install-tl-*

Launch the installer using a reliable global mirror:

./install-tl -repository http://mirror.ctan.org/systems/texlive/tlnet/

Installer Configuration (Text Menu)

Once the command-line interface menu appears, you must modify specific installation paths to point to your local directories:

Directories (D): Change TEXDIR to: /home/porfirio/.local/opt/texlive/2025
Options (O): * Toggle Paper Size to letter. * Toggle Create symlinks (L) to [X]. * Binaries: /home/porfirio/.local/bin * Manpages: /home/porfirio/.local/share/man * Info: /home/porfirio/.local/share/info
Install (I): Start the process.

Note

The full installation can take approximately 3 hours depending on your connection. It is highly recommended to run this inside a tmux session to prevent the installation from terminating if your terminal session disconnects.

Post-Installation: Environment Setup

Fedora needs to be instructed where to find these new binaries. We will create a modular profile script in ~/.bashrc.d/ to keep our ~/.bashrc clean.

Create the configuration file:

vi ~/.bashrc.d/texlive.sh

Paste the following content into the file:

# TeX Live 2025 Path Configuration
TEX_BIN_PATH="/home/porfirio/.local/opt/texlive/2025/bin/x86_64-linux"

if [ -d "$TEX_BIN_PATH" ]; then
    # Prioritize TeX binaries and your local bin folder
    export PATH="$TEX_BIN_PATH:/home/porfirio/.local/bin:$PATH"
    export INFOPATH="/home/porfirio/.local/opt/texlive/2025/texmf-dist/doc/info:$INFOPATH"
    export MANPATH="/home/porfirio/.local/opt/texlive/2025/texmf-dist/doc/man:$MANPATH"
fi

Apply the changes to your current session:

source ~/.bashrc

Finalizing the Formats

With the paths set and Perl dependencies satisfied, manually execute the tasks that occasionally fail during the installer's finalization stage due to environment restrictions:

# Set default paper size
tlmgr paper letter

# Build the format files (Engines)
fmtutil-sys --all

Verification and Troubleshooting

To ensure your system is ready for document production, perform a "smoke test" by checking the binary location and compiling a dummy document.

Check the Binary Path

which pdflatex

This should return a path pointing to your ~/.local/bin/ or ~/.local/opt/ folder rather than /usr/bin/.

Compile a Test Document

echo "\documentclass{article}\begin{document}LaTeX 2025 is Live on Fedora!\end{document}" > test.tex

pdflatex test.tex

Summary of Locations

For future maintenance on your system, keep these paths in mind:

The System Root: ~/.local/opt/texlive/2025/
The Binaries/Symlinks: ~/.local/bin/
The Configuration: ~/.bashrc.d/texlive.sh
The Logs: ~/.local/opt/texlive/2025/texmf-var/web2c/fmtutil.log

You now have a professional-grade, upstream LaTeX installation that is completely independent of Fedora's system packages, making it easy to back up or migrate to a new machine in the future.

Patchutils 0.4.5 released

2026-02-27T16:04:38Z

I have released version 0.4.5 of patchutils and also built it in Fedora rawhide. This is a stability-focused update fixing compatibility issues and bugs, some of which had been introduced in 0.4.4.

Compatibility Fix: Git Extended Diffs

Version 0.4.4 added support in the filterdiff suite for Git’s extended diff format. Git diffs without content hunks (such as renames, copies, mode-only changes and binary files) were included in the output. This broke compatibility with 0.4.3.

For 0.4.5 this functionality is now gated with a --git-extended-diffs=include|exclude parameter. The default for 0.4.x is to exclude files in Git extended diffs with no content. There were also some fixes relating to file numbering for these types of diffs.

Note: in 0.5.x this default will change to include.

Status Indictors for grepdiff

Previously grepdiff --status showed ! for all matching files, but now it correctly reports them as additions (+), removals (-) or modifications (!).

As always, bug reports and feature requests are welcome on GitHub. Thanks to everyone who reported issues and helped to test fixes!

The post Patchutils 0.4.5 released appeared first on PRINT HEAD.

Community Update – Week 9, 2026

2026-02-27T15:13:10Z

Week: 20 Feb – 27 Feb 2026

Fedora Infrastructure

This team is taking care of day to day business regarding Fedora Infrastructure.
It’s responsible for services running in Fedora infrastructure.
Ticket tracker

Migration of pagure.io repositories continues
Ansible repository now has CI running in runner
Resolved errors on dist-git that were spamming sysadmin-main mailbox – ticket

CentOS Infra including CentOS CI

Release Engineering

This team is taking care of day to day business regarding Fedora releases.
It’s responsible for releases, retirement process of packages and package builds.
Ticket tracker

Release engineering is currently in Beta Freeze.
Releng has provided the first beta release candidate after the QE request.
Request for creating detached signature has been handled by Samyak for ignition 2.26.0 release.

RISC-V

This is the summary of the work done regarding the RISC-V architecture in Fedora.

Continued to go through the list of packages to investigate (failing to build; requires patching, and more).
- Got a PR merged for ‘libkrunfw’ (a low-level library for process isolation) and built it in RISC-V Koji; Marcin (“hrw”) also got a couple more merged.
Work is progressing well on Fedora RISC-V unified kernels (Jason Montleon is doing most of the heavy-lifting here). Currently hosted in Copr.

AI

This is the summary of the work done regarding AI in Fedora.

Worked on the AI-powered This Week in Fedora.
Updated ai-code-review to use Gemini 3.1 Pro by default
Nexus team: interesting experience using Claude to refactor fedora-easy-karma, on review we found it made multiple unrequested functionality changes which complicated the review, after discussion decided to re-prompt it to avoid functionality changes

QE

This team is taking care of quality of Fedora. Maintaining CI, organizing test days
and keeping an eye on overall quality of Fedora releases.

Fedora 44 Beta testing in full swing
Kernel 6.19 test week in progress, please contribute
Help wanted: if anyone can help figure out why F44 boot occasionally just hangs for no very obvious reason that’d be great (details in bug)

Forgejo

This team is working on introduction of https://forge.fedoraproject.org to Fedora
and migration of repositories from pagure.io.

Create and present Fedora -> Forgejo efforts during FOSDEM 2026 Distributions Devroom [Video] [Followup]
Pagure migration doc: add follow-up checklist based on Nexus team’s [Review]
Create new organisation and namespace on Fedora Forge for Fedora Join [Followup] [Namespace]
Rename “translations” into “localization-docs” [Followup][Namespace]
- Remedy mapping after renaming translations to localization-docs [Commit]
Migrating NeuroFedora SIG [Followup][Namespace][Update]
- Perform mapping for NeuroFedora SIG teams and groups [Commit] [Followup]
New Organization and Teams Request: Fedora KDE [Followup][Triaged]
- Perform mapping for Fedora KDE teams and groups [Commit]
REST API: Create private issue & foundational work [Followup]
Runner for forge organization [Followup]
Refactor the ansible-role-forgejo-runner role task now configured and deploy the runners using ansible-pull
Clean out the spam from Pagure.io [Followup]
Refactor internal APIs for public vs. private issue identity [Followup][Feature]

EPEL

This team is working on keeping Epel running and helping package things.

Completed EPEL 10.2 mass branching in preparation for the upcoming RHEL 10.2 release (announcement)

UX

This team is working on improving User experience. Providing artwork, user experience,
usability, and general design services to the Fedora project

Madeline’s F45 Wallpaper Process Update: blog post / discussion post
The Fedora Design Team has switched to using Google Meet for our weekly meetings to utilise Gemini for note-taking. Discussion took place on this ticket and on our weekly call. Updated Fedocal entry can be found here.
Forgejo migration: Sprint board created, tickets currently being imported, remaining assets currently being imported / reorganised.

If you have any questions or feedback, please respond to this report or contact us on #admin:fedoraproject.org channel on matrix.

The post Community Update – Week 9, 2026 appeared first on Fedora Community Blog.

Mise en place d’Anubis sur l’instance Scaleway de Fedora-Fr

2026-02-27T11:22:00Z

Pour rappel, l’architecture de Fedora-Fr repose sur un VPS, ou plutôt une Instance, comme l’appelle Scaleway. Cette machine (virtuelle) correspond à nos besoins actuels et nous est gracieusement fournie par notre partenaire Scaleway. Côté sécurité, nous utilisons le firewall natif proposé par Scaleway, appelé Security Groups. Nous n’avons pas activé le WAF, car je n’ai […]

Cet article Mise en place d’Anubis sur l’instance Scaleway de Fedora-Fr est apparu en premier sur Guillaume Kulakowski's blog.

🎲 PHP version 8.4.19RC1 and 8.5.4RC1

2026-02-27T07:12:00Z

Release Candidate versions are available in the testing repository for Fedora and Enterprise Linux (RHEL / CentOS / Alma / Rocky and other clones) to allow more people to test them. They are available as Software Collections, for parallel installation, the perfect solution for such tests, and as base packages.

RPMs of PHP version 8.5.4RC1 are available

as base packages in the remi-modular-test for Fedora 42-44 and Enterprise Linux ≥ 8
as SCL in remi-test repository

RPMs of PHP version 8.4.19RC1 are available

as base packages in the remi-modular-test for Fedora 42-44 and Enterprise Linux ≥ 8
as SCL in remi-test repository

ℹ️ The packages are available for x86_64 and aarch64.

ℹ️ PHP version 8.3 is now in security mode only, so no more RC will be released.

ℹ️ Installation: follow the wizard instructions.

ℹ️ Announcements:

Parallel installation of version 8.5 as Software Collection:

yum --enablerepo=remi-test install php85

Parallel installation of version 8.4 as Software Collection:

yum --enablerepo=remi-test install php84

Update of system version 8.5:

dnf module switch-to php:remi-8.5
dnf --enablerepo=remi-modular-test update php\*

Update of system version 8.4:

dnf module switch-to php:remi-8.4
dnf --enablerepo=remi-modular-test update php\*

ℹ️ Notice:

version 8.5.4RC1 is in Fedora rawhide for QA
EL-10 packages are built using RHEL-10.1 and EPEL-10.1
EL-9 packages are built using RHEL-9.7 and EPEL-9
EL-8 packages are built using RHEL-8.10 and EPEL-8
oci8 extension uses the RPM of the Oracle Instant Client version 23.9 on x86_64 and aarch64
intl extension uses libicu 74.2
RC version is usually the same as the final version (no change accepted after RC, exception for security fix).
versions 8.4.19 and 8.5.4 are planed for March 12th, in 2 weeks.

Software Collections (php84, php85)

Base packages (php)

Veritasium

2026-02-26T08:37:38Z

I was interviewed on Veritasium about the rise of Linux and the XZ hack.

Cockpit 357

2026-02-25T00:00:00Z

Cockpit is the modern Linux admin interface. We release regularly.

Here are the release notes from Cockpit 357, cockpit-files 37, cockpit-machines 350 and cockpit-podman 123:

Show browser context menu when holding `Shift`

To help with accessibility and user-specific items in your browser context menu we now follow the behavior seen in Firefox: holding the Shift key while right-clicking will skip our own context menu and instead follow the default behavior of your browser. Note that while Cockpit Files uses the Cockpit context menu it did not receive this update before the release, but will get it for the for next one.

Machines: Support for kernel boot arguments when creating VMs

Cockpit now supports the --extra-args option of virt-install. This allows you to pass arguments to the kernel when the installer is booted during the creation of a new virtual machine. Installers like Anaconda and Agama can be controlled in various ways via these arguments.

Thanks to Nykseli for this contribution!

Podman: Show description, version and documentation image labels

The image details tab now shows a select set of labels from the OpenContainers Annotations specification.

Try it out

Cockpit 357, cockpit-files 37, cockpit-machines 350 and cockpit-podman 123 are available now:

📝 Install PHP 8.5 on Fedora, RHEL, CentOS Stream, Alma, Rocky or other clone

2026-02-24T11:16:00Z

Here is a quick howto upgrade default PHP version provided on Fedora, RHEL, CentOS, AlmaLinux, Rocky Linux or other clones with latest version 8.5.

You can also follow the Wizard instructions.

Architectures:

The repository is available for x86_64 (Intel/AMD) and aarch64 (ARM).

Repositories configuration:

On Fedora, standards repositories are enough, on Enterprise Linux (RHEL, CentOS) the Extra Packages for Enterprise Linux (EPEL) and Code Ready Builder (CRB) repositories must be configured.

Fedora 44

dnf install https://rpms.remirepo.net/fedora/remi-release-44.rpm

Fedora 43

dnf install https://rpms.remirepo.net/fedora/remi-release-43.rpm

Fedora 42

dnf install https://rpms.remirepo.net/fedora/remi-release-42.rpm

RHEL version 10.1

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-10.noarch.rpm
dnf install https://rpms.remirepo.net/enterprise/remi-release-10.rpm
subscription-manager repos --enable codeready-builder-for-rhel-10-x86_64-rpms

RHEL version 9.7

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-9.noarch.rpm
dnf install https://rpms.remirepo.net/enterprise/remi-release-9.rpm
subscription-manager repos --enable codeready-builder-for-rhel-9-x86_64-rpms

RHEL version 8.10

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm
dnf install https://rpms.remirepo.net/enterprise/remi-release-8.rpm
subscription-manager repos --enable codeready-builder-for-rhel-8-x86_64-rpms

Alma, CentOS Stream, Rocky version 10

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-10.noarch.rpm
dnf install https://rpms.remirepo.net/enterprise/remi-release-10.rpm
crb install

Alma, CentOS Stream, Rocky version 9

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-9.noarch.rpm
dnf install https://rpms.remirepo.net/enterprise/remi-release-9.rpm
crb install

Alma, Rocky version 8

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm
dnf install https://rpms.remirepo.net/enterprise/remi-release-8.rpm
crb install

PHP module usage

With Fedora and EL, you can simply use the remi-8.4 stream of the php module

With Fedora (dnf5 has partial module support)

dnf module reset php
dnf module enable php:remi-8.5
dnf install php-cli php-fpm php-mbstring php-xml

Other distributions (dnf4)

dnf module switch-to php:remi-8.5/common

PHP upgrade

By choice, the packages have the same name as in the distribution, so a simple update is enough:

dnf update

That's all :)

$ php -v
PHP 8.5.3 (cli) (built: Feb 10 2026 18:25:51) (NTS gcc x86_64)
Copyright (c) The PHP Group
Built by Remi's RPM repository  #StandWithUkraine
Zend Engine v4.5.3, Copyright (c) Zend Technologies
    with Zend OPcache v8.5.3, Copyright (c), by Zend Technologies

Known issues

The upgrade can fail (by design) when some installed extensions are not yet compatible with PHP 8.5.

See the compatibility tracking list: PECL extensions RPM status

If these extensions are not mandatory, you can remove them before the upgrade; otherwise, you must be patient.

Warning: some extensions are still under development, but it seems useful to provide them to upgrade more people and allow users to give feedback to the authors.

More information

If you prefer to install PHP 8.5 beside the default PHP version, this can be achieved using the php85 prefixed packages, see the PHP 8.5 as Software Collection post.

You can also try the configuration wizard.

This is also documented as the community way to install PHP 8.5 on the official PHP web site.

The packages available in the repository were used as sources for Fedora 44.

By providing a full feature PHP stack, with about 150 available extensions, 11 PHP versions, as base and SCL packages, for Fedora and Enterprise Linux, and with 300 000 downloads per day, the remi repository became in the last 21 years a reference for PHP users on RPM based distributions, maintained by an active contributor to the projects (Fedora, PHP, PECL...).

First speech of Chanellor Hitler, Andreas Tille & Debian denounce Branden Robinson

2026-02-22T18:30:00Z

Hitler became Chancellor of Germany in January 1933.

On 10 February 1933, Hitler gave his first official speech as Chancellor.

He used the occasion to attack his predecessors for everything that had transpired against German interests since the end of World War One.

Many copies of the speech are available online:

Thatâ€™s what Germany looks like today! Under the rule of these parties who have ruined our Volk for fourteen years. The only question is, for how much longer?

...

And this brings us thus to our sixth item, clearly the goal of our struggle: the preservation of this Volk and this soil, the preservation of this Volk for the future, in the realization that this alone can constitute our reason for being.

The word Volk, which means People or Population, appears forty two times in the speech.

On 10 February 2026, the Debianist leader Andreas Tille reprimanded one of his predecessors G. Branden Robinson, just as Adolf Hitler had reprimanded the predecessors who ruled Germany in the fourteen years before the Nazis took power.

While Hitler claimed to represent the Volk / People, Andreas Tille claims to represent the user, whoever that is, in this spiteful reply:

... directly harms our ability to serve our users

...

the project owes first and foremost to its users.

...

Andreas Tille was, on two successive occasions, elected on Hitler's birthday. Go figure.

misc fedora bits 3nd week of feb 2026

2026-02-21T19:27:52Z

Well, another saturday, time for another bit of longer form recapping what has been going on in fedora infrastructure and other areas for me.

Infrastructure Fedora 44 Beta freeze

We started the beta freeze in infrastructure. This is to make sure that we don't cause any problems for the release building and distributing pipeline. We require some acks for any changes to things that might impact those things until the day after the Beta is released.

I think this has served us fine over the years. Every once in a while I wonder if we could just stop doing it as we are usually pretty good about not breaking things day to day, but having the extra eyes on changes and slowing down a bit is a good thing I think.

Forge migrations

We have been busy working on migrating things from pagure.io to forge.fedoraproject.org. On tuesday just before the freeze we finally got our ansible repo moved over. I've really been looking forward to this as the review interface in forgejo is a good deal nicer than the pagure one. I've already used it to great effect.

We do still have a few more things to migrate, but overall it's moving along nicely.

Last bits of rdu2-cc move

We finally finished off the last things (at least that I am aware of) for things we moved in last december from rdu2-cc to rdu3.

There was a very strange and difficut to figure out problem for copr builders on ipv6 that I wasn't able to track down, but luckily Pavel worked with networking and finally did so! It seems to have been a odd caching bug in the switches. Hopefully it's now gone once and for all.

There was some hardware issues to sort out: some bad network cards that had to be replaced, a machine that didn't actually move when it was supposed to, etc.

Anyhow I hope all that work is all finally done.

signing work

Finally got back to deploying / testing the new signing path for secure boot signing. I got it all deployed, just need to get things tested now and hopefully we can switch over after the freeze.

This should hopefully allow us to sign aarch64 kernels for secure boot as well as removing reliance on an old smart card for signing.

comments? additions? reactions?

As always, comment on mastodon: https://fosstodon.org/@nirik/116110354434738317

Pourquoi j’ai basculé de Portainer vers Arcane pour gérer les conteneurs sur le NAS

2026-02-21T09:37:36Z

Si tu me lis un peu, tu sais que j’ai une attitude assez pragmatique vis-à-vis de mes outils. Je n’aime pas changer pour changer, mais je n’hésite pas non plus à bouger quand une solution stagne trop ou prend une direction qui ne me plaît pas. Récemment, j’ai d’ailleurs confirmé mon choix de rester sous […]

Cet article Pourquoi j’ai basculé de Portainer vers Arcane pour gérer les conteneurs sur le NAS est apparu en premier sur Guillaume Kulakowski's blog.

Community Update – Week 8 2026

2026-02-20T10:00:00Z

Week: 16 Feb – 20 Feb 2026

Fedora Infrastructure

This team is taking care of day to day business regarding Fedora Infrastructure.
It’s responsible for services running in Fedora infrastructure.
Ticket tracker

[GSoC Project Idea 2026] Revamp Fedora Badges project with modern fullstack architecture and dedicated MCP support [Ticket] [Followup]
[Infra] Added package and installed size to package metadata [Review] [Lint]
[Infra] Improve vagrant setup instructions and add container-based setup [Followup A] [Followup B]
Migration of pagure.io repositories to forge.fedoraproject.org continues (9 more repositories migrated)
Resolved authentication issues with wordpress instances (thanks to misc)
Fixed database connection issues on Dist-Git
Dep updates and CI fixes for our apps in Github
Worked on the port of bugzilla2fedmsg to Kafka (since the UMB deprecation), deployed it to staging, asked RHIT for firewall ports.

CentOS Infra including CentOS CI

Release Engineering

This team is taking care of day to day business regarding Fedora releases.
It’s responsible for releases, retirement process of packages and package builds.
Ticket tracker

Fedora 44 Beta Freeze is now in effect.

RISC-V

This is the summary of the work done regarding the RISC-V architecture in Fedora.

(Not a lot to report this week, besides the routine on-going work.)
Started a discussion with the RISC-V team about RHEL builders for Konflux. (This is not about general Konflux support, that’s out of scope)
Continued to investigate Fedora 44 build failures and all that entails — working with relevant upstream maintainers to get changes reviewed, merged, etc.
Sorted out a build-timeout issue with Copr upstream. (Jason Montleon is currently used to build some board-specific kernels.)

QE

This team is taking care of quality of Fedora. Maintaining CI, organizing test days
and keeping an eye on overall quality of Fedora releases.

TestDays App was updated in production.
Anubis no longer breaks actions in Forge thanks to our debugging (and Infra fixing it, of course).
Blockerbugs meetings and the whole blocker review process has started since this week.
Ran test days: Grub OOM fix, GNOME 50

Forgejo

This team is working on introduction of https://forge.fedoraproject.org to Fedora
and migration of repositories from pagure.io.

Spike of zabbix installation for ansible-role-forgejo-runner researched
Forge image build and versioning fully automated
Runners updated to use image version 12
More migrations from pagure and troubleshooting
Private Issues: Fixed tests, tied up loose ends and consolidated feature branch.
Packit team got access to staging distgit deployment
Create new organisation and namespace on Fedora Forge for Fedora Join [Followup] [Namespace]

UX

This team is working on improving User experience. Providing artwork, user experience,
usability, and general design services to the Fedora project

CLE logo complete!
F45 wallpaper mindmap session took place. Mindmap created can be found here.
Continuing Forgejo migration.

If you have any questions or feedback, please respond to this report or contact us on #admin:fedoraproject.org channel on matrix.

The post Community Update – Week 8 2026 appeared first on Fedora Community Blog.

GNOME is participating in Google Summer of Code 2026!

2026-02-20T06:58:52Z

Potential GSoC contributors may reach out with questions about our project ideas or GNOME internships in general. Please direct them to gsoc.gnome.org to learn more.

You can find our proposed project ideas at gsoc.gnome.org/2026.

Project proposal submissions are open from March 16th to 31st.

GNOME Foundation Update, 2026-02-19

2026-02-19T15:01:05Z

Welcome to another GNOME Foundation update post, covering highlights from the past two weeks (this week and last week). It’s been a busy time, particularly due to conference planning and our upcoming audit – read on to find out more!

Linux App Summit 2026

We were thrilled to be able to announce the location and dates of this year’s Linux App Summit this week. The conference will happen in Berlin on the 16th and 17th of May, at Betahaus Berlin. More information is available on the LAS website.

As usual, we are very pleased to be collaborating with KDE on this year’s LAS. Our partnership on LAS has been a real success that we hope to continue.

Travel sponsorship for LAS 2026 is available for Foundation members through the Travel Committee, so head over to the travel page if you would like to attend and need financial support.

February’s Board meeting

The Board of Directors it’s regular monthly meeting last week, on 9th February. Highlights from the meeting included:

We finally caught up on our minutes, approving the minutes from a total of nine meetings. This was a big relief, and hopefully we will be able to stay on top of the minutes now that we’re caught up.
The Board was thrilled to formally add Nirbheek Chauhan as a member of the Travel Committee. Many contributors will know Nirbheek as a longstanding GStreamer hacker, and he’s already been doing some great work to help with travel. Thanks Nirbheek!
The Board approved a new document retention and destruction policy, which is something that we are encouraged to have by regulators.
I gave an update on the operational highlights from the last month, including fundraising, conference planning, and audit preparation.
The Board considered a proposal for an exciting new program that we’re hoping to launch very soon. More details to follow soon.

The next Board meeting is scheduled for March 9th.

Audit submissions

As I’ve mentioned in previous updates, the GNOME Foundation is due to be audited very soon. This is a routine occurrence for non-profits like us, but this is our first formal audit, so there’s a good deal of learning and setup to be done.

Last week was the deadline to submit all the documentation for the audit, which meant that many of us were extremely busy finalising numbers, filling in spreadsheets, and tidying up other documentation ready to send it all to the auditors.

Our finance team *really* went the extra mile for us to get everything ready on time, so I’d like to give them a huge thank you for helping us out.

The audit inspection itself will happen in the first week of March, so preparations continue, as we assemble and organise our records, update our policies, and so on.

GUADEC 2026

Planning for this summer’s conference has continued over the past two weeks. In case you missed it, the location and dates have been announced, and accommodation bookings are open at a reduced rate. In the background we are gearing up to open the call for papers, and the sponsorship effort is on its way. Now is a good time to start thinking about any talk proposals that you’d like to submit.

Membership certificates

A cool community effort is currently underway to provide certificates for GNOME Foundation members. This is a great idea in my opinion, as it will allow contributors to get official recognition which can be used for job applications and so on. More volunteers to help out would definitely be welcome.

That’s it for this week. Thanks for reading, and feel free to ask questions in the comments.

Community Edition Explained

2026-02-18T08:20:00Z

This page explains what Kiwi TCMS Community Edition is, how it ships and what risks are associated with it. Please read about the details below.

What is Kiwi TCMS Community Edition

This is the official version of the Kiwi TCMS application as produced by our own team with the help of many contributors. It may also be referred to upstream or the community edition version and comes packaged as a container image which is publicly available and can be downloaded via the docker pull pub.kiwitcms.eu/kiwitcms/kiwi command!

Who is this for

Community Edition is suitable for developers, teams and organizations which are happy to run Kiwi TCMS without any warranty and safeguards for development, testing and even production purposes!

What do you get

You get access the publicly available container image; public documentation and source code but not much else!

What do all of the individual items mean

Upstream container image: Kiwi TCMS is packaged as a single container image. See Running Kiwi TCMS as a container to get started!

Only x86_64 build: the community edition container is built on Linux, suitable for Intel/AMD 64bit processors only.

No version tags: the community edition container is always the latest version! There are no other versions available!

GPL-2.0 license: Kiwi TCMS is an open source software, with very long history, with its primary core licensed under the GNU GPL-2.0 license!

No warranty: the community edition version of Kiwi TCMS is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

EthicalAds: the community edition version of Kiwi TCMS comes with built-in advertisement from EthicalAds rewards from which are paid out to opencollective.com/kiwitcms for transparency.

You assume all risk: you must understand and accept that open source comes with associated risks. To name just a few: our team disappears and stops development; security issues stay unresolved; code slowly becomes outdated & may incur technical debt in the case of one-off patches! All of these have happened in the past with the predecessor of what is now Kiwi TCMS.

There is no charge but please understand that software isn't zero cost! You pay for it indirectly with time and effort invested in engineering, operations, maintenance and resolving any issues which may arise from the use of said software, regardless of whether it is open source or not!

IMPORTANT: open source and Kiwi TCMS by extension, provide you with more options when it comes to mitigating risks and cost compared to proprietary software. However the easiest way to secure the future of Kiwi TCMS is to become a customer and help us sustain development!

Happy Testing!

If you like what we're doing and how Kiwi TCMS please help us!

Give â� on GitHub;
Join our newsletter and follow all project news;
Become a contributor and an awesome open source hacker;
Become a customer and help us sustain development

Private Tenant Extras Subscription Explained

2026-02-17T08:35:00Z

This page explains what Private Tenant Extras is and how it brings more value to your existing Kiwi TCMS subscription. Please read about the details below.

What is Private Tenant Extras by Kiwi TCMS

This is an optional subscription tier which combines our existing Private Tenant SaaS hosting with access to the underlying data in its raw format!

IMPORTANT: any Private Tenant subscription with unit count > 1 entitles you to this Extras add-on! Please contant support for setup!

Who is this subscription for

Private Tenant Extras is suitable for teams using the SaaS version of Kiwi TCMS which require access to their underlying data in machine readable format!

What do you get

Everything from lower tier subscription plans plus access to a database and file exports in case you would like to keep your own backup copy or provide in-house integration with other tools.

What do all of the individual items mean

Raw SQL database export: you will receive a database export in SQL format, suitable for the Postgres database engine. It includes all tables which constitute your own namespace under the *.tenant.kiwitcms.org domain name. This also includes information about user accounts authorized to access your tenant!

IMPORTANT: due to technical and security limitations we cannot give you direct access to the underlying database cluster in real-time!

File uploads included: means exactly that - all attachments uploaded to your private tenant will be included as part of this subscription!

Encrypted access: all data we export is stored encrypted and may be accessed using the popular open source tool restic. We will provide you with read-only level access and an unique public/private keypair!

IMPORTANT:

Export frequency: 1/day (max 3/day)
Exports on disk: last 3 days
Retention period: 7 days

Storage region of your choice: means that we can publish to a geographic region and data center of your choice. Exact locations are subject to availability.

NOTE: At the time of writing our preferred storage backend is Amazon S3.

Happy Testing!

If you like what we're doing and how Kiwi TCMS please help us!

Give â� on GitHub;
Join our newsletter and follow all project news;
Become a contributor and an awesome open source hacker;
Become a customer and help us sustain development

The Performance and Scale Booth at DevConf 2026

2026-02-16T09:03:47Z

This initiative was initially proposed by Jaison Raju, who had successfully organized a similar booth at DevConf 2025. As Jaison was unavailable to lead this year, he requested that I take charge of the effort.

Despite an already demanding schedule, my commitment to continuing this valuable presence led me to accept the leadership role.I attended a Booth Organizers meeting with Rajan Shah to establish a clear understanding of the necessary tasks and responsibilities.

From the wide array of projects within Performance and Scale meetup, we decided to feature the Krkn project for the booth, given its recent inclusion in the CNCF organization.

Further information regarding the Krkn project is available here: Link.

Darshan Jain, who recently began working on Krkn, demonstrated exceptional dedication to the booth's success. He took the lead on critical tasks, including planning the presentation slides, recording demonstrations, and onboarding Shashank Kestwal as a member of the booth staff.

On the day of the event, a parallel talk on the LogAn project was also scheduled: Link. I had initially planned for Darshan and Shashank to manage the booth while I attended the LogAn talk. However, the impressive level of engagement and attendance at the Krkn booth prompted me and Nikhil Jain to join the staff. This provided an opportunity for stimulating discussions covering Krkn, general performance and scale topics, guidance on contributing to open source, and various other subjects.

Booth swag is often an essential component of these events. Unfortunately, a shortage of Performance and Scale stickers necessitated an alternative, and we opted to distribute chocolates as our promotional item, coincidentally aligning with Valentine's Day. This solution proved to be well-received by attendees.

I extend my gratitude to Krishna Magar, Aman Vishwakarma,

Rajaditya Chauhan and Nikhil Jain for their presence and support at the booth.

Significant credit is due to Darshan Jain for the meticulous planning and smooth execution of the booth, and to Shashank Kestwal for his excellent support from planning to execution.

misc fedora bits 2nd week of feb 2026

2026-02-14T18:20:52Z

Another weekly recap of happenings around fedora for me.

Strange long httpd reload times on proxy11

I spent a fair bit of time looking at one of our proxies. We have them all to a reload (aka 'graceful restart') every hour when we update a ticketkey on them. For the vast majority of them, thats fine and works as expected. However, proxy11 decided to start taking a while (like 12-15seconds) to reload, causing our monitoring to alert that it was down... then back up.

In the end, it seemed the problem was somehow related to some old tls certificates that were present, but not used anywhere. All I can think of is that it's doing some kind of parsing of all certs and somehow those old ones cause it undue processing time. I removed those old certs and reload times went way back down again.

I'm tempted to try and figure out what it's doing exactly here, but I already spent a fair bit of time on it and it's working again now, so I guess I will just shrug and move on.

Anubis and download servers

A while back I had to hurredly deploy anubis in front of our download servers. This was due to the scrapers deciding to just download every rpm / iso from every fedora release since the dawn of time at a massive concurrency. This was saturating one of our 10G links completely, and making another somewhat full. So, I deployed anubis and it dropped things back to 'normal' again.

Fast forward to this last week, and my rush in deploying anubis came back to bite me. We have a cloudfront distribution that uses our download servers as it's 'origin'. Then we point all aws network blocks to use that for any fedora instances in aws. This is a win for us as then everything for them is cached on the aws side saving bandwith, and a win for aws users as that traffic is 'local' to them so faster and doesn't cause them to need to be billed for ingress either.

Last week, anubis started blocking CloudFront, so uses in aws would get a anubis challenge page instead of the actual content they were expecting. But why did it this just happen now? well, as near as I could determine, someone/scrapers were hitting the CloudFront endpoints and crawling our download server (fine, no problem there), but then they hit a directory that they handled poorly.

The directory was used/last updated about 11 years ago with a readme file explaining that the content was moved and no longer there. Great. However, also it had previous subdirectories as links to '.' (ie, the current directory). Since scrapers don't use any of the 20 years of crawling code, and instead just brute force things, this resulted in a bunch of requests like:

GET /foo/ GET /foo/foo/ GET /foo/foo/foo/

and so on. These are all really small (just a directory listing), so that meant it could make requests really really fast. So, after some point anubis started challenging those CloudFront connections and boom.

So, the problem with the hurred deployment I had made there was that The policy file I had deployed was not actually being used. I had allowed CloudFront, but it didn't seem to help any, and it took me far too long to figure out that anubis was starting up, printing one error about not being able to read the policy file and just running with the default configuration. ;( It turned out be a podman/selinux interaction and is now fixed.

I also removed those . links and set that directory tree to just 403 all requests to it.

Anubis and forge

Also this week, folks were reporting problems with our new forgejo forge. Anubis was doing challenges when people were trying to submit comments and it was messing them up.

In the end here, I just needed to adjust the config to allow POSTs through. At least right now scrapers aren't doing any POSTS and just allowing those seems to fix the issues people were having.

Some more scrapers

Friday we had them hitting release-monitoring.org. This time it was what I am calling a 'type 0' scraper. It was all coming from one cloud ip and I could just block them.

This morning a bit ago, we had a group hit/find the 'search' button on koji.fedoraproject.org, taking it offline. I was able to block the endpoint for a few hours and they went away, but no telling if they will be back. These were the 'type 2' kind (botnet using users ip's/browsers from 100's of thousands of different ips).

I am sad that the end game here sounds like there's not going to be so much of a open internet anymore. ie, for self defense sites will all have to go to requiring registration of some kind before working. I can only hope business models change before it comes to that.

comments? additions? reactions?

As always, comment on mastodon: https://fosstodon.org/@nirik/116070476999694239

drm subsystem AI patch review

2026-02-13T06:56:33Z

This topic came up at kernel maintainers summit and some other groups have been playing around with it, particularly the BPF folks, and Chris Mason's work on kernel review prompts[1] for regressions. Red Hat have asked engineers to investigate some workflow enhancements with AI tooling, so I decided to let the vibecoding off the leash.

My main goal:

- Provide AI led patch review for drm patches

- Don't pollute the mailing list with them at least initially.

This led me to wanting to use lei/b4 tools, and public-inbox. If I could push the patches with message-ids and the review reply to a public-inbox I could just publish that and point people at it, and they could consume it using lei into their favorite mbox or browse it on the web.

I got claude to run with this idea, and it produced a project [2] that I've been refining for a couple of days.

I started with trying to use Chris' prompts, but screwed that up a bit due to sandboxing, but then I started iterating on using them and diverged.

The prompts are very directed at regression testing and single patch review, the patches get applied one-by-one to the tree, and the top patch gets the exhaustive regression testing. I realised I probably can't afford this, but it's also not exactly what I want.

I wanted a review of the overall series, but also a deeper per-patch review. I didn't really want to have to apply them to a tree, as drm patches are often difficult to figure out the base tree for them. I did want to give claude access to a drm-next tree so it could try apply patches, and if it worked it might increase the review, but if not it would fallback to just using the tree as a reference.

Some holes claude fell into, claude when run in batch mode has limits on turns it can take (opening patch files and opening kernel files for reference etc), giving it a large context can sometimes not leave it enough space to finish reviews on large patch series. It tried to inline patches into the prompt before I pointed out that would be bad, it tried to use the review instructions and open a lot of drm files, which ran out of turns. In the end I asked it to summarise the review prompts with some drm specific bits, and produce a working prompt. I'm sure there is plenty of tuning left to do with it.

Anyways I'm having my local claude run the poll loop every so often and processing new patches from the list. The results end up in the public-inbox[3], thanks to Benjamin Tissoires for setting up the git to public-inbox webhook.

I'd like for patch submitters to use this for some initial feedback, but it's also something that you should feel free to ignore, but I think if we find regressions in the reviews and they've been ignored, then I'll started suggesting it stronger. I don't expect reviewers to review it unless they want to. It was also suggested that perhaps I could fold in review replies as they happen into another review, and this might have some value, but I haven't written it yet. If on the initial review of a patch there is replies it will parse them, but won't do it later.

[1] https://github.com/masoncl/review-prompts

[2] https://gitlab.freedesktop.org/airlied/patch-reviewer

[3] https://lore.gitlab.freedesktop.org/drm-ai-reviews/

⚙️ PHP version 8.4.18 and 8.5.3

2026-02-13T05:42:00Z

RPMs of PHP version 8.5.3 are available in the remi-modular repository for Fedora â‰¥ 42 and Enterprise Linux â‰¥ 8 (RHEL, Alma, CentOS, Rocky...).

RPMs of PHP version 8.4.18 are available in the remi-modular repository for Fedora â‰¥ 42 and Enterprise Linux â‰¥ 8 (RHEL, Alma, CentOS, Rocky...).

â„¹ï¸� These versions are also available as Software Collections in the remi-safe repository.

â„¹ï¸� The packages are available for x86_64 and aarch64.

â„¹ï¸� There is no security fix this month, so no update for versions 8.2.30 and 8.3.30.

Version announcements:

â„¹ï¸� Installation: Use the Configuration Wizard and choose your version and installation mode.

Replacement of default PHP by version 8.5 installation (simplest):

On Enterprise Linux (dnf 4)

dnf module switch-to php:remi-8.5/common

On Fedora (dnf 5)

dnf module reset php
dnf module enable php:remi-8.5
dnf update

Parallel installation of version 8.5 as Software Collection

yum install php85

Replacement of default PHP by version 8.4 installation (simplest):

On Enterprise Linux (dnf 4)

dnf module switch-to php:remi-8.4/common

On Fedora (dnf 5)

dnf module reset php
dnf module enable php:remi-8.4
dnf update

Parallel installation of version 8.4 as Software Collection

yum install php84

And soon in the official updates:

Fedora Rawhide now has PHP version 8.5.3
Fedora 44 - PHP 8.5.3
Fedora 43 - PHP 8.4.18
Fedora 42 - PHP 8.4.17

âš ï¸� To be noticed :

EL-10 RPMs are built using RHEL-10.1
EL-9 RPMs are built using RHEL-9.7
EL-8 RPMs are built using RHEL-8.10
intl extension now uses libicu74 (version 74.2)
mbstring extension (EL builds) now uses oniguruma5php (version 6.9.10, instead of the outdated system library)
oci8 extension now uses the RPM of Oracle Instant Client version 23.26 on x86_64 and aarch64
A lot of extensions are also available; see the PHP extensions RPM status (from PECL and other sources) page

â„¹ï¸� Information:

Base packages (php)

Software Collections (php83 / php84 / php85)

Cockpit 356

2026-02-11T00:00:00Z

Cockpit is the modern Linux admin interface. We release regularly.

Here are the release notes from Cockpit 356:

Services: Timers created by Cockpit run shell commands and can be edited

The systemd timer units created by Cockpit now run their command via the shell. Previously, the text you entered into the “Command” field was used directly for the ExecStart value in the systemd service unit for the timer, and was thus subject to systemd specifier expansion and other idiosyncrasies. Now the text you enter is executed directly by /bin/sh.

Systemd timer units that have been created by this version of Cockpit (or later) can now also be edited in Cockpit.

Thanks to Miguel Ribeiro for contributing this feature!

Try it out

Cockpit 356 is available now:

Zettelkasten on a Supernote

2026-02-10T19:25:20Z

I love the Ratta Supernote e-ink devices. They are a delight to use for writing, planning and all sorts of things. Writing things out by hand helps me connect with them on a deeper level because I need to choose the words more carefully than I would when typing, simply because of the speed difference. The Supernote’s onboard handwriting recognition works really well for my handwriting style.

At work I also love using Logseq, an open source personal knowledge management tool and outliner. I use it to write down things I learn and ideas I have, then connect things together in interesting ways in order to gain deeper understanding.

That’s the experience I wanted on the Supernote. Then I discovered Zettelkasten and knew it was what I wanted to use. ZK is a paper-based method and there have been adaptations to all sorts of environments, both physical and virtual. When I started I found it difficult to figure out how to make it feel native on a Supernote.

I also knew I wanted to be able to export my Supernote Zettelkasten as text in case I wanted to process it on a computer in different ways, or even run an AI assistant on it.

My first attempt was pretty messy! I’d create a new handwritten notebook file for each idea. Each filename had a timestamp (pre-filled by Supernote when you create a new file) along with a short title.

I found pysn-digest, a tool which is able to convert Supernote notebooks into Markdown files. So I worked out an elaborate system in which each notebook (each idea) had a level one heading for the title, then level two headings for metadata field names like “type” or “relates to”, then a level one heading again to start the content. The filenames were handwritten (timestamps and all) and I made them links to the idea files. It was really process-heavy and I didn’t stick with it for long before realising I needed to improve it.

Many months later I have evolved this system into a framework I’m really pleased with. It uses only level one headings and all the ideas can be in a single notebook (or split across notebooks if preferred). There are no complicated timestamps or numbering systems, only unique titles.

Idea notes can link together arbitrarily but also cluster together neatly, as well as nest as deeply as I need them to. I don’t have to keep to the Supernote’s native limit of four heading levels. And thanks to a system of templates and stickers I designed, I can easily see how deep in the tree I am, even from the pages overview. This means it is fast to insert a new idea note into the right place.

I’ve called it Slipstream. If you have a Supernote and want to skip the trial and error, Iâ€™ve packaged the entire framework (PDF user guide, templates, stickers, quick reference and notebooks) into a ready-to-use kit. Download the full Slipstream kit here.

The post Zettelkasten on a Supernote appeared first on PRINT HEAD.

misc fedora bits 1st week of feb 2026

2026-02-07T18:34:43Z

Welcome to a bit of recap of the first week of February. It will be a shorter one today...

Fedora 44 Branching

The big news this week was the Fedora 44 branching off rawhide. This is by far the most complicated part of the release. There's updates that have to happen in a ton of places all in the right order and with the right content.

Things didn't start when they were supposed to (tuesday morning), because we had some last minute mass rebuilds (golang and ghc). Then, they didn't start wed morning because we were trying to get the gnome 50 update to pass gating. Finally on thursday we just ended up unpushing that update and starting the process.

This time the releng side was run by Patrik. It's the first time he's done this process, but he did a great job! He asked questions at each step and we were able to clarify and reorder the documetation so I hope things will be even more clear and easy next cycle.

You can see the current SOP on it (before changes from this cycle): https://docs.fedoraproject.org/en-US/infra/release_guide/sop_mass_branching/ Look at all those steps!

This was also a bit of a long week because I am in PST and patrik is in CET, so I had to get up early and he had to stay late. Timezones are anoying. :)

Anyhow, I think things went quite smoothly. We got rawhide and branched composes right away, and only a few minor items to clean up and figure out how to do better.

Sprint planning meeting again monday

We had our last sprint planning meeting almost two weeks ago, so on monday it's time for another one. We did manage to run it in matrix, and although we did run over time I think it went not too badly.

I'll probibly do some prep work on things this weekend for it.

But if anyone wants to join in/read back it will be in #meeeting-3:fedoraproject.org at 15UTC on matrix.

comments? additions? reactions?

As always, comment on mastodon: https://fosstodon.org/@nirik/116030844840004998

GNOME Foundation Update, 2026-02-06

2026-02-06T18:00:11Z

Welcome to another GNOME Foundation weekly update! FOSDEM happened last week, and we had a lot of activity around the conference in Brussels. We are also extremely busy getting ready for our upcoming audit, so there’s lots to talk about. Let’s get started.

FOSDEM

FOSDEM happened in Brussels, Belgium, last weekend, from 31st January to 1st February. There were lots of GNOME community members in attendance, and plenty of activities around the event, including talks and several hackfests. The Foundation was busy with our presence at the conference, plus our own fringe events.

Board hackfest

Seven of our nine directors met for an afternoon and a morning prior to FOSDEM proper. Face to face hackfests are something that the Board has done at various times previously, and have always been a very effective way to move forward on big ticket items. This event was no exception, and I was really happy that we were able to make it happen.

During the event we took the time to review the Foundation’s financials, and to make some detailed plans in a number of key areas. It’s exciting to see some of the initiatives that we’ve been talking about starting to take more shape, and I’m looking forward to sharing more details soon.

Advisory Board meeting

The afternoon of Friday 30th January was occupied with a GNOME Foundation Advisory Board meeting. This is a regular occurence on the day before FOSDEM, and is an important opportunity for the GNOME Foundation Board to meet with partner organizations and supporters.

Turn out for the meeting was excellent, with Canonical, Google, Red Hat, Endless and PostmarketOS all in attendance. I gave a presentation on the how the Foundation is currently performing, which seemed to be well-received. We then had presentations and discussion amongst Advisory Board members.

I thought that the discussion was useful, and we identified a number of areas of shared interest. One of these was around how partners (companies, projects) can get clear points of contact for technical decision making in GNOME and beyond. Another positive theme was a shared interest in accessibility work, which was great to see.

We’re hoping to facilitate further conversations on these topics in future, and will be holding our next Advisory Board meeting in the summer prior to GUADEC. If there are any organizations out there would like to join the Advisory Board, we would love to hear from you.

Conference stand

GNOME had a stand during both FOSDEM days, which was really busy. I worked the stand on the Saturday and had great conversations with people who came to say hi. We also sold a lot of t-shirts and hats!

I’d like to give a huge thank you to Maria Majadas who organized and ran our stand this year. It is incredibly exhausting work and we are so lucky to have Maria in our community. Please say thank you to her!

We also had plenty of other notable volunteers, including Julian Sparber, Ignacy KuchciÅ„ski, Sri Ramkrishna. Richard Litteaur, our previous Interim Executive Director even took a shift on the stand.

Social

On the Saturday night there was a GNOME social event, hosted at a local restaurant. As always it was fantastic to get together with fellow contributors, and we had a good turnout with 40-50 people there.

Audit preparation

Moving on from FOSDEM, there has been plenty of other activity at the Foundation in recent weeks. The first of these is preparation for our upcoming audit. I have written a fair bit about this in these previous updates. The audit is a routine exercise, but this is also our first, so we are learning a lot.

The deadline for us to provide our documentation submission to the auditors is next Tuesday, so everyone on the finance side of the operation has been really busy getting all that ready. Huge thanks to everyone for their extra effort here.

GUADEC & LAS planning

Conference planning has been another theme in the past few weeks. For GUADEC, accommodation options have been announced, artwork has been produced, and local information is going up on the website.

Linux App Summit, which we co-organise with KDE, has been a bit delayed this year, but we have a venue now and are in the process of finalizing the budget. Announcements about the dates and location will hopefully be made quite soon.

Google verification

A relatively small task, but a good one to highlight: this week we facilitated (ie. paid for) the assessment process for GNOME’s integration with Google services. This is an annual process we have to go through in order to keep Evolution Data Server working with Google.

Infrastructure optimization

Finally, Bart, along with Andrea, has been doing some work to optimize the resource usage of GNOME infrastructure. If you are using GNOME services you might have noticed some subtle changes as a result of this, like Anubis popping up more frequently.

That’s it for this week. Thanks for reading; I’ll see you next week!

Trying Claude Code prompt injections (Feb 2026)

2026-02-06T10:00:00Z

For quite some time I’ve wanted to test how prone agentic tools are to prompt injection. Let’s go.

I’ll be using Claude Code 2.1.5, 4.5 Opus in various different sessions.

💎 PHPUnit 13

2026-02-06T07:59:00Z

RPMs of PHPUnit version 13 are available in the remi repository for Fedora ≥ 42 and Enterprise Linux (CentOS, RHEL, Alma, Rocky...).

Documentation :

ℹ️ This new major version requires PHP ≥ 8.4 and is not backward compatible with previous versions, so the package is designed to be installed beside versions 8, 9, 10, 11, and 12.

Installation:

dnf --enablerepo=remi install phpunit13

Notice: This tool is an essential component of PHP QA in Fedora. This version should be available soon in the Fedora ≥ 43 official repository (19 new packages).

nouveau: a tale of two bugs

2026-02-04T21:04:16Z

Just to keep up some blogging content, I'll do where did I spend/waste time last couple of weeks.

I was working on two nouveau kernel bugs in parallel (in between whatever else I was doing).

Bug 1: Lyude, 2 or 3 weeks ago identified the RTX6000 Ada GPU wasn't resuming from suspend. I plugged in my one and indeed it wasn't. Turned out since we moved to 570 firmware, this has been broken. We started digging down various holes on what changed, sent NVIDIA debug traces to decode for us. NVIDIA identified that suspend was actually failing but the result wasn't getting propogated up. At least the opengpu driver was working properly.

I started writing patches for all the various differences between nouveau and opengpu in terms of what we send to the firmware, but none of them were making a difference.

I took a tangent, and decided to try and drop the latest 570.207 firmware into place instead of 570.144. NVIDIA have made attempts to keep the firmware in one stream more ABI stable. 570.207 failed to suspend, but for a different reason.

It turns out GSP RPC messages have two levels of sequence numbering, one on the command queue, and one on the RPC. We weren't filling in the RPC one, and somewhere in the later 570's someone found a reason to care. Now it turned out whenever we boot on 570 firmware we get a bunch of async msgs from GSP, with the word ASSERT in them with no additional info. Looks like at least some of those messages were due to our missing sequence numbers and fixing that stopped those.

And then? still didn't suspend/resume. Dug into memory allocations, framebuffer suspend/resume allocations. Until Milos on discord said you did confirm the INTERNAL_FBSR_INIT packet is the same, and indeed it wasn't. There is a flag bEnteringGCOff, which you set if you are entering into graphics off suspend state, however for normal suspend/resume instead of runtime suspend/resume, we shouldn't tell the firmware we are going to gcoff for some reason. Fixing that fixed suspend/resume.

While I was head down on fixing this, the bug trickled up into a few other places and I had complaints from a laptop vendor and RH internal QA all lined up when I found the fix. The fix is now in drm-misc-fixes.

Bug 2: A while ago Mary, a nouveau developer, enabled larger pages support in the kernel/mesa for nouveau/nvk. This enables a number of cool things like compression and gives good speedups for games. However Mel, another nvk developer reported random page faults running Vulkan CTS with large pages enabled. Mary produced a workaround which would have violated some locking rules, but showed that there was some race in the page table reference counting.

NVIDIA GPUs post pascal, have a concept of a dual page table. At the 64k level you can have two tables, one with 64K entries, and one with 4K entries, and the addresses of both are put in the page directory. The hardware then uses the state of entries in the 64k pages to decide what to do with the 4k entries. nouveau creates these 4k/64k tables dynamically and reference counts them. However the nouveau code was written pre VMBIND, and fully expected the operation ordering to be reference/map/unmap/unreference, and we would always do a complete cycle on 4k before moving to 64k and vice versa. However VMBIND means we delay unrefs to a safe place, which might be after refs happen. Fun things like ref 4k, map 4k, unmap 4k, ref 64k, map 64k, unref 4k, unmap 64k, unref 64k can happen, and the code just wasn't ready to handle those. Unref on 4k would sometimes overwrite the entry in the 64k table to invalid, even when it was valid. This took a lot of thought and 5 or 6 iterations on ideas before we stopped seeing fails. In the end the main things were to reference count the 4k/64k ref/unref separately, but also the last thing to do a map operation owned the 64k entry, which should conform to how userspace uses this interface.

The fixes for this are now in drm-misc-next-fixes.

Thanks to everyone who helped, Lyude/Milos on the suspend/resume, Mary/Mel on the page tables.

JuristGate is for sale: three billion Swiss francs for a domain name

2026-02-04T18:30:00Z

When the share price of Credit Suisse, one of Switzerland's two top banks, went into its death spiral, the Swiss authorities had some discussions among themselves and the inevitable result of that discussion was the sale of Credit Suisse to their main rival, UBS.

The public were told that UBS had offered three billion Swiss francs to compensate the shareholders of Credit Suisse for their shares. Some people were sceptical about the method used to reach this valuation. Nonetheless, it was important for Swiss national pride.

When writing the reports on the JuristGate web site, I've attempted to operate with the highest level of accuracy and integrity. Nonetheless, everybody has their price. If somebody offered me three billion Swiss francs for the domain then I would probably do like Urban Angehrn and Birgit Rutishauser and take a lengthy garden leave.

In the meantime, to make up for the wayward support from Swiss legal insurance I've started a crowdfunding campaign to resolve one of the disputes relating to Debian. This is vital because we all do a lot of work for Debian and we are all entitled to equal recognition.

In a previous blog, I looked at the prompt and efficient manner in which Bernice is undertaking enforcement action to protect the public from rogue health practitioners in the State of Victoria.

Bernice generated a lot of news stories when she banned a social media influencer, Emily Lal, who promoted herself as The Authentic Birthkeeper. Lal had promoted wild birthing. A number of mothers and babies have died around the world when following the advice of social media influencers.

I don't know how much Bernice gets paid but I found a job vacancy for her deputy. The salary is in the range up to $290,000 per year. That is approximately CHF 160,000 Swiss francs.

As Bernice is a nurse and keyworkers never get paid what they are really worth I'm guessing her salary is not too much bigger than that.

The FINMA annual reports reveal the salaries for the CEO and executive team. Urban Angehrn's salary was CHF 602,000. It looks like FINMA pays their CEO three or four times what the State of Victoria pays Bernice.

We can compare their performance and see how many banning orders each of them has published:

FINMA (Swiss financial regulator) banning orders: they published 24 bans between 2018 and 2025. Some bans are not reported publicly.

Health Complaints Commissioner (Victoria) banning orders: Bernice has personally signed over 100 interim or permanent bans since taking office in July 2022.

This was Urban Angehrn's payout, CHF 581,000, when he resigned at the same time that FINMA published their anonymous decision about Parreaux, Thiebaud & Partners. Clients/victims were offered nothing.

Generations (take N+1)

2026-02-03T21:49:36Z

Putting some rigour to generations

Recently a coworker posted that children born this year would be in Generation Beta, and I was like “What? That sounds like too soon…” but then thought “Oh its just that thing when you get older and time flies by.” I saw a couple of articles saying it again, so decided to look at what was on the wikipedia article for generations and saw that yes ‘beta’ was starting.. then I started looking at the lengths of the various generations and went “Hold On”.

Wikipedia_graphic

Let us break this down in a table:

Generation	Wikipedia	How Long
T (lost)	1883-1900	17
U (greatest)	1901-1927	26
V (silent)	1928-1945	17
W (boomer)	1946-1964	18
X	1965-1980	15
Y (millenial)	1981-1996	15
Z	1997-2012	15
alpha	2013-2025	12
beta	2026-2039	13
gamma	2040-???	??

So it is bad enough that Generation X,Millenials, and Z got shortened from 18 years to 15.. but alpha and beta are now down to 12 and 13? I realize that this is because all of this is a made up construct to make some people born in one age group angry/sad/afraid in another by editors who are needing to sell advertising for things which will solve the feelings of anger, sadness, or fear.. but could you at least be consistent.

I personally like some order to my starting and ending dates for generations so I am going to update some lists I have put out in the past with newer titles and times. We will use the definiton as outlined at https://en.wikipedia.org/wiki/Generation

A generation is all of the people born and living at about the same time, regarded collectively.[1] It also is “the average period, generally considered to be about 20–30 years, during which children are born and grow up, become adults, and begin to have children.”

For the purpose of trying to set eras, I think that the original 18 years for baby boomers makes sense, but the continual shrinkflation of generations after that is pathetic. So here is my proposal for generation ending dates outside. Choose which one you like the best when asked what generation you belong to.

Generation	Wikipedia	18 Years
T (lost)	1883-1900	1889-1907
U (greatest)	1901-1927	1908-1926
V (silent)	1928-1945	1927-1945
W (boomer)	1946-1964	1946-1964
X	1965-1980	1965-1983
Y (millenial)	1981-1996	1984-2002
Z	1997-2012	2002-2020
alpha	2013-2025	2021-2039
beta	2026-2039	2040-2058
gamma	2040-???	2059-2077

(*) I say wikipedia here, but they are basically taking dates from various other sources and putting them together.. which should be seen as more on the statement of social commentators who aren’t good at math.

دانلود و راه‌اندازی Psiphon Conduit؛ ابزار قدرتمند ضد سانسور اینترنت

2026-02-03T12:44:47Z

Psiphon Conduit یک پروژه متن‌باز است که توسط شرکت Psiphon Inc توسعه یافته و بر پایه‌ی هسته‌ی تونلینگ Psiphon (psiphon-tunnel-core) ساخته شده است. این پروژه شامل یک کلاینت موبایل و CLI کراس‌پلتفرم برای ایجاد تونل و پروکسی است که با هدف دور‌ زدن سانسور اینترنت و افزایش دسترسی آزاد به اینترنت طراحی شده است. برخی […]

The post دانلود و راه‌اندازی Psiphon Conduit؛ ابزار قدرتمند ضد سانسور اینترنت first appeared on طرفداران فدورا.

آموزش و معرفی Snowflake Tor برای عبور از محدودیت‌های اینترنت

2026-01-31T11:04:45Z

پروژه Snowflake یک فناوری ضدسانسور از پروژه Tor است که به کاربران کمک می‌کند حتی در کشورها یا شبکه‌هایی که Tor مسدود شده، به اینترنت آزاد و شبکه Tor متصل شوند. چند ویژگی‌ مهم Snowflake: یک Pluggable Transport برای Tor است که ترافیک را طوری پنهان می‌کند که شناسایی و فیلتر آن سخت‌تر شود. ارتباطات […]

The post آموزش و معرفی Snowflake Tor برای عبور از محدودیت‌های اینترنت first appeared on طرفداران فدورا.

Stacking Protocols

2026-01-30T21:54:43Z

I find myself writing a program in C that is supposed to handle multiple protocols. At its entry point, the protocol is Platform Communication Channel (extended memory, type 3 and type 4). Embedded in that is an Management Component Transport Protocol (MCTP) message, and embedded in that is one of many different protocols.

I might want to swap out the PCC layer in the future for….something else. MCTP can come over many different protocols, so there is a good be that the tool will be more useful if it can assume that the protocol outside of the MCTP layer is something other than MCTP.

One problem I have is that the MCTP header does not have a length field. We do not not know how long the payload is; all it has is version, source, destination, and flags. Thus, if we want to pass a buffer of type MCTP header along, and we want the length, we need to pass it in a separate field. This goes both for incoming (how many bytes to read) and outgoing (how many bytes to write).

My initial thought on writing this layer is to have a request/response pair for each layer of the protocol. For PCC, I could just do this. For all the other internal ones, I would need to pass length in and length out for each handler. Length in will not change, but length out might, so this needs to be passed as pointer. This leads to functions that look like this:

void handle_mctp_control_message(struct mctp_hdr * mctp_req,  int req_len,  struct mctp_hdr * mctp_resp,  int *resp_len)
{

}

I will also need to be converting from outer protocols to inner protocols. So I will need code like this:

struct mctp_hdr * mctp_req  = (struct mctp_hdr *)pcc_req->buffer_start;
int mctp_req_len = pcc_req->length - sizeof(MCTP_SIGNATURE);
struct mctp_hdr * mctp_rsp  = (struct mctp_hdr *)pcc_rsp->buffer_start;

The outgoing length would be initialized to 0, and grow as each later of the protocol stack adds its own data. However, I am planning on pre-allocating the buffer, and just passing a pointer to the location where the protocol is supposed to write its data. IN order to confirm we don’t want to write past the end of the function, we will have to pass the overall buffer length in, maybe shortened by the amount we need to reserve for the outer headers.

If each protocol header had a smart pointer, I could pass those around instead. Something like:

struct pcc_header_p {
    struct pcc_header * header;
    int buffer_length;
}

struct mctp_header_p {
    struct mctp_header * header;
    int length;
    int buffer_length;
}

Then it would be fairly easy to write a function that, given an *pcc_header_p populates a struct mctp_header_p that points to it.

void pcc_2_mctp(struct pcc_header_p *pcc, struct mctp_header_p *mctp)
{
   mctp->header = (struct mctp_header *)pcc->header[sizeof struct pcc_header];
   mctp->len =  pcc->header.length - sizeof(MCTP_SIGNATURE);
   mctp->buffer_length = pcc->buffer->length - sizof(struct pcc_header);
     
}

This seems like it would benefit from a set of preprocesser Macros. I know that Qemu does something like this. But for a first pass, I think I can just code it up like this.

Building an AI assistant for computational modelling with NeuroML

2026-01-30T16:19:10Z

Brain models are hard to build

While experiments remain the primary method by which we neuroscientists gather information on the brain, we still rely on theory and models to combine experimental observations into unified theories. Models allow us to modify and record from all components, and they allow us to simulate various conditions---all of which is quite hard to do in experiments.

Researchers model the brain at multiple levels of detail depending on what it is they are looking to study. Biologically detailed models, where we include all the biological mechanisms that we know of---detailed neuronal morphologies and ionic conductances---are important for us to understand the mechanisms underlying emergent behaviours.

These detailed models are complex and difficult to work with. NeuroML, a standard and software ecosystem for computational modelling in Neuroscience, aims to help by making models easier to work with. The standard provides ready-to-use model components and models can be validated before they are simulated. NeuroML is also simulator independent, which allows researchers to create a model and run it using a supported simulation engine of choice.

In spite of NeuroML and other community developed tools, a bottleneck remains. In addition to the biology and biophysics, to build and run models, one also needs to know modelling/simulation and related software development practices. This is a lot, presents quite a steep learning curve and makes modelling less accessible to researchers.

LLM based assistants provide a possible solution

LLMs allow users to interact with complex systems using natural language by mapping user queries to relevant concepts and context. This makes it possible to use LLMs as an interface layer where researchers can continue to use their own terminology and domain-specific language, rather than first learning a new tool's vocabulary. They can ask general questions, interactively explore concepts through a chat interface, and slowly build up their knowledge.

We are currently leveraging LLMs in two ways.

RAG

The first way we are using LLMs is to make it easier for people to query information about NeuroML.

As a first implementation, we queried standard LLMs (ChatGPT/Gemini/Claude) for information. While this seemingly worked well and the responses sounded correct, given that LLMs have a tendency to hallucinate, there was no way to ensure that the generated responses were factually correct.

This is a well known issue with LLMs, and the current industry solution for building knowledge systems using LLMs with correctness in mind is the RAG system. In a RAG system, instead of the LLM answering a user query using its own trained data, the LLM is provided with curated data from an information store and asked to generate a response strictly based on it. This helps to limit the response to known correct data, and greatly improves the quality of the responses. RAGs can still generate errors, though, since their responses are only as good as the underlying sources and prompts used, but they perform better than off-the-shelf LLMs.

For NeuroML we use the following sources of verified information:

the documentation from https://docs.neuroml.org
NeuroML publications
other sources, such as well validated models from our Open Source Brain platform

I have spent the past couple of months creating a RAG for NeuroML. The code lives here on GitHub and a test deployment is here on HuggingFace. It works well, so we consider it stable and ready for use.

Here is a quick demo screen cast:

We haven't dedicated too many resources to the HuggingFace instance, though, as it's meant to be a demo only. If you do wish to use it extensively, a more robust way is to run it locally on your computer. If you have the hardware, you can use it completely offline by using locally installed models via Ollama (as I do on my Fedora Linux installation). If not, you can also use any of the standard models, either directly, or via other providers like HuggingFace.

The package can be installed using pip, and more instructions on installation and configuration is included in the package Readme. Please do use it and provide feedback on how we can improve it.

Implementation notes (for those interested)

The RAG system is implemented as a Python package using LangChain/LangGraph. The "LangGraph" for the system is shown below. We use the LLM to generate a search query for the retrieval step, and we also include an evaluator node that checks if the generated response is good enough---whether it uses the context, answers the query, and is complete. If not, we iterate to either get more data from the store, to regenerate a better response, or to generate a new query.

The RAG system exposes a REST API (using FastAPI) and can be used via any clients. A couple are provided---a command line interface and a Streamlit based web interface (shown in the demo video).

The RAG system is designed to be generic. Using configuration files, one can specify what domains the system is to answer questions about, and provide vector stores for each domain. So, you can also use it for your own, non-NeuroML, purposes.

Model generation and simulation

The second way in which we are looking to accelerate modelling using LLMs is by using them to help researchers build and simulate models.

Unfortunately, off-the-shelf LLMs don't do well when generating NeuroML code, even though they are consistently getting better at generating standard programming language code. In my testing, they tended to write "correct Python", but mixed up lots of different libraries with NeuroML APIs. This is likely because there isn't so much NeuroML Python code out there for LLMs to "learn" from during their training.

One option is for us to fine tune a model with NeuroML examples, but this is quite an undertaking. We currently don't have access to the infrastructure required to do this, and even if we did, we will still need to generate synthetic NeuroML examples for the fine-tuning. Finally, we would need to publish/host/deploy the model for the community to use.

An alternative, with function/tool calls becoming the norm in LLMs, is to set up a LLM based agentic code generation workflow.

Unlike a free-flowing general-purpose programming language like Python, NeuroML has a formally defined schema which models can be validated against. Each model component fits in at a particular place, and each parameter is clearly defined in terms of its units and significance. NeuroML provides multiple levels of validation that give the user specific, detailed feedback when a model component is found to be invalid. Further, the NeuroML libraries already include functions to validate models, read and write them, and to simulate them using different simulation engines.

These features lend themselves nicely to a workflow in which an LLM iteratively generates small NeuroML components, validates them, and refines them based on structured feedback. This is currently a work in progress in a separate package.

I plan to write a follow up post on this once I have a working prototype.

While being mindful of the hype around LLMs/AI, we do believe that these tools can accelerate science by removing/reducing some common accessibility barriers. They're certainly worth experimenting with, and I am hopeful that the modelling/simulation pipeline will help experimentalists that would like to integrate modelling in their work do so, completing the neuroscience research loop.

MIT DEDP MicroMasters online learner's blog post about cover-up linked to resignation of Swiss financial regulator

2026-01-30T14:00:00Z

MIT's DEDP MicroMasters is about Data, Economics and Development Policy. It was recently renamed, Data Economics & Design of Policy although the focus remains on developing countries, not rich countries like Switzerland. From time to time, I've seen people asking what a MicroMasters certificate is really worth. Ironically, that would be a great question for an MIT economist to answer.

In one of the modules, 14.750x Political Economy, Prof Ben Olken begins by asking the online learners to read papers Hit or Miss? The Effect of Assassinations on Institutions and War and Do leaders matter? National leadership and growth since world war II. The above average rate of assassinations for people with this job category is used to give us insights. There is nothing in the paper about leaders kidnapped by President Trump. Trump himself was nearly assassinated before his own return to office. The paper is available online.

In earlier blogs, we were able to prove the anonymous document on the FINMA web site relates to Parreaux, Thiébaud & Partners, Switzerland's "Law Firm X". From there, we were able to prove that FINMA, the regulatory authority knew about it for some years and the Geneva bar association also seemed to know for a long time.

In the world of abuse, it seems that priests knew some of their colleagues were paedophiles but they have a code of conduct, the Crimen Sollicitationis which prevented priests from warning the public about their own colleagues. The Swiss jurists from the bar association and officials from FINMA appeared to be operating from the same playbook. Those who knew about the scandal didn't tell the clients. They put the reputation of their profession and the privacy of rogue colleagues ahead of the interests of public safety and justice.

We then went the next step, showing that FINMA participated in a cover-up. Reports on this web site meticulously reverse-engineering their cover-up tactics.

From the moment the illegal legal insurance launched in 2018, Birgit Rutishauser had been head of FINMA's surveillance of the insurance industry.

Rutishauser is a graduate of ETH Zurich. Many of the attacks on my family revolve around the death of Adrian von Bidder-Senn on our wedding day. He was also an ETH Zurich graduate like Rutishauser. Just as the priests and the jurists have sought to maintain strict silence about the wrongdoing of colleagues, it seems the ETH Zurich alumni are maintaining silence. After all, Adrian von Bidder-Senn's wife obtained a PhD in cybersecurity from the same institution but looking at the last email she sent to the Debianists, we can see that Dr Diana von Bidder-Senn failed to realize the extent to which her husband was a victim of social engineering. Dr von Bidder-Senn is now the mayor of Basel.

When Rutishauser was appointed in 2018, FINMA published the following comments about her background:

The 46-year-old mathematician and actuary has managed the Risk Management section of the Insurance division since June 2016. Prior to joining FINMA, Birgit Rutishauser spent many years working in a variety of management roles within the insurance sector, most recently as Chief Underwriting & Risk Management Officer at Nationale Suisse and, as such, a member of the Group Executive Management Board. Birgit Rutishauser is a Swiss citizen.

Rutishauser had also worked for Zurich insurance, the same company where Urban Angehrn had worked prior to joining FINMA.

By comparison, I am a holder of the MIT MicroMasters diploma in Data, Economics & Development Policy (DEDP). Well, I also worked for a few banks too. UBS in Zurich provided a reference letter:

Competitors and cyberbullies have spent an enormous amount of effort trying to trick people to believe that I'm a mentally ill person who only pretends to be a developer:

Subject: Re: Open Letter to Debian election candidates about Debian vendettas
Date: Sun, 20 Mar 2022 21:00:02 +0900
From: Hideki Yamane 
To: Daniel Pocock 
CC: debian-devel@lists.debian.org

Hi,

 [ ... snip gaslighting ... ]

 Before talking, you should get counseling for a while since it seems that
 you have some cognitive troubles now. You'd be better to hear about your
 opinion and current your mind status from professional 3rd parties, not Debian.
 (If they say you're very healthy and good, then that's good. Don't you think so?)

 Without that, we cannot make a constructive conversation.
  As I said in my platform, "Be calm, stay cool, stay safe" - Hope you stay "cool"
 a bit with help from professionals, and you would to be able to a "contributor"
 to floss again.

 Life is short - to waste our time for fighting. Let's make more values for users.

-- 
Hideki Yamane

The rumours about mental illness were obviously falsified. Nonetheless, it is important to remember they started falsifying these things at a time when I lost two family members. Is it an example of cybertorture or is it simply an example of how rude these people are after spending too much time in social control media?

Is it possible that a mentally ill fake developer with a MicroMasters was able to write a blog post that brought down the Deputy CEO of the Swiss financial market regulator?

Rather than finding me to be insane, Switzerland granted me citizenship in the Canton of Vaud in November 2023:

Early 2024, I came across a photo of Switzerland's attorney general visiting his peers at the Parquet National Financier in Paris, France to talk about cross-border crime. The illegal legal insurance was an example of cross-border crime because they recruited people from France to work for them and they promoted the insurance to French residents.

The meeting and photo has since been removed from the PNF web site but it can be found in the Wayback Machine.

Rencontre avec M. Stefan Blättler, procureur général de la Confédération helvétique

Le 13 février, les quatre chefs de juridiction du tribunal judiciaire de Paris ont reçu M. Stefan Blättler, procureur général de la Confédération helvétique.

Ces échanges stratégiques ont permis d’aborder les questions d’entraide pénale internationale et de lutte contre la criminalité organisée, le terrorisme, les crimes contre l'humanité et la délinquance financière.

Ils ont été l’occasion de rappeler la détermination commune des autorités judiciaires françaises et helvétiques à lutter contre les formes complexes de criminalité et de réaffirmer le caractère essentiel d'un dialogue soutenu et cordial entre pays voisins, partageant une frontière et une relation longue.

I wrote to the PNF and sent them some of the documents. This was their reply:

The PNF, like me, suspects that a crime was committed under French law but for technical reasons, they require a French citizen to bring a formal complaint to a local prosecutor first. The law in France would see me as a witness to the crime.

Later in 2024, I acquired the domain name www.michaelmcgrath.ie, that is the former domain name of Michael McGrath, the EU Commissioner for Democracy, Justice, the Rule of Law and Consumer Protection.

I used the EU Justice Commissioner's former domain name to publish information about the cross-border crime from the Swiss jurists.

The blogs published between January and March 2025 continue to build the case that FINMA not only knew about the illegal legal insurance but they also had a role in the cover-up.

The blog published on 8 March 2025, International Women's Day, considered the case of a French woman who was tricked to quit the job she had for seven years and come to work for the Swiss jurists in Geneva selling illegal legal insurance.

On 27 March 2025, I was discussing the case with an expert in Paris at the same time the BBC was interviewing former Archbishop of Canterbury Justin Welby about the cover-up of their jurist John Smyth QC. In the meeting in Paris, I was asked if there was a link between the paedophiles and the jurists. The BBC's report only appeared after the meeting and it prompted me to write a new blog post about the culture of cover-ups.

When famed whistleblower Trevor Kitchen exposed the Forex scandal, he didn't go as far as comparing anybody to a paedophile, nonetheless, they had him arrested in Portugal and tried to have him extradited back to Switzerland for criminal speech.

Finance chief who exposed currency scandal fights Swiss extradition bid for criminal defamation

...

Mr Kitchen, who worked as a financial controller for companies including Shell, Castrol and Black and Decker, lost his 700,000 franc pension due to the rigged currency fluctuations.

...

“Because I had specialised in finance all my life understanding policies and procedures in companies, I went and reported this to all the regulators,” he said.

Toby Cadman, head and co-founder of Guernica 37 International Justice Chambers, who is working pro bono for Mr Kitchen, said: “If he was in the UK, this would not even get past the issuing of a warrant, let alone the extradition process.

...

He was arrested by Portuguese police on January 19 and taken for questioning. “It was all maximum security stuff. Six policemen were around me and took me into a small room and told me to take all my clothes off. They threw me in prison for 48 hours all because of the words I used,” he said.

The Swiss embassy has been contacted for comment.

Yet Swiss authorities have made no such move against me. Without admitting or denying any of the allegations on this web site, Birgit Rutishauser's resignation was announced on 1 April 2025, barely two days after the strongest blog post about the complicity of the authorities in the cover-up.

First of all, FINMA published a notice about Rutishauser's resignation.

The same day, FINMA published a longer notice about their restructuring. The notice included a paragraph emphasizing that Birgit Rutishauser's departure was not part of the restructuring. In other words, there is some other specific reason she has chosen to tender her resignation and they are not going to give us any more detail about it.

FINMA adapts its organisation to meet future challenges

...

Independently of the introduction of the new organisational structure, Birgit Rutishauser, member of the Executive Board, Deputy CEO and Head of the Insurance division, has decided to leave FINMA. Vera Carspecken will assume the leadership of the Insurance division on an interim basis from 1 May 2025 (see separate press release of 1 April 2025).

Somebody with so many years of service would normally be entitled to three months of notice period. The abrupt departure of Martin Senn from Zurich insurance was referred to as a factor in his suicide. However, it is also possible that Rutishauser was placed on an extended period of garden leave to prevent her from immediately being able to use confidential information she has acquired in the course of her duties.

Rutishauser's LinkedIn profile tells us she is now at StenFo, that is the fund for a Swiss nuclear dump.

All the waste from five decades of Swiss nuclear programs is currently stored in a temporary facility at Zwilag. If they ever try to build a permanent underground storage facility there will be a referendum to stop it. It just sits in the temporary facility for the time being:

President Putin frequently reminds the world about nuclear punishments. What we see here is the power of nuclear teamwork.

Fifty percent of Swiss nuclear fuel comes from Russia, twenty five percent from Canada and twenty five percent was mined in Australia, where I was born.

After leaving her post as second in command at FINMA, head of the insurance division, Rutishauser enrolled at IMD, a well known Swiss business school, where she was awarded a certificate in blockchain.

While there are some companies doing completely legitimate projects with blockchain, it is ironic that the scammers who had been allowed to operate for years under FINMA have also pursued a new career path promoting cryptocurrency "investment" to victims in France.

I started the Software Freedom Institute in May 2021 in the middle of the pandemic. Many people were not working at all. I could have chosen not to work and claimed furlough payments from the government. Instead, by working, I was paying tax and contributing money to the social security system to support the rest of society.

I purchased and paid for various insurance services to protect the business. Invoices for Parreaux, Thiebaud & Partners legal services. When IBM Red Hat attacked my business, the Swiss legal protection did nothing to help. The ADR Forum legal panel eventually gave a ruling against IBM Red Hat, declaring that I was a victim of harassment from a much larger company/competitor.

Likewise, after the conviction of Cardinal George Pell, rogue Debianists attacked my family with rumours about abuse and they attacked us again after I founded the Software Freedom Institute. The Swiss jurists did nothing to help. I went to Italy by myself to speak to the Carabinieri and after years of inquiries, the Cardinal died four hours after I filed a report about exploitation. Despite Switzerland's reputation for privacy, the Swiss jurists provided no help whatsoever to protect my family and I from these intrusions.

Justin Welby had to resign as Archbishop of Canterbury due to his failure to take action against the paedophile jurist John Smyth QC. On 30 March 2025, we published the comparison of Church of England, Crimen Sollicitationis and the cover-up at FINMA and barely two days later, Rutishauser's resignation was announced.

As it is Switzerland, nobody is holding their breath waiting for authorities to confirm the Deputy CEO of FINMA resigned because of the cover-up disclosed in the JuristGate reports. However, we can now try to answer the question we started with, what is the real value of the DEDP MicroMasters from MIT? Priceless.

Pourquoi je suis resté sur n8n ?

2026-01-30T11:38:00Z

Ça fait maintenant un bon moment que je me pose la question : est-ce que je dois quitter n8n pour une autre solution d’automatisation ? n8n est un excellent outil, je l’utilise depuis longtemps, mais au fil des versions une tendance devient claire : de plus en plus de fonctionnalités sont réservées aux offres Enterprise […]

Cet article Pourquoi je suis resté sur n8n ? est apparu en premier sur Guillaume Kulakowski's blog.

🎲 PHP version 8.4.18RC1 and 8.5.3RC1

2026-01-30T06:16:00Z

RPMs of PHP version 8.5.3RC1 are available

as base packages in the remi-modular-test for Fedora 41-43 and Enterprise Linux ≥ 8
as SCL in remi-test repository

RPMs of PHP version 8.4.18RC1 are available

as base packages in the remi-modular-test for Fedora 41-43 and Enterprise Linux ≥ 8
as SCL in remi-test repository

ℹ️ The packages are available for x86_64 and aarch64.

ℹ️ PHP version 8.3 is now in security mode only, so no more RC will be released.

ℹ️ Installation: follow the wizard instructions.

ℹ️ Announcements:

Parallel installation of version 8.5 as Software Collection:

yum --enablerepo=remi-test install php85

Parallel installation of version 8.4 as Software Collection:

yum --enablerepo=remi-test install php84

Update of system version 8.5:

dnf module switch-to php:remi-8.5
dnf --enablerepo=remi-modular-test update php\*

Update of system version 8.4:

dnf module switch-to php:remi-8.4
dnf --enablerepo=remi-modular-test update php\*

ℹ️ Notice:

EL-10 packages are built using RHEL-10.1 and EPEL-10.1
EL-9 packages are built using RHEL-9.7 and EPEL-9
EL-8 packages are built using RHEL-8.10 and EPEL-8
oci8 extension uses the RPM of the Oracle Instant Client version 23.9 on x86_64 and aarch64
intl extension uses libicu 74.2
RC version is usually the same as the final version (no change accepted after RC, exception for security fix).
versions 8.4.18 and 8.5.3 are planed for February 12th, in 2 weeks.

Software Collections (php84, php85)

Base packages (php)

rpminspect-2.1 released

2026-01-29T19:19:00Z

rpminspect 2.1 is now available. The last release was on September 5, 2024. That was 511 days ago, so a bit longer than I really wanted to make it. Similar to the previous release, this release of rpminspect marks the first post on my blog for 2026. My plan for 2026 is to make more frequent releases even if they just contain one or two fixes. Development and maintenance of rpminspect comes in waves, so I think that will work out better while getting fixes out to people.

Since it has been 511 days since the last release, I will just paste the changes below. I am having a hard time remembering everything that went in to it, but that’s why we have logs!

Work on 2.1 has begun. Please file issues and feature requests on the GitHub project page: https://github.com/rpminspect/rpminspect. The changes below are in the 2.1 release:

General release and build process changes:

Remove duplicate AND from License tag
Drop utils/find-ninja.sh and update GNUmakefile and README.md
Remove explicit requires on libtoml
Fix FTBFS errors in the spec file for home.
Update the Copr srpm target to sub %autorelease correctly
Change the Release value used for Copr builds to 1.1….
Add ‘Recommends: system-rpm-config’ to the spec file template

Config file or data/ file changes:

Drop ‘id’ from the example license JSON database

Changes to the GitHub Actions CI scripts and files:

On Fedora rawhide, remove duplicate __debug_package in macros
Fix broken Debian Testing and Alpine Linux CI jobs
On Fedora stable, remove duplicate __debug_package in macros
Remove obsolete Fedora CI job files
Uninstall openssl-fips-provider-so in the RHEL UBI9 test environment
On Fedora systems, make sure /var/lib/clamav is 0755
Add clamav-freshclam to reqs.txt for Fedora
Do not run freshcleam on Fedora CI jobs
Workaround Gentoo bug with circular dependencies
Install ‘gawk’ on fedora and fedora-rawhide CI jobs
Default Fedora environment lacks awk now, so install it.
Fixes for some failing CI jobs
Fix how ‘rc’ compiles on Arch Linux
On OpenSUSE Leap builds, explicitly install which before git
Drop the modification of /usr/bin/xmlrpc-c-config on Debian testing
Fix ‘rc’ build and install Mageia Linux job
Recognize newer operating systems in determine-os.sh
Set /var/lib/clamav ownership on CentOS 10 CI job
Remove Oracle Linux jobs
Fix Debian GHA failures
Fix the Debian GitHub Action jobs
Minor fixes for the OpenSUSE and Ubuntu CI jobs on GitHub
Ignore changes in the .github/ and osdeps/ subdirectories
Do ‘emerge -uDN world’ in the Gentoo CI job before running
Try to reduce the amount of stuff built on Gentoo
Increase the timeout for Ubuntu GHA jobs
Add additional packages to Debian jobs and install annocheck
Upgrade Fedora rawhide test environments
Install libatomic.i686 on Fedora stable and rawhide jobs
Install lib64xxhash-devel on Mageia jobs
Build cc with CFLAGS=-std=gnu99 on Alpine Linux
Fix up the Gentoo CI jobs so they don’t fail

rpminspect(1) changes or improvements related to it:

Disable the ‘filesmatch’ inspection for now

Documentation changes:

Update examples in the rpminspect(1) man page

General bug fix in the library or frontend program:

Strip UNPACK_BASE directory in unicode inspection
Subpackage changes can cause some peer rpmdeps to be empty
Support PatchN specifications with URLs to the upstream patch
Make sure unapproved licenses are reported as such.
Refactor how remedy strings are handled in librpminspect
Correctly handle SPDX special keywords: AND, OR, WITH
Match 32-bit x86 RPMs when performing inspections
Add extra guards for hardlink handling in extract_rpm()
Prevent duplicate SRPMs in file lists when comparing tasks
Check if archive exists before closing it
Properly handle hard links on RPM payload extraction
Continued fixes for the new libxml2 API
Correct rebase build comparison in some inspections
Download failure does not exit with a error code
Memory leak fixes in strtrim()
Allow %{load} macros to work when calling rpmSpecParse()
Define _specdir in read_spec() for spec files that self-reference
Memory leak fixes in read_spec()
Retain empty lines when using strsplit()
Ignore the tokens within a %verify() expression in %files
Do not write untokenized %doc and %license lines to filtered spec
Skip debuginfo and debugsource packages in filesmatch
s/file->rpm_header/peer->after_hdr/g in filesmatch
Ignore .build-id subdirectories in the filesmatch inspection
Handle %dir and %files modifiers in %doc & %license lines
Invoke annocheck correctly so we do not get incorrect reporting
Fix some curl_easy_setop() calls that give warnings on FreeBSD
Correctly construct entry paths in extract_rpm()
Correct some formerly gcc warnings that are now errors

librpminspect feature or significant change:

Disable file and scan size limits in clamav
Honor locally defined ignore list for the ‘unicode’ inspection
Favor POSIX basename(3) throughout librpminspect
Support SPDX 3.0 expression specification
Drop FNM_PATHNAME for path matching from rule files
When extracting payloads, if archive_read_extract() fails, error out
Change ENODATA to EIO in libarchive read loop
Expand the remedy strings for the ‘patches’ inspection
Handle newer versions of libxml2 and deprecated struct members
Minor memory management cleanups in shellsyntax inspection
Make free_string_hash() part of the public API
Define PATH_SEP and rename joinpath() to joindelim()
Add read_spec() function and use librpm for macro expansion
Introduce the ‘filesmatch’ inspection
Add filtering ability to the read_spec() function
Minor cleanups for the filesmatch inspection
Refactor how %doc and %license lines are processed in filesmatch
Hook up the actual filesmatch inspection
Remove the abandoned ‘filesmatch’ inspection.
Ignore expected empty RPM payloads in ‘lostpayload’ (#1518)
In runcmd.c, return NOT FOUND for not found commands
Account for RPM payloads carrying paths not in the RPM metadata
Handle additional annocheck profile name matching

Test suite commits:

Work around a limitation on Arch Linux in test_debuginfo
Add a testcase for unicode ignored files
Adjust test_kmod to account for EXTRA_CFLAGS removal in 6.15
Increase test case timeouts to 5000 from 1500
Add —skip-implicit-values to annocheck test jobs
In test_elf.py add -Wl,-z,notext for TEXTREL and PIC tests
When writing to test rpminspect.yaml file, do not wrap lines

See https://github.com/rpminspect/rpminspect/releases/tag/v2.1 for more information.

Where to get this new release?

Fedora (42, 43, and rawhide), EPEL 8, EPEL 9, and EPEL 10 users can get new builds from the testing updates collection. If you install from the testing update, please consider a thumbs up in Bodhi. Without that it takes a minumum of two weeks for it to appear in the stable repo.

Copr builds continue to be available in my Copr collection. The Copr repos are updated each time a pull request is merged and testing passes, so it contains the latest build of what will become the next stable release. Often times I ask reporters to try a Copr build to ensure a fix is correct. I do not expect all users to rely on the Copr builds, but do be aware of them if you report a bug or feature requests for rpminspect.

Lastly, I will be moving the upstream location of this project likely to codeberg.org and off of github.com as GitHub continues to march towards AI in everything. I will update the project page on GitHub when that happens and the repo there will likely go read-only before I eventually archive it or delete it entirely.

Cockpit 355

2026-01-29T00:00:00Z

Cockpit is the modern Linux admin interface. We release regularly.

Here are the release notes from Cockpit 355:

Add systemd/polkit-based fallback for administrative privileges if sudo is not available

There are systems like Ubuntu 25.10 which don’t have sudo, or only the incompatible sudo-rs. Cockpit previously failed to gain administrator rights then. Cockpit 355 now both detects and ignores an incompatible sudo, as well as introduces a fallback authentication method: It starts the root bridge through systemd’s StartTransientUnit() API. This is inspired by systemd’s run0, and uses the exact same underlying mechanism. This is guarded by polkit, so you need to authorize with your user password similar to sudo.

ws: Remove obsolete pam_cockpit_cert module

Cockpit version 209 (Dec 2019) introduced a pam_cockpit_cert PAM module in /etc/pam.d/cockpit. Cockpit 248 (Jul 2021) made this module unnecessary, and replaced the module with a stub that did not do anything, plus a warning during package upgrade.

This version finally removes it. So if you have locally modified /etc/pam.d/cockpit and still have that module in your configuration, you need to manually remove it.

Try it out

Cockpit 355 is available now:

Kiwi TCMS 15.3

2026-01-28T20:44:00Z

Dear testers, we're happy to announce Kiwi TCMS version 15.3!

IMPORTANT:

This is a minor version release which includes updates and improvements, database migrations and many new API methods.

You can explore everything at https://public.tenant.kiwitcms.org!

---

Public container image (x86_64):
pub.kiwitcms.eu/kiwitcms/kiwi   latest  bad66695a117   732MB

IMPORTANT: version tagged and multi-arch container images are available only to subscribers!

Changes since Kiwi TCMS 15.2

Improvements

Update Django from 5.2.9 to 5.2.10
Update django-simple-captcha from 0.6.2 to 0.6.3
Update django-simple-history from 3.10.1 to 3.11.0
Update from 3.10 to 3.10.1
Update psycopg[binary] from 3.3.1 to 3.3.2
Update python-gitlab from 7.0.0 to 8.0.0
Update tzdata from 2025.2 to 2025.3
Update node_modules/pdfmake from 0.2.20 to 0.3.3
Models Classification, Product, Priority, Component and Version can now be translated in Admin pages (@nichicom-yasutake). Fixes Issue #4193
Display categories ordered by name. Fixes Issue #4166
Stricter checking for attributes in uploaded files
Strip newline characters from email subjects to avoid crashes

Database

Add migration management.0012_alter_classification_options_alter_component_options_and_more
Add migration testcases.0023_alter_category_ordering

API

Add Bug.create() API method
Add Bug.filter_canonical() API method
Add Bug.get_comments() API method
Add Bug.add_comment() API method
Add Bug.add_attachment() API method
Add Bug.list_attachments() API method
Add Bug.add_execution() API method
Add Group.filter() API method
Add Group.permissions() API method
Add Group.users() API method
Add TestExecution.add_attachment() API method
Add TestExecution.list_attachments() API method
Add TestExecution.create() API method
Add TestExecution.add_property() API method
Add TestRun.list_attachments() API method
Add TestRun.add_property() API method
Add TestRun.get_cc() API method
Allow TestCase.create() API method to override TestCase.create_date
Allow TestPlan.create() API method to override TestPlan.create_date
Allow super-user to override comment author/submission date for TestCase.add_comment() API method
Allow super-user to override comment author/submission date for TestExecution.add_comment() API method
For BugTracker.filter() API method order results by id field
For Category.filter() API method order results by id field
For Classification.filter() API method order results by id field
For Component.filter() API method order results by id field
For Group.filter() API method order results by id field
For PlanType.filter() API method order results by id field
For Priority.filter() API method order results by id field
For Tag.filter() API method order results by id field
For Template.filter() API method order results by id field
For TestCase.filter() API method order results by id field
For TestCaseStatus.filter() API method order results by id field
For TestExecution.filter() API method order results by id field
For TestExecutionStatus.filter() API method order results by id field
For TestRun.filter() API method order results by id field
For User.filter() API method order results by id field
TestCase.comments() API method changes the following fields in its response:
- site -> site_id
- user -> user_id

Refactoring and testing

Update actions/upload-artifact from 5 to 6
Update black from 25.11.0 to 25.12.0
Update locust from 2.42.6 to 2.43.1
Update node_modules/brace-expansion from 1.1.11 to 1.1.12
Update node_modules/webpack from 5.103.0 to 5.104.1
Update URL to website article
Remove redundant code snippet in TestRun.create() API method
Add more assertions for TestRun.create() API test scenario

Changes since Kiwi TCMS Enterprise v15.2-mt

Based on Kiwi TCMS v15.3
Update certbot from 5.2.1 to 5.2.2
Update django-ses from 4.4.0 to 4.6.0
Update dj-database-url from 3.0.1 to 3.1.0
Update kiwitcms-tenants from 4.3.0 to 4.4.1
Update sentry-sdk[django] from 2.47.0 to 2.50.0
Update social-auth-app-django from 5.6.0 to 5.7.0
Update workaround for .map files b/c of newer pdfmake

Private container images

hub.kiwitcms.eu/kiwitcms/version          15.3 (aarch64)          3d15693bb229    28 Jan 2026     749MB
hub.kiwitcms.eu/kiwitcms/version          15.3 (x86_64)           0718496358e5    28 Jan 2026     732MB
hub.kiwitcms.eu/kiwitcms/enterprise       15.3-mt (aarch64)       235d645ffdfa    28 Jan 2026     1.02GB
hub.kiwitcms.eu/kiwitcms/enterprise       15.3-mt (x86_64)        b580a3c04ef3    28 Jan 2026     997MB

IMPORTANT: version tagged, multi-arch and Enterprise container images are available only to subscribers!

How to upgrade

Follow the Upgrading instructions from our documentation.

Happy testing!

---

If you like what we're doing and how Kiwi TCMS supports various communities please help us grow!

Give â� on GitHub;
Join our newsletter and follow all news;
Become a subscriber and help us sustain development

Fedora RISC-V 64-bit now available in Copr

2026-01-28T00:00:00Z

We are happy to announce that Fedora RISC-V 64-bit (riscv64) build targets are now available in Copr! The following chroots have been added to mock-core-configs and enabled in Copr:

fedora-42-riscv64
fedora-43-riscv64

Important notes

We currently do not have native RISC-V hardware in our infrastructure. All riscv64 builds are performed using QEMU emulation on x86_64 machines. Because of this:

Expect longer build times compared to native architectures
Only userspace calls are possible. If your package needs to make a kernelspace call during the build time or the check phase, then your build will fail. Only a few packages are limited by this.

If you encounter any issues, please let us know!

Installing Windows games on Linux - Arknights: Endfield

2026-01-27T00:31:19Z

You can easily run any Windows game on Linux nowadays without too much work. Let’s look at an example - Arknights: Endfield - using only Steam + Proton.

Proton

I usually run things through GE-Proton, however this is the rare case when that has disappointed me. While the game installed fine for me and the launcher was semi functionaly, it wouldn’t actually start the game.

So this time around, it’s gonna be dwproton - installing it is pretty simple, grab ProtonPlus from flathub using your software center of convenience. Then simply run it and find dwproton in the list, hit install and you’re done.

Make sure to restart Steam if you had it running.

Steam

If you’re on SteamDeck (or other Steam gamemode session) make sure to switch to Desktop first.
Download the windows installer: https://endfield.gryphline.com/en-us - bottom right corner of the page and I recommend you to immediately mute the tab in your browser ;)
Add it to Steam as a non-steam game (Menu -> Games -> Add a non-steam game to my Library)
Right click the added installer and go to its Properties -> Compatibility -> Force the use of dwproton.
Play -> it should now run the installer, follow the steps to install it, all the defaults are fine. If the launcher automatically starts it at the end, close it.

We’ve now installed the launcher but we can’t exactly execute it directly just yet. Let’s sort that out:

Right click the installer in steam again, this time we’re gonna change the Target field to "/home/deck/.steam/steam/steamapps/compatdata//pfx/drive_c/Program Files/GRYPHLINK/Launcher.exe"
- You may have noticed the there - this is a generated ID for your non steam game and it’s probably different from mine. To find out what it is, simply navigate to that location and take a look at which folder was modified last. That’s the correct one.
- If you’re not on Steam Deck the /home/deck is gonna be different - that’s your home folder and it will be your username.
- And don’t forget the quotes!
You’re actually done now. It will work with gamescope, mangohud, or any other shenanigans that you’re used to. Probably. Use the Launch options same as with any other Steam game.
Use the SteamGridDB Decky Plugin to change the artwork on Steam, or SDGBoop (flatpak) in combination with steamgriddb.com
Cleanup - you can now delete the installer.

Enjoy!

Find me in the Fedora Linux Discord if you have any issues or questions.

Endfield on Fedora Linux

Introducing Amutable

2026-01-26T23:00:00Z

Today, we announce Amutable, our âœ¨ new âœ¨ company. We â€“ @blixtra@hachyderm.io, @brauner@mastodon.social, @davidstrauss@mastodon.social, @rodrigo_rata@mastodon.social, @michaelvogt@mastodon.social, @pothos@fosstodon.org, @zbyszek@fosstodon.org, @daandemeyer@mastodon.social @cyphar@mastodon.social, @jrocha@floss.social and yours truly â€“ are building the ðŸš€ next generation of Linux systems, with integrity, determinism, and verification â€“ every step of the way.

For more information see â†’ https://amutable.com/blog/introducing-amutable

replyfast a python module for signal

2026-01-26T12:16:49Z

replyfast is a Python module to receive and send messages on Signal.

You can install it via

python3 -m pip install replyfast

uv pip install replyfast

I have to add Windows builds to CI though.

I have a script to help you to register as a device, and then you can send and receive messages.

I have a demo bot which shows both sending and rreceiving messages, and also how to schedule work following the crontab syntaxt.

    scheduler.register(
        "*/5 * * * *",
        send_disk_usage,
        args=(client,),
        name="disk-usage",
    )

This is all possible due to the presage library written in Rust.

We Make Fedora

Urgent : vérification du bien-être, appelez maintenant pour l’équipe féminine de football iranienne

Urgent: welfare check, call now Iran womenâ€™s football team

misc fedora bits first week of march 2026

GPU machine

Fedora 44 Beta is GO for release next tuesday

Secure boot signing

Just missed it

comments? additions? reactions?

Ideal Laptop

Laptop ideal

Bon Ap : Lieu de naissance des ANZAC, ancêtre de la guerre hybride, lié à un restaurant français

Bon Ap: ANZAC birthplace, Grandfather of hybrid warfare linked to French restaurant

GNOME Foundation Update, 2026-03-06

Audit time

GUADEC 2026

LAS 2026

Infrastructure

Final Reminder: Flock 2026 Financial Assistance Applications Close Sunday, March 8th

Note for Flock 2026 speakers

Apply for Flock 2026 financial assistance

Community Update – Week 10 2026

Fedora Infrastructure

CentOS Infra including CentOS CI

Release Engineering

RISC-V

AI

QE

Forgejo

UX

Logsqueak: Rescuing Insights from the Logseq Journal

How Logsqueak Works

1: Extraction (Signal vs. Noise)

2: Refinement (Making it Evergreen)

3: Integration (Filing it Away)

Try it out!

Taming the Knowledge Graph

Uniqlo T-Shirt Bash Script Easter Egg

Mauvaise foi : Hugo Roy savait que la FSFE se faisait passer pour la FSF devant le tribunal français, ses collègues ont été trompés.

Fedora Documentation translations not available from March 4th, 2026

Honey, I Shrunk the Model (Maybe): blk-archive vs AI Data

Because “It Should Work” Isn’t Data

LImiting What an Agent can do

misc fedora bits last week of feb 2026

Lots of small day to day items

Secure boot signing work

comments? additions? reactions?

Installing TeX Live 2025 on Fedora: A Minimalist User-Space Guide

Pre-Installation: System Dependencies

Create the Directory Structure

Download and Launch the Installer

Installer Configuration (Text Menu)

Post-Installation: Environment Setup

Finalizing the Formats

Verification and Troubleshooting

Check the Binary Path

Compile a Test Document

Summary of Locations

Patchutils 0.4.5 released

Compatibility Fix: Git Extended Diffs

Status Indictors for grepdiff

Community Update – Week 9, 2026

Fedora Infrastructure

CentOS Infra including CentOS CI

Release Engineering

RISC-V

AI

QE

Forgejo

EPEL

UX

Mise en place d’Anubis sur l’instance Scaleway de Fedora-Fr

🎲 PHP version 8.4.19RC1 and 8.5.4RC1

Veritasium

Cockpit 357

Show browser context menu when holding Shift

Machines: Support for kernel boot arguments when creating VMs

Podman: Show description, version and documentation image labels

Try it out

📝 Install PHP 8.5 on Fedora, RHEL, CentOS Stream, Alma, Rocky or other clone

Show browser context menu when holding `Shift`