TGM

Fabulous Brussels

2026-02-24T00:00:00+00:00

Go

2026-02-24T00:00:00+00:00

go</span> fix</span></span>
</span>
go</span> get</span> -u</span></span>
</span>
go</span> mod tidy</span></span></code></pre>


Thinkpad
2026-02-05T00:00:00+00:00

rthinkpad Wiki Your Guide to ThinkPads</a></li>
ThinkWiki</a></li>
</ul>


Cameron n’est qu’une incarnation de plus de la prédation du capitalisme, cynique au point de nous faire passer un monde synthétique comme un geste politique
2026-01-01T00:00:00+00:00

J’écrivais récemment pour un texte que je prépare : l’expérience se partage, la consommation se satisfait. Et je trouve dans les commentaires des panégyristes de Cameron beaucoup de (auto)satisfaction.</p>
La question n’est pas de savoir si Télérama a raison de critiquer les carences et l’insuffisance scénaristique de ce film, la question est de savoir ce qui est véhiculé par Cameron, et ce depuis des décennies. Au hasard : une certaine fascination pour la guerre et la violence. Paul Virilio pointait ça déjà du doigt dans les années 90, comme lorsqu'il soulevait qu'Aliens n'est qu'un genre de répétition générale de la Première Guerre du Golfe (sans parler de la manipulation qui va avec, comme par exemple le personnage de Newt qui n'est qu'un pis-aller scénaristique, une justification émotionnelle au déferlement de violence qui s'en suit.) C'est ce qu'on appelle communément le « soft power » états-unien. (Le regretté Mark Fisher fait aussi une excellente lecture du T-1000 en tant que nouvelle forme, après celle du T-800, que prendrait la prédation capitaliste et les formes du travail, mais c'est une autre histoire.)</p>
Au lieu d’être employés à la concordance imaginaire du collectif humain au milieu naturel, les récits sont sempiternellement façonnés pour l’alliance des relations entre collectifs humains : individus, générations, civilisation. L’être humain en tant qu’individu, les humains en tant que sociétés, en tant qu’espèce sont devenus, dans l’Occident moderne, le héros incessant de tous les récits. C’est un culte de la personnalité dont il nous faut sortir désormais. Jean-Christophe Cavallin nous rappelle que le rôle de la fiction devrait être d’assurer l’entretien du régime symbolique, de faire en sorte qu’il reste en prise avec la réalité sans mettre en péril notre vie psychique. Si la fiction répare quelque chose, ce n’est pas le monde abîmé, mais la machine symbolique qui permet au monde et à la pensée de co-évoluer et de co-produire.</p>
Je fais partie de celles et ceux qui persistent et signent en disant que, par-delà l'hypocrisie du concept (profondément colonialiste) d'Avatar, il y a un discours mégalomane qui se veut démiurgique, voire messianique (cf. son livre Histoire de la SF). Je le redis : Cameron, dans sa course au progrès à tout prix — science sans conscience —, a une foi aveugle et irresponsable en la technologie. Faire un film qui se veut « écologiste » (ou « progressiste » ou « humaniste », que les laudateurs choisissent l'adjectif qu'ils veulent, je garde en tous cas les guillemets ironiques pour chacun d’entre eux) en le séparant aussi intrinsèquement du vivant n’est plus un paradoxe, c’est le symptôme d’une schize qui frise la pathologie – celle du désir de se passer des humains (ou tout du moins de les réifier) au profit d’images idéalisées (revoir à ce sujet l’entretien entre Godard et Daney en 1988). C'est sur ce point que le fond et la forme de Cameron se marient et engendrent des monstres. Cameron n’est qu’une incarnation de plus de la prédation du capitalisme, cynique au point de nous faire passer un monde synthétique comme un geste politique.</p>
La difficulté pour ses fans est de réussir à passer outre l'éventuelle jouissance ressentie dans un de ses films (jouissance que je ne nie pas, mais on ne peut pas s’en contenter…) pour tenter de voir par-delà, chercher ce qui est « réellement » dit.</p>
Et toutes celles et ceux qui nous diront qu'il est un « grand storyteller », un « superbe technicien » ou un « grand réalisateur » n'y changeront rien à mes yeux : il est proche des gouvernements US, d'Elon Musk et autres ordures de la Silicon Valley. Il est donc du côté de nos ennemis. (On ne peut se plaindre que le cordon sanitaire avec l'extrême droite ait été coupé sans faire le ménage jusque dans nos imaginaires et nos goûts.)</p>
La question n'est donc pas « scénario ou pas scénario ». S'il il y a une ligne de démarcation entre deux camps, elle est peut-être plutôt là : celle que Paul B. Preciado définissait très bien en 2014 dans l'Entretien infini : « Soit on est du côté de l'esthétique du capital, soit on est du côté de la transformation sociale. Et au milieu, il n'y a rien. »</p>
Et puisqu'on parle de transformation sociale, mentionnons Ursula K. Le Guin grâce à un article de Mark Fisher sur son blog k-punk : « Avatar is Le Guin-lite, a degraded version of the scenario that Le Guin developed in novels such as The Word For World Is Forest, The Dispossessed and City Of Illusions, but stripped of all Le Guin's ambivalence and intelligence »</p>
</blockquote>
— Vincent Capes de la chaîne Zoe Anima</a></p>
via Le Fossoyeur de Film @YT</a></p>


Nix
2025-12-29T00:00:00+00:00
nix is a language/tool used to handle reproducible declarative packages and system configurations.</p>
It can be used as a package manager nearly</em> everywhere, as an OS, both for server and user machines, and as a dev env and services hosting deployements.</p>
Useful Links</h2>
Package Search & configuration options</h3>
Packages</strong></p>

Official nixpkgs search</a></li>
</ul>
Options</strong></p>

MyNixOS</a></li>
Home Manager OPTION SEARCH</a>, by the ExtraNix Project</a></li>
</ul>
Passing Mentions</strong></p>

Flakehub</a>, by Determinate Systems</a></li>
</ul>
Wikis</h3>

Official NixOS wiki</a></li>
nixos.wiki</a>, unofficial user wiki</li>
nix.dev</a></li>
</ul>
Manuals</strong></p>

nixpkgs</code></a></li>
NixOS</a></li>
nix</code> at nix.dev</a></li>
</ul>
Learning Ressources</h3>

Nix Pills</a></li>
nix.dev's NixOS</a> tutorial</li>
XeIaso's blog</a></li>
NixOS</a> Series</li>
Nix Flakes</a> Series</li>
NixOS & Flakes Book</a></li>
Zero to Nix</a>, by Determinate Systems (again)</li>
A tour of Nix</a></li>
Nixology</a> playlist</li>
NixOS for the Impatient</a></li>
Everything I Know</a> → key</a></li>
NixOS in Production</a></li>
</ul>
Passing Mention</strong></p>

NixOps</a></li>
</ul>

Sources: Reddit</a></p>

Installation</h2>
System-wide/userspace-wide</h3>
There are two ways to install nix to use as your system package manager etc:</p>

home-manager</a>, source code</a></li>
</ul>

You do not necessarily need home-manager, but it automates nicely setting up your environment, handle dotfiles etc.</p>
On a non-nix machine, I first install nix, then home-manager (root access needed), in single-user mode.</p>
I then setup my home.nix file, and then run :</p>
home-manager</span> switch</span></span></code></pre>
This installs and sets up everything as defined in the home.nix file.</p>
Specifics</h3>
Missing /etc/nix/nix.conf</code> or ~/.config/nix/nix.conf</code></h4>
It's okay to create those files.</p>
Flatpaks</h4>

wiki page on Flatpaks in nix</a></li>
</ul>
Apparently, installing the package is enough to add flatpaks to the system.</p>


Fripes à Bruxelles
2025-12-04T00:00:00+00:00



Bonne question pour savoir comment l'autre personne va
2025-11-30T00:00:00+00:00
Liste de question à poser à la place de «Comment ça va ?»</strong>.</p>

T'as des belles choses de prévus aujourd'hui/cette semaine ?</li>
Est-ce que tu as fait des belles choses aujourd'hui/cette semaine ?</li>
À quoi penses-tu ces derniers temps ?</li>
Y a-t-il eu quelque chose d'intéressant (excitant, cool, amusant) récemment ?</li>
Alors, à quoi ressemble l'avenir ?</li>
Comment te sens-tu ces derniers jours ?</li>
Comment était ta semaine ?</li>
Qu'est-ce qui occupe ton temps ces jours-ci ?</li>
</ul>

Quel était le highlight de ta semaine ?</li>
Tu fais quoi de tes journée ?</strong></li>
</ul>


</ol>
</section>


Python
2025-11-30T00:00:00+00:00
Learning Ressources</h2>

Mathspp</a></li>
Pydon'ts – Write elegant Python code</a></li>
</ul>


Fabulous Hack
2025-11-18T00:00:00+00:00

Rockbox - Free Music Player Firmware</a></li>
kavishdevar/librepods: AirPods liberated from Apple's ecosystem.</a></li>
</ul>


Thunderbird
2025-11-18T00:00:00+00:00
Troubleshooting</h2>
Inbox not working correctly</h3>
→ Thunderbird Unified Inbox Not Showing All Accounts</a> web archive</a></p>


Fabulous Manuals
2025-11-11T00:00:00+00:00


Easy RISC-V</a>, an interactive introduction to RISC-V assembly programming, by dramforever</a>.</p>
</li>

Helix Editor | Helix</a>, alternate pedagogie helix</a> manual.</p>
</li>
</ul>


Android
2025-11-11T00:00:00+00:00

Comparison of Android-based Operating Systems</a> → great analysis of security in different android-based ROMs, notably GrapheneOS</a> and CalyxOS</a>.</li>
</ul>


Flour Glue
2025-11-11T00:00:00+00:00



Euro Cloud Providers
2025-11-04T00:00:00+00:00


European cloud computing platforms | European Alternatives</a></p>
</li>

Prices of the gridscale cloud services</a></p>
</li>

VPS Hosting UK | Cheap » from £1/mth | IONOS</a></p>
</li>

Europe's empowering cloud provider | Scaleway</a></p>
</li>

European virtual private server (VPS) hosters | European Alternatives</a></p>
</li>

VPS configurator: configure your VPS yourself</a></p>
</li>

UpCloud Cloud Server Infrastructure</a></p>
</li>

VPS - Your virtual private server in the cloud | OVHcloud Worldwide</a></p>
</li>

Cloud Portfolio - gridscale</a></p>
</li>

Cyso European Cloud Pricing – Transparent & Flexible</a></p>
</li>

My Cyso | Cloud</a></p>
</li>

Configuration du serveur | IONOS</a></p>
</li>

Cheap hosted VPS by Hetzner: our cloud hosting services</a></p>
</li>
</ul>


Forward Proxy
2025-10-23T00:00:00+00:00


Linux Proxy Server Settings - Set Proxy For Command Line - ShellHacks</a></p>
</li>

💤 linux - How do I set systemwide connection over a proxy server - Unix & Linux Stack Exchange</a></p>
</li>

💤 squid : Optimising Web Delivery</a></p>
</li>

💤 docker-tinyproxy/docker-compose.yml at master · kalaksi/docker-tinyproxy</a></p>
</li>

💤 tarampampam/3proxy-docker: 🥷 Docker image with 3proxy - Tiny free proxy server</a></p>
</li>

💤 3proxy/3proxy: 3proxy - tiny free proxy server</a></p>
</li>

💤 Install — Caddy Documentation</a></p>
</li>

💤 Keep Caddy Running — Caddy Documentation</a></p>
</li>

Build from source — Caddy Documentation</a></p>
</li>
</ul>


ShadowDark
2025-10-16T00:00:00+00:00
System</h2>

ShadowDark</a></li>
The Arcane Librarie's Youtube Channel</a></li>
</ul>
Ressources</h2>

ShadowDarklings</a></li>
</ul>


Fabulous Rust
2025-09-28T00:00:00+00:00
Tools</h2>
Language</h3>


facet.rs</a> - facet is a derive macro and a trait that gives runtime (and to some extent, const-time) knowledge about the shape, trait implementations, and characteristics of arbitrary types.</em></p>
</li>

malva</a> - Configurable, smart and fast CSS, SCSS, Sass and Less formatter with dprint integration.</em></p>
</li>
</ul>
Document creation</h3>

typst</a> - Typst is a new markup-based typesetting system that is designed to be as powerful as LaTeX while being much easier to learn and use</em></li>
hayro</a> - An experimental, work-in-progress PDF interpreter and renderer.</em></li>
oxidize-pdf</a> - A pure Rust PDF generation and manipulation library with zero external PDF dependencies. Production-ready for basic PDF functionality. Generate PDFs 2x faster than PDFSharp, with memory safety guarantees and a 5.2MB binary size.</em></li>
mupdf-rs</a> - Rust binding to mupdf</a>, MuPDF is a lightweight open source software framework for viewing and converting PDF, XPS, and E-book documents.</em></li>
</ul>
Authors of note</h2>
This is  a listing of people that work on rust, whose work is too extensive to list in other sections. I however do not consider those people better then the devs working on projects solo etc.</p>

Armin Ronacher / mitsuhiko</a></li>
</ul>


Maps
2025-09-24T00:00:00+00:00
Commercial Maps</h2>

Czepeku</a></li>
Tom Cartos</a></li>
</ul>


Virtual Table Top
2025-09-24T00:00:00+00:00
Online VTT</h2>
Roll20</h3>
Owlbear</h3>
Owlbear Rodeo</a> is new kid in the block online VTT.
Launched in ... by two australians, it aims to be a minimalist tool that focuses on maps and tokens.</p>
It is a React web app that uses the Material You design style.</p>
Performance</strong></p>
In my opinion, it is snappier and faster then Roll20.</p>


Audio in TTRPG
2025-09-23T00:00:00+00:00
Original Soundtracks</h2>
Video Games</h3>
all souls games</strong></p>

Dark Souls 1</a></li>
Dark Souls 3</a></li>
Bloodborn</a></li>
Elden Ring</a></li>
</ul>
Other Games</strong></p>

The Witcher 3: The Wild Hunt</a></li>
</ul>
Films</h3>

King Arthur</a></li>
</ul>
TV Shows</h3>

Carnival Row</a></li>
</ul>


Fabulous Communication
2025-09-16T00:00:00+00:00
Visioconférence1</a></sup></h2>
Liste des instances gratuites et publiques de Jitsi / BigBlueButton</em></p>
Jitsi Meet via CHATONS</h3>
Collectif des Hébergeurs Alternatifs,Transparents, Ouverts, Neutres et
Solidaires</em></p>


Colibris Outils Libres (AS24940 Hébergé en Allemagne en IPv4 uniquement) :
https://visio.colibris-outilslibres.org/</a></p>
</li>

TeDomum (AS12876 Hébergé en France et IPv6 dispo) :
https://meet.tedomum.net/</a></p>
</li>

Artifaille (AS12322 Hébergé sur une Freebox FTTH en IPv4 uniquement) :
https://meet.artifaille.fr/</a></p>
</li>

Alolise (AS199422 hébergé en France en IPv4 uniquement)
https://meet.alolise.org</a></p>
</li>

roflcopter.fr (AS24940 hébergé en Allemagne en IPv4 uniquement)
https://meet.roflcopter.fr/</a></p>
</li>

HADOLY (Hébergeur Associatif Décentralisé et Ouvert à Lyon) (AS51083 Hébergé
en France et IPv6 dispo) :
https://jitsi.hadoly.fr/</a></p>
</li>

devloprog.org (AS5410, hébergé sur une Bbox FTTH en IPv4 uniquement) :
https://meet.devloprog.org</a></p>
</li>

Pâquerette (AS29222 hébergé en Suisse en IPv4 uniquement)
https://visio.paquerette.eu/</a></p>
</li>

Algoo (AS12876 Hébergé en France en IPv4 uniquement)
https://suricate.tv/</a></p>
</li>

DryCat (AS24940 Hébergé en Allemagne et IPv6 dispo)
https://meet.drycat.fr/</a></p>
</li>
</ul>

Jitsi Meet via autres acteurs</h3>


WebConférence de l'État (AS60855 Hébergé en France en IPv4 uniquement) :
https://webconf.numerique.gouv.fr/</a>

Restriction : obligation de créer le salon depuis le RIE (Le réseau
interministériel de l'État)</p>
</li>

Renater :
https://rendez-vous.renater.fr/home/</a>

Restriction : Service réservées aux établissement d’enseignement supérieur.
obligation de s'identifier.</p>
</li>

Rezopole (AS199422 Hébergé en France et IPv6 dispo) :
https://meeting.rezopole.net/</a></p>
</li>

Hivane NetWork (AS34019 Hébergé en France et IPv6 dispo) :
https://jitsi.hivane.net/</a></p>
</li>

Elkaïm Élie (AS12876 Hébergé en France et IPv6 dispo) :
https://framatalk.jesuisfrancobelge.eu/</a></p>
</li>

Nextmap.io (AS35426 Hébergé en France et IPv6 dispo) :
https://jitsi.nextmap.io/</a></p>
</li>

Association Viviers Fibre (AS16276 Hébergé en France en IPv4 uniquement) :
https://webconf.viviers-fibre.net/</a></p>
</li>

SilentT – Clément Breuil (AS62000 Hébergé en France en IPv4 uniquement) :
https://jitsi.silentt.fr/</a></p>
</li>

MilkyWan (AS57199 Hébergé en France et IPv6 dispo) :
https://jitsi.milkywan.fr/</a> (cf
Jitsi/Framatalk : Un service de visioconférence pour aider les télétravailleurs</a>)</p>
</li>

Grifon Fédération FDN (AS204092 Hébergé en France et IPv6 dispo) :
https://framatalk.grifon.fr/</a></p>
</li>

Vevox (AS16276 Hébergé en France en IPv4 uniquement) :
https://videoconf.wevox.eu</a></p>
</li>

Florian / minzord (AS197922 Hébergé en France et IPv6 dispo) :
https://jitsi.minzord.eu.org/</a></p>
</li>

Vidéos du Libre (AS197422 Hébergé en France et IPv6 dispo) :
https://jitsi.videodulib.re/</a></p>
</li>

BIM (Bienvenue sur l'Internet Montpellierain) (AS200162 Hébergé en France en
IPv4 uiquement) : https://allo.bim.land/</a></p>
</li>

thedark (AS29222 Hébergé en Suisse en IPv4 uiniquement) :
https://jitsi.thedark.xyz/</a></p>
</li>

BrainMill (AS48514 Hébergé en Suède et IPv6 dispo) :
https://jitsi.brainmill.com/</a></p>
</li>

Freifunk Düsseldorf (AS207871 Hébergé en Allemagne et IPv6 dispo) :
https://jitsi.freifunk-duesseldorf.de/</a></p>
</li>
</ul>
Jitsi via Grosse Instance</h3>


Framasoft (redirige vers un pool de serveurs chez différents hébergeurs en
Europe) :
https://framatalk.org/accueil/fr/</a>
(attention, il peut être surchargé compte tenu de sa popularité)</p>
</li>

Celle des développeurs Jitsi hébergé chez Amazon en IPv4 uniquement (AS16509)
: https://jitsi.org/jitsi-meet/</a> (qui redirige
vers https://meet.jit.si/</a> ) (attention, il peut être
surchargé compte tenu de sa popularité)</p>
</li>

Jitsi hébergé chez l'opérateur Free:
https://jitsi.free.org/</a></p>
</li>
</ul>




Auteur (n.d.</em>). Visioconférence: Liste des instances gratuites et
publiques de Jitsi</em>. Visioconférence: Liste des instances gratuites et
publiques de Jitsi / BigBlueBu, Retrieved September 16, 2025 at 12:02:54
from
https://lafibre.info/navigateurs/visioconference/</a> -
🏛 archive</a> ↩</a></p>
</li>
</ol>
</section>


Fabulous Board Games
2025-09-09T00:00:00+00:00

regles du jeu.net</a></li>
</ul>


Fabulous Web
2025-09-09T00:00:00+00:00

Eternal Struggle</a></li>
DarkPattern.games » Healthy Gaming « Avoid Addictive Dark Patterns</a></li>
Acrylic Match</a></li>
Browse - Harry Ransom Center Digital Collections</a></li>
How Did I Get Here</a></li>
XSLT.RIP - Google are killing XSLT!</a></li>
About - Project Euler</a></li>
</ul>
Platforms</h2>

OpenBenches Welcome!</a></li>
Bookshelf Town</a></li>
</ul>
Personal Websites</h2>

Chase McCoy</a></li>
Design History | Kev Quirk</a></li>
Hi, I'm Heather Burns – — yes, that one.</a></li>
</ul>


My setup on linux
2025-09-09T00:00:00+00:00
General</h2>
Those couple of pages have pretty good recommendation for tools to use on your
system.</p>

Useful Add Ons for Sway</a></li>
Useful Utilities - hyprland wiki</a></li>
Are we Wayland yet</a></li>
</ul>
Sway</h2>
Media</h3>

avizo</a> -

a simple notification daemon, mainly intended to be used for multimedia keys
for example with Sway.</p>
</blockquote>
</li>
</ul>
Utility</h3>


lswt</a> - list Wayland toplevels. →
AUR</a></p>
You can alternatively use: swaymsg -t get_tree</code></p>
for app_ids:</p>
swaymsg</span> -t</span> get_tree</span> |</span> jq</span> -r</span> '</span>.. | .name? // empty</span>'</span></span></code></pre>
or for names:</p>
swaymsg</span> -t</span> get_tree</span> |</span> jq</span> '</span>.. | try .app_id? // empty</span>'</span></span></code></pre></li>
</ul>


date
2025-09-05T00:00:00+00:00
Use:</p>
man date</code></p>
or if you have a tldr client (like tealdeer</code>):</p>
tdlr date</code></p>
Open a file with today's date:</h2>
# Your editor + the date expression + the rest of the file name + file extension </span></span>
hx</span> $(</span>date</span> '</span>+%Y-%m-%d</span>')</span>_something.md</span></span></code></pre>


Créer un ficher avec la date
2025-09-05T00:00:00+00:00
Créer un fichier avec la commande date</h1>
Il est courant d'utiliser la commande [[date]] pour générer dynamiquement un nom
de fichier basé sur la date ou l'heure actuelle. Cette technique est utile pour
organiser des fichiers ou des journaux.</p>
Créer un fichier nommé avec la date</h2>
Pour créer un fichier utilisant la date actuelle dans son nom, combinez date</code>
avec une commande comme touch</code> (création de fichiers) ou un éditeur de texte.
Exemple :</p>
touch</span> "$(</span>date</span> '</span>+%Y-%m-%d</span>')</span>.txt</span>"</span></span></code></pre>
Cela crée un fichier nommé 2024-11-20.txt</code>.</p>
Si vous voulez inclure l'heure, utilisez un format étendu :</p>
touch</span> "$(</span>date</span> '</span>+%Y-%m-%d_%H-%M-%S</span>')</span>.txt</span>"</span></span></code></pre>
Ce fichier inclura également l'heure, par exemple : 2024-11-20_14-30-45.txt</code>.</p>
Utiliser un éditeur de texte</h2>
Si vous voulez créer et ouvrir un fichier directement dans un éditeur comme
nano</code>, vim</code>, ou micro</code>, utilisez la commande suivante :</p>
micro</span> "$(</span>date</span> '</span>+%Y-%m-%d</span>')</span>.txt</span>"</span></span></code></pre>
Cela ouvrira le fichier 2024-11-20.txt</code> dans l'éditeur.</p>
Automatisation avec un script</h2>
Pour simplifier le processus, créez un script Bash ou Zsh :</p>
Exemple :</strong></p>
#!/bin/bash</span></span>
</span>
# Créer ou ouvrir un fichier nommé avec la date actuelle</span></span>
</span>
FILENAME</span>="$(</span>date</span> '</span>+%Y-%m-%d</span>')</span>.txt</span>"</span></span>
</span>
touch</span> "</span>$FILENAME</span>"</span></span>
</span>
echo</span> "</span>Fichier créé : </span>$FILENAME</span>"</span></span></code></pre>
Rendez le script exécutable :</p>
chmod</span> +x script_date.sh</span></span></code></pre>
Puis exécutez-le :</p>
./script_date.sh</span>&</span></span></code></pre>Applications pratiques</h2>


Gestion de journaux</strong> : Créez des fichiers de log ou des rapports
quotidiens automatiquement nommés.</p>
</li>

Organisation de projets</strong> : Archivez vos notes ou tâches par date.</p>
</li>

Sauvegardes</strong> : Nommez les fichiers de sauvegarde avec une date et une
heure pour faciliter le suivi des versions.</p>
</li>
</ol>
Ressources</h2>


Consultez man date</code> pour explorer les options de format.</p>
</li>

Explorez des scripts avancés pour des tâches automatisées avec des dates.</p>
</li>
</ul>


Rust Syntax
2025-09-05T00:00:00+00:00

Primitives</strong> = the categories/types themselves (like i32</code>, bool</code>, char</code>).</p>
Literals</strong> = the actual written values you put in your code (like 5</code>,
"hello"</code>, true</code>).</p>
aka</p>
Primitive</strong> = “money</em>” (the concept of a type of thing)</p>
Literal</strong> = “$10 bill in my hand</em>” (a specific value of that type)</p>

Type</strong> = Dog</code> (what it is)</p>
Trait</strong> = Speak</code> (what it can do)</p>
Method</strong> = lick</code> and run</code> (actions a Dog knows how to do)</p>
Function</strong> = double</code> (general action not tied to Dog)</p>
Expression</strong> = x * 2</code>, Dog::new(...)</code>, self.age += 1</code> (the little building
blocks inside everything)</p>


date
2025-09-05T00:00:00+00:00
DIY Sacs de vélo et autres matos</h1>
General</h2>

https://bikepacking.com/news/the-functional-sewing-project/</a></li>
https://www.thefunctionalsewingproject.com/</a></li>
https://bikepacking.com/tag/myobg/</a></li>
</ul>
Projects Spécifique:</h2>


https://bikepacking.com/gear/a-diy-waxed-canvas-frame-bag/</a></p>
</li>

https://bikepacking.com/gear/make-your-own-stem-bag/</a></p>
</li>

https://bikepacking.com/gear/how-to-make-a-bikepacking-framebag-krampus/</a></p>
</li>

https://www.youtube.com/watch?v=oWDRiexLA0I</a></p>
</li>

https://bikepacking.com/gear/make-trail-wallet/</a></p>
</li>

https://bikepacking.com/gear/make-your-own-fanny-pack/</a></p>
</li>

https://bikepacking.com/plog/bike-tire-belt/</a></p>
</li>

https://bikepacking.com/gear/make-cycling-cap/</a></p>
</li>
</ul>
Accessoire Vélo</h3>
https://bikepacking.com/index/add-cage-mounts-bike/</a> *
https://bikepacking.com/gear/tubus-vega-surly-ecr-rack-mods-extra-h20/</a>
https://bikepacking.com/news/usps-mailer-bikepacking-bags/</a></p>


sac
2025-09-05T00:00:00+00:00
Stuff sack</h3>


Workspace
2025-09-01T00:00:00+00:00
</p>
</p>
</p>
</p>
</p>


Shell tricks
2025-09-01T00:00:00+00:00
Fuzzy-Finding and Opening Files in Helix (fzf / skim)</h3>


Open file with fzf:</p>
hx</span> $(</span>fzf</span>)</span></span></code></pre></li>

Define a shell function (in ~/.bashrc</code> or ~/.zshrc</code>):</p>
fhx</span>() {</span> hx</span> "$(</span>fzf</span>)";</span> }</span></span></code></pre>
Then just run fhx</code>.</p>
</li>

Limit to project files with fd</code> or rg</code>:</p>
hx</span> $(</span>fd</span> -t</span> f</span> |</span> fzf</span>)</span></span>
# or</span></span>
hx</span> $(</span>rg</span> --files</span> |</span> fzf</span>)</span></span></code></pre></li>

Open multiple files:</p>
hx</span> $(</span>fzf</span> -m</span>)</span></span></code></pre></li>
</ul>
🔀 Alternative:</strong> You can use
skim (sk)</strong></a>, a Rust-based fzf clone, in
the same way:</p>
hx</span> $(</span>sk</span>)</span></span></code></pre>Find and remove file</h3>
rm</span> $(</span>sk</span>)</span></span></code></pre>


Maps & Routing
2025-08-29T00:00:00+00:00
General Maps found online (desktop)</h2>
FOSS (privacy friendly ?)</h3>

OpenStreetMap</a></li>
GraphHopper</a></li>
Valhalla</a></li>
gpxStudio</a></li>
</ul>
Proprietary (non-privacy-friendly)</h3>

Google Maps</a></li>
AppleMaps</a></li>
Bing Maps, by Microsoft</a></li>
Waze</a></li>
ViaMichelin</a></li>
MapQuest</a></li>
</ul>


Site de Rando et Balades
2025-08-29T00:00:00+00:00
Générique</h2>

Decathlon Outdoor</a></li>
</ul>
Bretagne</h2>
Finistère</h3>

Tout Commence au Finistère</a></li>
Pay de Landerneau</a></li>
Presqu'île de Crozon</a></li>
Monts d'Arrée</a></li>
Parc Naturel Régional d'Armorique</a></li>
Menez-Hom Atlantique</a></li>
La Côte des Légendes</a></li>
roscoff - cote des légendes</a></li>
Baie de Morlaix</a></li>
</ul>


Fabulous Static Site Generators
2025-08-24T00:00:00+00:00
What are Static Site Generators →
The Wikipedia Article</a>.</p>
You can then also have a look at the following list for inspiration:</p>

Jamstack - Site Generators complete list</a></li>
</ul>
I've listed below my favorite SSG (Static Site Generators):</p>

Zola</a>, written in Rust, very fast, relavitely easy to
use. I use it for this website, and it has served me well.</em></li>
Eleventy</a>, written in Javascript, also quite fast,
you need to learn JS to be able to build something with this. It is however
very very VERY extensible, lots of template languages to choose from etc.</em></li>
Zine SSG</a>, written in Zig, in alpha version as of time
of writing, should be VERRRRY fast, probably in the ballpark of zola.
Implements it's own templating language and style using custom html tags, it's
own markdown superset that makes it more extensible. I'm very curious to see
where this goes.</em></li>
</ul>
Worthy mentions</h2>

mdBook</a>, similar to Gitbook, but in
Rust. Used by The Rust Book</em></li>
oranda</a>, build single page dev project
landing pages; I have not tried this, but it seems interesting. Written in
Rust</em></li>
</ul>


Bretagne
2025-08-22T00:00:00+00:00


Bretagne Vivante - Une voix pour la nature</a> -
la principale association de protection de la nature et de la biodiversité en
Bretagne.</p>

Tout commence en Finistère - balade</a></li>
</ul>
</li>
</ul>


cargo
2025-08-22T00:00:00+00:00
cargo</code> comes preinstalled with rustup.</p>
Usage</h2>


cargo new</code> → either creates a new folder with a Cargo.toml</code>, a src/main.rs</code>
and initializes a git repo (by default).</p>
</li>

cargo init</code> in a preexisting folder.</p>
</li>

cargo new --vcs</code> → to change the version control tool.</p>
</li>

cargo build</code> to compile and create an executable. --debug</code> (which is by
default) compiles faster then --release</code>.</p>
</li>

cargo run</code> to compile and immediatly run the resulting executable, without
making an executable. Faster then build</code></p>
</li>

cargo check</code> to check the code without making an executable nor running the
program. Serves to check if the program still compiles. It is faster then
run</code> and build</code>.</p>
</li>

cargo run/build --offline</code> to use local offline dep instead of online.</p>
</li>

cargo add</code> to add dependencies to the project (and automatically in the
Cargo.toml</code>).</p>
</li>
</ul>
External tooling</h2>

cargo-upgrades</a></li>
cargo-edit</a></li>
cargo-bloat</a></li>
cargo-diet</a></li>
</ul>
Both tools, when installed, can help automatically upgrade dependencies to latest compatible version.</p>


rustup
2025-08-20T00:00:00+00:00
Installation</h2>
Installation documentation can be found on the rustup.rs</a>
website.</p>
Alternatively, you can also use your distros packages version or
other means</a>.</p>
Usage</h2>
rustup update</code></p>
More on rustup</code> in my dedicated wiki page</a>.</p>
Once rustup</code> is installed, you can access The Rust Book</strong>, Rust by
Examples</strong> and the std library documentation</strong> locally and offline via the
cmd:</p>
rustup</span> doc</span></span></code></pre>rustup</span> doc</span> --book</span></span></code></pre>
Alternatively, the book mentions a interactive version of the Rust Book, via
Experiment: Improving the Rust Book</em></a>.</p>


Mini PCs
2025-08-14T10:06:08+02:00

Miniforum</a></li>
Beelink</a>, via
D.H.Hansson</a>, who is
also behind omarchy.</li>
Anker Prime Power Bank</a>,
also via
D.H.Hansson</a></li>
</ul>


License & Copyright
2025-08-14T00:00:00+00:00
Handling Third-Party Code and License Notices</h2>
When incorporating external code (directly or indirectly) into a multi licensed
or open source project, the following rules could apply:</p>
File-Level License Headers</h3>

Do not remove or alter</strong> existing SPDX headers or copyright notices in
third-party source files.</li>
If you modify such a file, add your own attribution</strong> line beneath the
original one:</li>
</ul>
// SPDX-FileCopyrightText: © 2023 Original Author <author@example.com></span></span>
// SPDX-FileCopyrightText: © 2025 Your Name <you@example.com></span></span>
// SPDX-License-Identifier: \[original license]</span></span></code></pre>

The original license identifier (e.g. MIT</code>, Apache-2.0</code>, AGPL-3.0-only</code>)
must remain unchanged unless the upstream license explicitly permits
re-licensing.</li>
</ul>
Project-Level Notices</h3>

In addition to per-file headers, the project must include a LICENSE</code> and/or
ThirdPartyNotices.txt</code> file that:</li>
Lists all third-party components</strong> bundled or redistributed.</li>
Preserves their license texts (MIT, Apache, BSD, GPL, AGPL, etc.).</li>
Includes copyright statements where required.</li>
</ul>
Transitive Dependencies</h3>

License obligations extend to both:</li>
Direct dependencies</strong> you intentionally import.</li>
Transitive dependencies</strong> bundled within them.</li>
Use tools such as license-checker</code> or oss-attribution-generator</code> to generate
a complete ThirdPartyNotices.txt</code>.</li>
</ul>
Special Considerations</h3>

Permissive licenses</strong> (MIT, BSD, Apache-2.0): generally require only
attribution and preservation of notices.</li>
Copyleft licenses</strong> (GPL, AGPL): may obligate distribution of full source
code if your package is deployed or accessed over a network.</li>
SPFx (SharePoint Framework)</strong>:</li>
Solutions must ship with a ThirdPartyNotices.txt</code> inside the .sppkg</code>
package.</li>
Avoid strong-copyleft libraries (GPL/AGPL) unless you are prepared to release
the entire SPFx solution’s source code.</li>
</ul>


Rclone
2025-08-14T00:00:00+00:00
sync</p>
bisync</p>
Alternatives</h2>

syncrclone</a> - alternative to
rclone bisync</code>. Implementations
differ</a>.</li>
</ul>


LLM serpent
2025-08-14T00:05:15+02:00

Embryology of a Language Model</a>, via
Robin Sloan</a>
→ Superbe generated visualisations of points in an LLM model. I love how it
looks like dry pastel, and the colors work really well together.</li>
</ul>

Image from a baby dragon, stolen from Robin Sloan's blog.</em></p>

Use Ollama with any GGUF Model on Hugging Face Hub</a>,
via my wiki entry on LLMs</a> → A welcome feature, that enables
pulling in custom quantized models.</li>
</ul>


State Surveillance, Arch and Psych
2025-08-13T11:09:22+02:00

Fight Chat Control</a> - Here we go again. I'm
definitly sending an email to my MP.</li>
Omarchy</a>, via 37signals</a> - a
"custom archlinux + hyperland setup". I'm definitly going to go look at the
repo</a> to see how they setup the whole
system.</li>
DARVO</a> - "deny, attack, reverse victim
and offender", common manipulation tactique.</li>
</ul>


Fabulous Audio
2025-08-13T00:00:00+00:00
Synth stuff</h2>

Composer's Desktop Project</a> via
Benn Jordan</a> - a mature and
wide-ranging suite of sound-manipulation programs. #cli</code></li>
</ul>


LLM
2025-08-13T00:00:00+00:00
Prompt Engineering</h2>
Context Engineering</h2>
Local inference and model libraries</h2>

Ollama</a></li>
LM Studio</a></li>
vllm</a> I have not tested this system.</em></li>
</ul>
How to download huggingface models via Ollama</h3>
via
Use Ollama with any GGUF Model on Hugging Face Hub</a></p>
ollama</span> run hf.co/{username}/{repository}:{quantization}</span></span>
# It also works with pull</span></span></code></pre>Interface</h2>
→ Ollama and LM Studio provide and inteface</p>

OpenWebUI</a> → this project is being used by a bunch of
LLM startups, notably z.ai and kimi</li>
</ul>
GGUF sources</h2>
Hugginface</a> is as of writing this the biggest models
repo.</p>
Here are some notable LLM model publishers:</p>

Bartowski</a></li>
Unsloth AI</a></li>
team mradermacher</a></li>
</ul>


Rust
2025-08-13T00:00:00+00:00
Installation</h2>
Install using rustup</a>.</p>
Usage</h2>
rustup</h3>
rustup</code> packages all things rust, including rustc</code>,cargo</code>,rust-analyzer</code>,
and the std</code> lib.</p>
rustc</h3>
rustc</code> compiles a binary. Usually we use cargo</code> instead of rustc</code> directly.</p>
cargo</h3>
cargo</code> is the package manager and utility to build/run/check a rust project.</p>
See cargo</a></p>
rust-analyzer</h3>
Syntax</h2>
fn main()</code> is always the first function to run in a rust program (probably true
for most programming languages).</p>
Most lines of Rust code end with a semicolon.</strong></p>
Rust importst the std
prelude</a>. Other std libraries
also have preludes, but need to be explicitly brought in to scope with use</code>.</p>
println!()</code> is a macros</a> that prints a string to the screen.</p>
Variables are immutable by default. To make a mutable variable, you need to
prefix with mut</code> → let mut apples = 5;</code>.</p>
::</code> → associated function of a type. For example, String::new()</code> is the
new()</code> function of String</code>.</p>

An associated function</em> is a function that’s implemented on a type[...].
You’ll find a new</code> function on many types because it’s a common name for a
function that makes a new value of some kind.</p>
</blockquote>
diff between types, functions and methods ? (and traits ...)</p>
A handle</strong> is just an indirect reference to a resource.</p>
Instead of directly giving you the resource (like the actual keyboard device, or
a file), the system gives you a small object or identifier that you can use to
interact with it.</p>
You can think of it as a ticket or remote control that lets you operate
something else.</p>
use std::io</code> could also be directly called with std::io::Sdin</code> or somthing.</p>

The &</code> indicates that this argument is a reference, which gives you a way to
let multiple parts of your code access one piece of data without needing to
copy that data into memory multiple times.</p>
</blockquote>

It’s often wise to introduce a newline and other whitespace to help break up
long lines when you call a method with the .method_name()</code> syntax.</p>
</blockquote>
Enum & methods</h3>
Result</code> return value is an enum</code> type, which means it can have multiple
possible states, called variants</strong>.</p>
Result</code>'s variants are Ok</code> and Err</code>.</p>

Values of the Result type, like values of any type, have methods defined on
them.</p>
</blockquote>
Macros</h3>
println()</code> is a function, println!()</code> is a macros. Rust macros are a way to
write code that generates code to extend Rust syntax. Macros don’t always follow
the same rules as functions.</p>

Dependencies</h2>
Where to look for crates a.k.a. libraries</h3>

docs.rs</a></li>
crates.io</a></li>
</ul>
Update a dependency to the latest version</h3>
# 1. Edit Cargo.toml</span></span>
[</span>dependencies</span>]</span></span>
crate_name</span> =</span> "</span>x.y</span>"</span>   # set to desired/latest version</span></span>
</span>
# 2. Update</span></span>
cargo</span> update</span> -p</span> crate_name</span></span>
</span>
# 3. Verify</span></span>
cargo</span> tree</span> |</span> grep</span> crate_name</span></span></code></pre>

cargo update -p crate_name</code></strong> → updates only that dependency</li>
cargo update</code></strong> → updates all dependencies</li>
Loosen version in Cargo.toml</code> if updates are blocked by constraints</li>
</ul>
Compiling</h2>
Types of builds</h3>
--dev</code> builds are optimized at level 0, which means it's unoptimized, but is
also faster to compile.</p>
cargo</span> build</span> --dev</span></span></code></pre>
--release</code> flag indicates an optimization level of 3, the maximum. It takes
longer to compile.</p>
cargo</span> build</span> --release</span></span></code></pre>Cross Compiling from Archlinux</h3>
Requirements</h4>
You first need rustup</code> to be able to add targets.</p>
Windows</strong></p>
sudo</span> pacman</span> -S</span> mingw-w64-gcc</span></span></code></pre>
Linux</strong></p>
Since we already on linux, we don't need much. However, to compile to ARM</strong>,
we need:</p>
sudo</span> pacman</span> -S</span> aarch64-linux-gnu-gcc</span></span></code></pre>
and then set it as the linker for the ARM target in .cargo/config.toml</code>(local):</p>
[</span>target</span>.</span>aarch64-unknown-linux-gnu</span>]</span></span>
linker</span> = "</span>aarch64-linux-gnu-gcc</span>"</span></span></code></pre>
Cargo ignores linker settings in Cargo.toml. It only reads them from
.cargo/config.toml</code> (project-local) or ~/.cargo/config.toml</code> (global).</p>
MacOS</strong></p>
I don't know yet...</p>
Maybe this
MacOS Cross-Compiler - github.com</a></p>
https://studios.ptilouk.net/superfluous-returnz/blog/2022-03-16_macos.html
https://doc.rust-lang.org/beta/rustc/platform-support/apple-darwin.html
https://doc.rust-lang.org/beta/rustc/platform-support.html
https://wapl.es/rust/2019/02/17/rust-cross-compile-linux-to-macos.html/ ← seems
promissing</p>
https://github.com/cross-rs/cross</p>
https://github.com/tpoechtrager/osxcross</p>


Timeline visualisations
2025-08-12T11:34:37+02:00


Cursed Knowledge - immich.app</a> - a
funny list of cursed knowledge</em> from building the immich.app. It made me
discover notably goja</a>, a
ECMAScript/JavaScript engine in pure Go. I'm supprised that runtime was not
mentioned in
The many, many, many JavaScript runtimes of the last decade</a>
article I read last month.</p>

Historical Tech Tree</a> - a cool
visualisation of technology overtime.#depnum</code></li>
</ul>
</li>
</ul>


Interesting Links #1
2025-08-11T00:00:00+00:00
A while back, I stumbled</a> upon this personal website:
webb.page</a>.</p>
I found the horizontal layout quite interesting. Unfortunately, I find it also
jarring to alternate scrolling directions when making a single (in my case)
downwards motion.</p>


First microblog post
2025-08-11T00:00:00+00:00

Wplace - Paint the world</a> a new project that is similar
to reddit /r/places.</li>
Horizon at ...</a></li>
</ul>


HTML
2025-08-11T00:00:00+00:00
Syntax</h2>

the output</a> tag, via Den Odell</a></li>
the hgroup</a> tag</li>
the meter</a> tag, via Terence Eden</a></li>
</ul>
Howto's & Tricks</h2>
Open link in new tab</strong></p>
<</span>p</span>></span></span>
  Check out </span><</span>a</span></span>
    href</span>="</span>https://example.com/</span>"</span></span>
    target</span>="</span>_blank</span>"</span></span>
    rel</span>="</span>noopener noreferrer</span>"</span></span>
  ></span>SomeLink</span></</span>a</span>></span>.</span></span>
</</span>p</span>></span></span></code></pre>
target="_blank"</code> to open in new tab, rel="noopener noreferrer"</code> to avoid
tabnabbing</a>, a type of phishing
attack via malecious login page doppelganger.</p>

Source</a></li>
</ul>
Auto-Reload</strong></p>

Makeshift hot reload - HTMHell</a></li>
</ul>


Events Bruxelles
2025-07-07T00:00:00+00:00
Where to find out about events in Bruxelles, excluding facebook.</p>


Surveillance Capitalism
2025-06-19T00:00:00+00:00



javascript
2025-06-17T00:00:00+00:00

Resolving EACCES permissions errors when installing packages globally | npm Docs</h1>
If you see an EACCES</code> error when you try to
install a package globally</a>,
you can either:</p>


Reinstall npm with a node version manager (recommended),</p>
or</strong></p>
</li>

Manually change npm's default directory</p>
</li>
</ul>
Reinstall npm with a node version manager</a></h2>
This is the best way to avoid permissions issues. To reinstall npm with a node
version manager, follow the steps in
"Downloading and installing Node.js and npm</a>".
You do not need to remove your current version of npm or Node.js before
installing a node version manager.</p>
Manually change npm's default directory</a></h2>
Note:</strong> This section does not apply to Microsoft Windows.</p>
To minimize the chance of permissions errors, you can configure npm to use a
different directory. In this example, you will create and use hidden directory
in your home directory.</p>


Back up your computer.</p>
</li>

On the command line, in your home directory, create a directory for global
installations:</p>
mkdir -p ~/.npm-global/lib</code></p>
</li>

Configure npm to use the new directory path:</p>
npm config set prefix '~/.npm-global'</code></p>
</li>

In your preferred text editor, open or create a ~/.profile</code> file and add
this line:</p>
export PATH=~/.npm-global/bin:$PATH</code></p>
</li>

On the command line, update your system variables:</p>
source ~/.profile</code></p>
</li>

To test your new configuration, install a package globally without using
sudo</code>:</p>
npm install -g jshint</code></p>
</li>
</ol>
Instead of steps 3-5, you can use the corresponding ENV variable (e.g. if you
don't want to modify ~/.profile</code>):</p>
NPM_CONFIG_PREFIX=~/.npm-global</code></p>
npx: an alternative to running global commands</strong></p>
If you are using npm version 5.2 or greater, you may want to consider
npx</a> as an alternative way to run
global commands, especially if you only need a command occasionally. For more
information, see
this article about npx</a>.</p>
References</h3>

(n.d.</em>). Resolving EACCES permissions errors when installing packages
globally | npm Docs</em>. Resolving EACCES permissions errors when installing
packages globally | npm Docs, Retrieved June 17, 2025 at 13:19:20 from
https://docs.npmjs.com/resolving-eacces-permissions-errors-when-installing-packages-globally#manually-change-npms-default-directory</a> -
🏛 archive</a></li>
</ul>


Setting Up Local HTTPS Services with Custom Domains Using Caddy on Arch Linux
2025-06-05T00:00:00+00:00
This guide explains how to serve multiple local services on custom domains (e.g.
.local</code> → service1.local</code>) with HTTPS</strong>, using Caddy</strong> and no external DNS
or certificates.</p>

🧱 Prerequisites</h3>

Arch Linux (or derivative)</li>
Services running locally on different ports (e.g., 8080, 9090)</li>
caddy</code> installed:sudo</span> pacman</span> -S</span> caddy</span></span></code></pre></li>
</ul>

🔧 Step 1: Map Local Hostnames</h3>
Edit /etc/hosts</code>:</p>
sudo</span> nano /etc/hosts</span></span></code></pre>
Add:</p>
127.0.0.1  service1.local service2.local service3.local</span></span></code></pre>

📁 Step 2: Configure Caddyfile</h3>
Edit or create your Caddyfile (default path: /etc/caddy/Caddyfile</code>):</p>
service</span>1</span>.local </span>{</span></span>
    tls internal</span></span>
    reverse_proxy 127.0.0.1</span>:</span>8080</span></span>
}</span></span>
</span>
service</span>2</span>.local </span>{</span></span>
    tls internal</span></span>
    reverse_proxy 127.0.0.1</span>:</span>9090</span></span>
}</span></span></code></pre>
This tells Caddy to:</p>

Use internal TLS (self-signed CA)</li>
Reverse-proxy requests based on hostname and forward to corresponding ports</li>
</ul>

🚀 Step 3: Start Caddy</h3>
Restart the service to apply changes:</p>
sudo</span> systemctl restart caddy</span></span></code></pre>

🔐 Step 4: Trust Caddy’s Local CA (One-Time Setup)</h3>
1. Copy the root certificate:</h4>
sudo</span> cp /var/lib/caddy/pki/authorities/local/root.crt /etc/ca-certificates/trust-source/anchors/caddy-local.crt</span></span></code></pre>2. Update the system trust store:</h4>
sudo</span> trust extract-compat</span></span></code></pre>3. (Optional) Trust in Firefox:</h4>

Preferences → Privacy & Security → Certificates → View Certificates →
Authorities → Import</li>
Choose /var/lib/caddy/pki/authorities/local/root.crt</code></li>
Check “Trust this CA to identify websites”</strong></li>
</ul>

✅ Done!</h3>
Now you can access:</p>

https://service1.local</code> → 127.0.0.1:8080</code></li>
https://service2.local</code> → 127.0.0.1:9090</code></li>
</ul>
All with trusted HTTPS</strong>, locally.</p>


Accessing Host Services from Inside Podman Compose Containers
2025-06-04T00:00:00+00:00
Accessing Host Services from Inside Podman Compose Containers</h1>
Overview</h2>
When running containers with Podman Compose</strong>, software inside the container may need to access services running on the host machine</strong> (e.g., a database, API, or development server). However, localhost</code> inside a container refers to the container itself—not the host.</p>
This guide explains how to correctly route traffic from a container to the host.</p>

ℹ️ Note</strong>: This method also works with Docker Compose</strong>, as long as the container runtime supports host.docker.internal</code> or equivalent DNS resolution.</p>
</blockquote>

✅ Recommended Method: host.containers.internal</code></h2>
Podman provides a special internal DNS name to reference the host from inside a container:</p>
</span>
host.containers.internal</span></span>
</span></code></pre>
This works with Podman 3.4+</strong> and is the most reliable and secure method.</p>
Example</h3>
If your host is running a service on localhost:8000</code>, you can access it inside the container as:</p>
http://host.containers.internal:8000</a></p>
No additional configuration is required.</p>

🔧 Using Podman Compose</h2>
1. Run Your Host Service</h3>
Ensure the service you want to access is running on the host (e.g., localhost:1234</code>).</p>
2. Configure Your Container</h3>
Update your application or environment variables to reference the special host address. Example docker-compose.yml</code> (compatible with Podman Compose):</p>
services</span>:</span></span>
  app</span>:</span></span>
    image</span>:</span> your-image</span></span>
    environment</span>:</span></span>
      -</span> API_URL=http://host.containers.internal:1234</span></span></code></pre>3. Start Podman Compose</h3>
podman-compose</span> up</span></span></code></pre>
Your containerized application will now connect to the host service via the internal alias.</p>

🛠 Troubleshooting</h2>


If host.containers.internal</code> doesn't resolve:</p>

Ensure you're using Podman 3.4+.</li>
Test DNS inside the container: ping host.containers.internal</code></li>
Manually add a host entry via volume mount if needed.</li>
</ul>
</li>
</ul>

⚠️ Alternative: Host Network Mode (Linux Only)</h2>
Use network_mode: host</code> to fully expose the host network to the container:</p>
services</span>:</span></span>
  app</span>:</span></span>
    image</span>:</span> your-image</span></span>
    network_mode</span>:</span> host</span></span></code></pre>

❗ Warning</strong>: This disables network isolation and may cause port conflicts. Only use when necessary.</p>
</blockquote>

✅ Summary</h2>
To allow a container to access services on the host:</p>

Use host.containers.internal:<port></code> inside the container.</li>
Works with Podman Compose</strong> and Docker Compose</strong> (on compatible systems).</li>
Avoid network_mode: host</code> unless isolation is not required.</li>
</ul>


Scribus
2025-05-18T00:00:00+00:00
Import an SVG</h2>
File > Import > Import an SVG</code></p>


*Fabulous LLM
2025-05-16T00:00:00+00:00
Here I list all sorts of tools that use LLMs (Large Language Models</em>). I feel
ambivalent about hosting a ressource list that use those insidious tools. When I
get around to it, I will link a blog post that talks about the problems inherent
to LLMs and other «AI» tools.</p>
In the mean time, here is a non-exhaustive list of articles on the subject of
LLM assisted creation (mostly coding).</p>

Here’s how I use LLMs to help me write code</a></li>
A cheat sheet for why using ChatGPT is not bad for the environment</a></li>
A trick to feel less like cheating when you use LLMs</a></li>
My LLM codegen workflow atm | Harper Reed's Blog</a></li>
programming with llms- crawshaw - 2025-01-06</a></li>
Everything I've learned so far about running local LLMs</a></li>
An example of LLM prompting for programming</a></li>
How I write code using Cursor: A review</a></li>
</ul>
VERY IMPORTANT →
Hallucinations in code are the least dangerous form of LLM mistakes</a></p>
Providers</h2>
To check prices for different models and providers, you can refer to this
incredible LLM Pricinc Calculator</a>. It has the
pricing for Input</strong> and Output</strong> costs for different models, as well as a
calculator that takes as parameters the number of input</strong> and output</strong>
tokens. This tool is made and maintained by the amazing
Simon Willison</a>.</p>
Cloud</h3>
Openrouter.ai</a> is a platform that enables you to have a
single api key to be used with multiple models, including private ones.</p>
Chats</h3>
Most, if not all services listed in the Chat section, will use your data for
training purposes. They are not suitable for privacy sensitive queries.</strong></p>


Claude</a>, by Anthropic - $</p>
</li>

Chatgpt</a> by OpenAI - $</p>
</li>

Gemini</a>, by Google - $</p>
</li>

AI Studio</a>, also by Google.</p>
</li>

Le Chat</a>, by Mistral ($)</p>
</li>

Deepseek</a>, by Deepseek $</p>
</li>

Qwen</a>, by QwenAI $</p>
</li>

GLM</a>, by z.ai</p>
</li>

Grok</a>, by x.ai → use is ethical very questionable, since
it is built by a E. Musk company. Know for being extreme right oriented, and
will actively provide erroneous opinions as facts.</p>
</li>

Phind</a>, by Phind. Creates a lot of diagrams</em></p>
</li>

Perplexity</a>, by PerplexityAI ($)</p>
</li>

Kimi</a></p>
</li>

Poe</a>, access to multiple models. #freetier</p>
</li>

Meta ($)</p>
</li>

Amazon $</p>
</li>

LMArena</a>, free random models</p>
</li>
</ul>
Local</h3>

Llama.ccp</a>, inference engine in C/C++,
very efficient and is the basis for most other serving tools.</li>
Ollama</a>, go based hosting tool, with a good DL directory
of models. It serves an OpenAI compatible API endpoint.</li>
Llamafile</a>
cross-platform model in a file, that runs both with CLI interface, web
interface, and can also serve as an OpenAI compatible API endpoint.</li>
LM Studio</a>, electron based local interface, model
curated, local server capable app. Not entirely open source though.</li>
</ul>
CLI</h2>

llm</a>, a python CLI tool with a
really cool range of features and uses, also by
Simon Willison</a>.</li>
Aider.chat</a>, a python cli llm «agent» that sort of
automates programming. You can connect all sorts of providers to it, including
your own locally served models (via Ollama,Llama.ccp, VLLM, LM Studio).</li>
Anthropic Code</a></li>
OpenAi Codex</a></li>
</ul>
Editors</h2>

Cursor</a>, electron VSCode-based LLM-enabled editor,
not FOSS.</li>
Void</a>, Cursor clone, FOSS.</li>
zed.dev</a>, rust IDE with AI agent, auto-complete. I really
like this IDE, since it is not electron based, but developed in rust
directly.</em></li>
</ul>
Training</h2>

p-e-wheretic Fully automatic censorship removal for language models</a></li>
</ul>


Stupid Violent Media
2025-05-14T00:00:00+00:00
Films that celebrate violence, especially really stupid nonsensical violence.
Which means most violence, maybe?</p>
Real Nonsense</h2>

Nobody</li>
John Wick</li>
</ul>
Political Nonsense</h2>

Warfare</li>
Civil War</li>
</ul>


Ereader
2025-05-14T00:00:00+00:00
Kobo Clara HD</h2>
Setup</h3>

Connect the device to a computer with the help of a microUSB cable.</li>
Bypass Registration</a></li>
Open the device, to access the microsd card that serves as storage.</li>
</ol>
Bypassing Registration1</a></sup></h3>
Connect the Clara HD to a computer as directed, but instead of downloading the
Kobo app, instead open the Clara HD internal storage on the computer, which
should have become available as a USB device.</p>
Navigate to the (possibly hidden by default) .kobo</code> directory, where there
should be a KoboReader.sqlite</code> file. This is an SQLite database, and we will
need to install the SQLite tools to access it (e.g. sudo pacman -S sqlite</code>).</p>
Once SQLite is installed, open a command prompt in this directory and run:</p>
sqlite3 KoboReader.sqlite</span></span></code></pre>
From the SQLite prompt, run:</p>
<span class="k">INSERT</span> <span class="k">INTO</span> <span class="k">user</span><span class="p">(</span><span class="n">UserID</span><span class="p">,</span><span class="n">UserKey</span><span class="p">)</span> <span class="k">VALUES</span><span class="p">(</span><span class="s1">'1'</span><span class="p">,</span><span class="s1">''</span><span class="p">);</span></span></span></code></pre>
And then exit SQLite. Then eject the Clara HD internal storage from the
computer, and disconnect it. The Clara HD should continue as if it were
successfully registered!</p>
Using this method enables all functionality of the Clara HD, including the beta
web browser, and firmware updates. To check for updates, connect to WiFi, then
tap the sync icon in the top right corner of the screen and tap ‘Sync now’. The
sync will fail, but the device should still check for firmware updates, and if
one is found, direct you to install it.</p>
Backup and restore</h3>
Remove the microSD</h4>
Remove the back cover of the e-reader and follow the safety measures if you
haven't yet. Then carefully remove the microSD card and connect it to your
computer.</p>
Backup2</a></sup></h4>
Now backup all the partitions at once using dd</code> and xz</code>.</p>
sudo</span> dd if=/dev/sde bs=4M conv=sync,noerror status=progress</span> |</span> xz</span> ></span> kobo_sdcard_</span>"$(</span>date</span> '</span>+%Y-%m-%d_%H-%M-%S</span>')"</span>.img.xz</span></span></code></pre>Flags</h4>


status=progress</code> enables progress information</p>
</li>

you can find the block size (bs</code>) of the device by running
stat -fc %s /dev/<microsd></code></p>
</li>
</ul>

Note: conv=sync,noerror</code> tells dd</code> that if it can't read a block due to a
read error, then it should at least write something to its output of the
correct length1</a></sup></p>
</blockquote>
Restore</h3>
To restore the backup, simply uncompress the image and write it to the microSD.</p>
xz</span> -dc</span> kobo-backup.img.xz</span> |</span> sudo</span> dd of=/dev/</span><</span>micros</span>d</span>></span> bs=4M status=progress</span></span></code></pre>Software</h2>
KOReader</h3>

Main Project Portal</a></li>
Github Repo</a></li>
Github Wiki Page</a> for Kobo devices which leads to the link below.</em></li>
Kobo Installation Instruction Posts</a></li>
</ul>
The last link is the most important and relevant one.</p>
Basically, installing KOReader on a Kobo device, after a couple of steps detailed in the instruction post, is as follows:</p>

Download the relevant .zip file → I use the package containing KoReader and Plato with KFmon</em></li>
Extract its content to the root of your device. This step is important, since there are hidden files (.add</code>, .koreader</code>, etc) that need to be added. You might miss them if you simply extract the zip file and copy it's content by hand at the root of your ereader.</li>
Reboot the device,</li>
Open the new Kfmon button on the bottom right,</li>
Wait for the Downloading page to disappear.</li>
Open KoReader by opening the dialoge with the KFmon button, and selecting KOReader.</li>
</ol>
Inkbox</h3>
Website:</em> https://inkbox.ddns.net</a></p>
Repo:</em> Inkbox Gitlab</a></p>
Here is the
MobileRead annoucement</a>
of Inkbox. It also features the Kobo Clara HD link.</p>
As of 2024-04-12 1</a></sup>, the provided InkBox OS image for the Clara HD (N249)</strong>
could only work with the following board revisions:</p>

37NB-E60K00+4A4</code></li>
37NB-E60K00+4A4+.C.</code> (may not fully work, but should boot)</li>
</ul>
This image will not work for the following board revision and could permanently
damage your device if you try to boot it:</p>

37NB-E60K00+4A4+.B.</code></li>
</ul>
For more details, please look here:
Kobo Clara not booting with mainline kernel - postmarketOS gitlab</a></p>
Miscaleanous</h2>
Custom Fonts3</a></sup></h3>

Bookerly() → AUR</li>
Volkorn (serif)</a></li>
LibreBaskerville (serif)</a> →
GoogleFonts</a></li>
LInux LIbertine</li>
</ul>
https://dmpop.github.io/koreader-compendium/01-custom-fonts/</p>
https://web.archive.org/web/20230610222521if_/https://thepenguin.eu/2020-01-02-some-nice-fonts-for-your-ebook-reader/</p>



https://yingtongli.me/blog/2018/07/29/kobo-sd.html</a> ↩</a> ↩2</a> ↩3</a></p>
</li>

https://www.inference.org.uk/saw27/notes/backup-hard-disk-partitions.html</a> ↩</a> ↩2</a> ↩3</a></p>
</li>

https://github.com/aartoni/n249/blob/main/backup.md</a> ↩</a></p>
</li>

https://web.archive.org/web/20230610222521if_/https://thepenguin.eu/2020-01-02-some-nice-fonts-for-your-ebook-reader/</a> ↩</a></p>
</li>
</ol>
</section>


Browser Extensions
2025-05-14T00:00:00+00:00
Development</h2>
Signing an Extension</h3>


Add-on signing in Firefox | Firefox Help</a></p>
</li>

Signing and distribution overview | Firefox Extension Workshop</a></p>
</li>
</ul>
Extensions I use:</h2>
MarkDownload</a>**</h3>
Back-matter template</strong></p>
I use it to get a decent reference to copy in my documentation or other markdown
documents for which I need a citation.</p>
### References</span></span>
</span>
-  {byline} (*n.d.*). *{title}*. {pageTitle}, Retrieved {date:MMMM D, YYYY [at] HH:mm:ss} from [{baseURI}]({baseURI}) - [🏛 archive](https://web.archive.org/web/{baseURI})</span></span></code></pre>
Example</strong></p>
### References</p>
- n.a. (n.d.</em>). How to Cite Sources | Citation Generator & Quick Guide</em>. How
to Cite Sources | Citation Generator & Quick Guide, Retrieved April 20, 2024
at 21:22:22 from
https://www.scribbr.com/category/citing-sources/</a> -
🏛 archive</a></p>


Laptops
2025-05-14T00:00:00+00:00
MSI GL63-8RE</h2>

BIOS Keybinding: Suppr</kbd>/Del</kbd></li>
Boot Menu Keybinding: F11</kbd></li>
</ul>
Issue with the Super</kbd> and Fn</kbd> keys switching sides, despite the option being changed in the BIOS.</p>
Software</h3>

MControlCenter</a> is a Free and
Open Source GNU/Linux application that allows you to change the settings of MSI
laptops.</li>
</ul>


waybar
2025-05-14T00:00:00+00:00
Waybar</h1>
Waybar is a</p>
It can use FontAwesome</a> &
Nerd Fonts</a> to display icons next to
your modules.</p>
For both you need to have them installed on your system and use them as fonts in
your CSS.</p>
Note that you can use multiple encoding formats for those icons, like UTF, or
the name of the icon (for FontAwesome).</p>


VPN
2025-05-14T00:00:00+00:00
From multiple sources, there seem to be too good vpns with decent prices and
freedom to do what you want:</p>

Mullvad</a></li>
AirVPN</a></li>
</ul>
Both boast decent prices, desktop applications to run the VPN and Config
Generator Tools for system configuration (Wireguard or OpenVPN).</p>
I have had trouble with the GUI clients of both of them:</p>

mullvad-vpn</code> app, found in the AUR, seems to disconnect every one and off,
and had multiple javascript error popup</li>
eddie-ui</code> from AirVPN just doesn't launch under archlinux > sway</li>
</ul>
I used the wireguard</code> generated config files. I used
wg-quick up my_conf_interface_name.conf</code> and the same with down to shut it
down.</p>
NOTE: rename the config file to an appropriate name, it will be used to generate
the VPN network interface you need to input in qBittorrent.</p>


Wallpapers
2025-05-14T00:00:00+00:00
Some good wallpaper sources</h1>
Github</h2>
catppuccin-wallpapers - zhichaoh</a></p>
aesthetic-wallpapers - D3Ext</a></p>
nordic-wallpapers - linuxdotexe</a></p>


Linus Custom Timer
2025-05-13T00:00:00+00:00
How to set up a 20-minute computer break</h1>
Every 20 minutes, take a 20-second break and focus your eyes on something at
least 20 feet away.</p>
So let's setup a scheduled notification every twenty minutes of using the
computer, to help follow the 20-20-20 rule to lessen eyestrain.</p>
It uses
systemd timers</a>:</p>

Set up a service unit for the notification</li>
Set up a timer unit to schedule the service</li>
Run commands to enable the timer</li>
</ul>
The service unit:</p>
# ~/.config/systemd/user/twenty.service</span></span>
</span>
[</span>Unit</span>]</span></span>
Description</span>=</span>20-20-20</span> rule</span> notifier</span></span>
</span>
[</span>Service</span>]</span></span>
ExecStart</span>=</span>/usr/bin/notify-send</span> --hint</span> int:transient:1</span> '</span>20-20-20 rule</span>' '</span>Take a break!</span>'</span></span></code></pre>
This is a declarative equivalent of a shell script that contains the command to
run. The important part is the last line, which contains the actual command.
Notice that it has a 'hint', a typed key-value pair setting which will ensure
that the notification doesn't stick around in the desktop environment's
notification history after it disappears from the screen. This hint is
documented here</a>.</p>
Remember to place the file in exactly the directory shown above. You will
probably need to create it first, with:</p>
mkdir</span> -p</span> ~/.config/systemd/user</span></span></code></pre>
Second, the timer unit:</p>
# ~/.config/systemd/user/twenty.timer</span></span>
</span>
[</span>Unit</span>]</span></span>
Description</span>=</span>20-20-20</span> rule</span></span>
</span>
[</span>Timer</span>]</span></span>
OnCalendar</span>=</span>*:0,20,40:00</span></span>
</span>
[</span>Install</span>]</span></span>
WantedBy</span>=</span>timers.target</span></span></code></pre>
This declares a schedule to run the twenty.service</code> unit on (the service name
is implicitly understood from the name of the timer file).</p>
The two significant parts of this timer unit are:</p>

The OnCalendar=</code> property which sets the schedule. This property can
understand various date and time formats, but in this case we are using
hh:mm:ss</code>. The hh</code> component is *</code> meaning run every hour; the mm</code>
component is 0,20,40</code> meaning run on the zeroth, twentieth, and fortieth
minutes; and the ss</code> component is 00</code> meaning run on the zeroth second. This
translates to: run every twenty minutes.</li>
The WantedBy=</code> property which describes when the timer should be enabled</em> on
system boot. In this case it says: enable this timer when enabling all other
timers on boot.</li>
</ul>
Finally, enable the timer immediately without needing to reboot the system:</p>
systemctl</span> daemon-reload</span> --user</span> # Tell systemd about the new units</span></span>
systemctl</span> enable</span> --user --now</span> twenty.timer</span> # Install the timer</span></span></code></pre>
This should set up a symbolic link to the timer, thus installing and enabling
it. Note that we are using the --user</code> flag which means enable this timer for
this user only, not for the entire system (i.e. all users, which would require
admin privileges). The --now</code> flag also immediately activates the timer.</p>
Check that the timer is active:</p>
 systemctl</span> list-timers</span> --user</span></span>
NEXT</span>                        LEFT       LAST                        PASSED      UNIT                         ACTIVATES</span>                     </span></span>
Sun</span> 2020-10-11 20:40:00 EDT 15min left Sun 2020-10-11 20:20:50 EDT 4min 7s ago twenty.timer                 twenty.service</span>                </span></span>
Mon</span> 2020-10-12 10:33:34 EDT 14h left   Sat 2020-10-10 20:20:28 EDT 24h ago     systemd-tmpfiles-clean.timer systemd-tmpfiles-clean.service</span></span>
n/a</span>                         n/a        Sat 2020-10-10 20:17:29 EDT 24h ago     grub-boot-success.timer      grub-boot-success.service</span>     </span></span>
</span>
3</span> timers listed.</span></span>
Pass</span> --all</span> to see loaded but inactive timers, too.</span></span></code></pre>
This listing shows when the timer will run next.</p>
If you need to disable the timer:</p>
systemctl</span> disable</span> --user</span> twenty.timer</span> # Install the timer</span></span></code></pre>
If you need to edit and reload the timer, you can disable, then edit, then
daemon-reload, then enable as shown above.</p>
References</h4>

Yawar Amin (n.d.</em>). How to set up a 20-minute computer break on Linux</em>. How
to set up a 20-minute computer break on Linux - DEV Community, Retrieved April
20, 2024 at 12:57:26 from
https://dev.to/yawaramin/how-to-set-up-a-20-minute-computer-break-on-linux-2kna</a> -
🏛 archive</a></li>
CAO (n.d.</em>). 20-20-20 Rule | Canadian Association of Optometrists</em>. 20-20-20
Rule | Canadian Association of Optometrists, Retrieved April 20, 2024 at
12:57:54 from
https://opto.ca/eye-health-library/20-20-20-rule</a> -
🏛 archive</a></li>
</ul>


Distrobox
2025-04-29T00:00:00+00:00
distrobox.it</a></p>
Installation</h2>
official docs</a></p>
Distrobox is widely available through package managers.</p>
Alternativelyj, you can use the script at:</p>
  curl</span> -s</span> https://raw.githubusercontent.com/89luca89/distrobox/main/install</span> |</span> sudo</span> sh</span></span></code></pre>
You also need to install either podman</code>, lilipod</code>(or docker</strong>).</p>
As of latest update to this page, I have tried lilipod, which worked well with the archlinux image.</p>
Usage</h2>
distrobox</span> create</span> -n</span> your_container_name</span> --image</span> quay.io/toolbx/arch-toolbox:latest</span></span></code></pre>
Usually,  distrobox</code> asks you to pull the image, which you should.</p>
Then to enter the container, run</p>
  distrobox</span> enter your_container_name</span></span></code></pre>
You still have access to all the files while using the container system executables, tools, etc.</p>


Équipement randonnée
2025-04-25T00:00:00+00:00
Rando d'une journée</h2>

moyen de navigation;

boussol</li>
carte</li>
système GPS/tel</li>
carte offline de la zone ou l'on va</li>
</ul>
</li>
vêtement

veste imperméable / coupe-vent</li>
polaire</li>
chemise/tshirt de rando</li>
sous-vêtement thermique</li>
chausettes de rando ?</li>
tour coup</li>
casequette/chapeau</li>
chaussures de rando/sandales</li>
</ul>
</li>
protection solaire

crème solaire</li>
baume à lèvres</li>
lunettes de soleil</li>
</ul>
</li>
éclairage

lampe frontal</li>
batterie de rechange/portable</li>
</ul>
</li>
soins et 1er secours

trousse de premier secours</li>
couverture de survie (feu de camps → alumettes/briquets,allume-feu)</li>
</ul>
</li>
nourriture</li>
nourriture lyophilisée</li>
barres énergétiques</li>
hydratation ( en moyenne 2L par jours, augmente si il faut chaud ou
l'itinéraire est difficile)

poche à eau</li>
gourde</li>
bouteil isotherme</li>
filtre à eau ?</li>
</ul>
</li>
sac entre 15L et 30L</li>
accessoires

mouchoirs jetables</li>
sac à déchets</li>
gel désinfectants</li>
anti-insecte</li>
serviette microfibre</li>
</ul>
</li>
</ul>


Prompts
2025-04-25T00:00:00+00:00
This is a list loos library of prompt templates I use with LLMs.</p>


caps2esc
2025-04-09T00:00:00+00:00
caps2esc</a> is a plugin
for the interception-tools</a>. It
enables a user to capture key presses at a very low level in Linux, and change
them to something else. As the name suggests, caps2esc</code> is used to switch the
capslock</code> key with the esc</code> key.</p>
The escape</code> key is extremely important in the vim/Emacs workflow. The user
needs it to switch out of modes like insert</code>.</p>
I came upon it after wanting to change my EDITOR</code> to helix. I remembered the
Comment arrêter de se servir de vim comme un goret - Nuit de l'Info 2018</a>.</strong></p>
I followed this
page from the Archlinux wiki</a>
after installing intercept-tools</code> and intercept-caps2esc</code>.</p>
→ From the page:</p>
Create the following udevmon job configuration file:</p>
/etc/interception/udevmon.d/caps2esc.yml</code></p>
-</span> JOB</span>:</span> intercept -g $DEVNODE | caps2esc | uinput -d $DEVNODE</span></span>
  DEVICE</span>:</span></span>
    EVENTS</span>:</span></span>
      EV_KEY</span>: [</span>KEY_CAPSLOCK</span>,</span> KEY_ESC</span>]</span></span></code></pre>
By default, this remaps the following:</p>

Escape as Caps Lock.</li>
Caps Lock as Escape when pressed alone and as Ctrl when pressed together with
another key.</li>
</ul>
Next, start/enable udevmon.service</code>.</p>

There are many more configuration options, that can be found in the
Archlinux Intercept Tools</a>
and the Gitlab page (second link from the top</em>).</p>


interesting horizontal layout blog
2025-04-03T00:00:00+00:00
A while back, I stumbled</a> upon this personal website: webb.page</a>.</p>
I found the horizontal layout quite interesting. Unfortunately, I find it also jarring to alternate scrolling directions when making a single (in my case) downwards motion.</p>


Single Link #5: Drum Patterns
2025-03-28T00:00:00+00:00
Here is my entry for this weeks Singlelink</a>:</p>
Drums Patterns</a> is a website that collects and shares drum patterns. You can play user made patterns, or make your own.</p>
I'm not a drummer myself, but visually, the site in interesting to me. Notably, I find the way the drum patterns are displayed quite cool.</p>


Hey look, similar color scheme
2025-03-27T00:00:00+00:00
Hey, look, a similar blog color scheme:</p>
Baldur Bjanrnason</a></p>


Docker
2025-03-25T00:00:00+00:00
Docker on Archlinux for testing</h2>
First install docker</code>, docker-compose</code> and from the AUR
docker-rootless-extras</code></p>
sudo</span> pacman</span> -S</span> docker docker-compose</span>;</span></span>
yay</span> -S</span> docker-rootless-extras</span></span></code></pre>
and then start the docker.socket</code> service with:</p>
system</span> --user</span> enable docker.socket</span></span>
system</span> --user</span> start docker.socket</span></span></code></pre>
To finish, add the following line to your .bashrc</code> or .zshrc</code>:</p>
echo</span> "</span>export DOCKER_HOST=unix://</span>$XDG_RUNTIME_DIR</span>/docker.sock</span>" ></span> .zshrc</span></span>
source</span> .zshrc</span>		#reload shell</span></span></code></pre>
Reference:</p>


https://wiki.archlinux.org/title/Docker</a></p>
</li>

https://wiki.archlinux.org/title/Docker#Rootless_Docker_daemon</a></p>
</li>
</ul>
Docker Compose</h2>
Environment</h2>
This is the section where you specify environment variables globally inside the
container.</p>
An equivalent is export Variable=value</code> in your Unix shell.</p>
From inside a container, you can find out which Env Variables are being set by
using :</p>
printenv</span></span></code></pre>
This works in any shell, be it container or host.</em></p>
!!!!!! Don't forget, it's - VARIABLE_NAME=value</code>, not with :</code>.</p>
Network</h3>
Local Deployement</h4>
You don't have to set networks:</code> for local deployement.</p>
# networks:</span></span>
#   my-external-network:</span></span>
#     external: true</span></span></code></pre>
Docker Compose automatically creates a default network for your services, and all services in the same compose file can communicate with each other using their service names as hostnames. See below</em></p>
You might use networks for:</p>

Multiple isolated networks</li>
Connecting to external networks</li>
Custom network configuration - specific drivers, subnets, etc.</li>
</ul>
Requests from one container to another using the same network</h4>
http://your_container_name:PORT</span></span></code></pre>
Example:</strong></p>
Say you have two services: app</code> and api</code>, the latter using port 8966</code>. You can use http://api:8966 in app</code> to connect to api</code>.</p>
Networking: Access outside host service from inside a container</h4>

network_mode: host</code></strong> option. → however, this defeats the purpose of the
container, which is to be isolated from the host system.</li>
extra_host: - host.docker.internal:docker.gateway</code></strong> → this gives the
outside container IP address. Each container has an outside IP address.</li>
</ol>
To find it, just use:</p>
# Gateway IP address</span></span>
  docker</span> network your_network inspect</span> |</span> grep</span> Gateway</span></span>
# Internal Subnet range, useful for the command below</span></span>
  docker</span> network your_network inspect</span> |</span> grep</span> Subnet</span></span></code></pre>

ufw</code> allow container external IP range with port</strong></li>
</ol>
This one worked! Usually, something like :</p>
sudo</span> ufw allow from 172.31.0.0/16 to any port</span> 56789</span></span></code></pre>
*Replace the IP address with your containers outside IP subnet range, follow the
command outside.</p>
entrypoint</code>, command</code></h2>
Those two run the program/command/script inside the container at the start or
running the container.</p>
Apparently, entrypoint</code> is harder to modify on the fly, while command</code> is more
flexible.</p>
If you need to run something that needs to run endlessly, use:</p>
command</span>: ["</span>/bin/bash</span>", "</span>-c</span>", "</span>/path/to/script/or/binary && tail -f /dev/null</span>"]</span></span></code></pre>
tail</code> enables it main process of the script/binary to exit while keeping the
container alive.</p>
If you get something exited with code 0</code> this means the container stopped. The
command above might help. This is mostly relevant for custom/modified
containers.</em></p>
Devices</h3>
GPU</h4>
For Intel GPU:</p>
# Total GPU capabilities</span></span>
devices:</span></span>
  - /dev/dri:/dev/dri</span></span>
  </span></span>
# Just for computation purposes</span></span>
devices:</span></span>
  - /dev/render128:/dev/render128</span></span></code></pre>Docker Compose commands</h2>
stats</h3>
The stats</code> command displays the system usage stats of the current docker
compose container.</p>
docker</span> compose stats</span></span></code></pre>


Karmic Cycle Quote
2025-03-10T00:00:00+00:00

“You have touched my heart, and I hope you will hear me when I say: You are not stuck, you understand? You can let go of your story. You can escape the karmic cycle. Find peace, in this life. I have hope for you.”</p>
</blockquote>
The White Lotus</em>, Episode 4 of Season 3, 15min 4sec.</p>
This one really hit me.</p>
The way the actress plays out those lines, it makes me feel that something important is happening right there and then. Important both for the characters, and for the story (and that part makes me a bit anxious, not going to lie).</p>
I hope someday I can use that when playing TRPG (Tabletop Role-Playing Games</em>), both as a player or a GM.</p>
Maybe even say that I myself, to someone I care about. I certainly don't feel emotionally mature to say something as deep at this moment of my life.</p>


helix: LSP
2025-03-06T00:00:00+00:00
LSP/FORMATER</h2>
helix</span> --health</span> lang_name</span></span>
</span>
# for example</span></span>
</span>
helix</span> --health</span> markdown</span></span></code></pre>
This flag enables an info dump that shows you what you can set up.</p>
Formaters</h2>
For
the official lists</a>.
I recommend using anything other than JavaScript/TypeScript LSP's and formaters.
Beware high memory usage.</p>
Specific Language</h2>
Spellchecking</h3>
Spellchecking can be done via a multitude of tools, LSP and non-LSP.</p>
Notable mentions:</strong></p>

ltex-ls-plus</a> → uses LanguageTool</li>
Harper</a></li>
typos-lsp</a></li>
Vale.sh</a></li>
</ul>
ltex-ls-plus</strong> has a high memory and CPU usage. It is also a bit slow. But it
handles [FR].</p>
[</span>language-server</span>.</span>ltex-ls-plus</span>.</span>config</span>]</span></span>
ltex</span>.</span>diagnosticSeverity</span> = "</span>warning</span>"</span></span>
ltex</span>.</span>disabledRules</span> = {</span> "</span>en-US</span>"</span> =</span> [</span>"</span>PROFANITY</span>"</span>]</span>,</span> "</span>en-GB</span>"</span> =</span> [</span>"</span>PROFANITY</span>"</span>]</span> }</span></span>
ltex</span>.</span>dictionary</span> = {</span> "</span>en-US</span>"</span> =</span> [</span>"</span>builtin</span>"</span>]</span>,</span> "</span>en-GB</span>"</span> =</span> [</span>"</span>builtin</span>"</span>]</span> }</span></span></code></pre>
There is :</p>

codebook-lsp</a></li>
spellbook</a></li>
</ul>
Markdown</h3>
I used dprint</a> for Markdown combined with markdownlint</code>.
But dprint</code> can do much more.</p>
Jinja</h3>

jinja-lsp</a></li>
</ul>
Tera (rust templating language)</h3>

tree-sitter-tera</a></li>
</ul>
Javascript/Typescript</h3>

Biome</a></li>
</ul>


Linux Networking
2025-03-06T00:00:00+00:00
Programs to manage networking in linux:</h2>
https://wiki.archlinux.org/title/ConnMan</p>

iwd</li>
NetworkManager</li>
</li>
</ul>
Wifi: IWD or wpa_supplicant</h2>
https://www.reddit.com/r/archlinux/comments/eusuif/the_insanity_of_wifi_iwd_wpa_supplicant_iw_ip/
https://wiki.archlinux.org/title/Iwd
https://archive.kernel.org/oldwiki/iwd.wiki.kernel.org/gettingstarted.html
https://wiki.gentoo.org/wiki/Iwd
https://wiki.archlinux.org/title/Iwd
https://wiki.archlinux.org/title/Wpa_supplicant</p>
https://discussion.fedoraproject.org/t/howto-replace-wpa-supplicant-with-iwd-a-newer-wireless-daemon/144111</p>
AP</h2>
https://archive.kernel.org/oldwiki/iwd.wiki.kernel.org/ap_mode.html</p>


Notes on Intro to Psychology
2025-03-06T00:00:00+00:00
Introduction to Psychology - Coursera/Yale</a></p>
Module 1</h1>
Introductory Reading</h2>

History of Psychology By David B. Baker and Heather Sperry University of Akron, The University of Akron</a></li>
Research Designs By Christie Napa Scollon Singapore Management University</a></li>
</ul>
Text 1</h3>
American-centric</p>
Historical period:</p>

For the purposes of this module, we will examine the development of psychology in America and use the mid-19th century as our starting point.</p>
</blockquote>
History (ours) is important to define who we are.</p>
Precursors to psychology:</p>

philosphers (such as John Locke and Thomas Reid) who promoted empiricism</strong> → the idea that all knowledge comes from experience.</li>
</ul>

the role of the human observer and the primacy of the senses in defining how the mind comes to acquire knowledge.</p>
</blockquote>
early 1800s, universities, empiricism is taught courses on mental and moral philosophy. (hehehe morality and mental stuff...)</p>
19th century, philosophical questions on mind & knowledge is matche by physiological investigations → sensory system of humans.</p>
H. von Hemboltz (German) → measures speed of neuron impulse</p>
senses can be deceived + mind is measurable by methods of "science"</p>
psychological reality ø physical reality</p>

philosophical speculation about the nature of mind became subject to the rigors of science.</p>
</blockquote>
psychophysics:</p>

The question of the relationship between the mental (experiences of the senses) and the material (external reality)</p>
</blockquote>
The formal development of modern psychology = invention of experimental psychology.</p>
German physician, physiologist, and philosopher Wilhelm Wundt :</p>

Students were trained to offer detailed self-reports of their reactions to various stimuli, a procedure known as introspection.</p>
The goal was to identify the elements of consciousness.</p>
</blockquote>
Studied:</p>

Sensation and perception</li>
Mental chronometry, more commonly known as reaction time.</li>
</ul>
Wundt → America via Edward Bradford Titchener: structuralism</strong>: describe the elements of concious experience</p>

For Titchener, the general adult mind was the proper focus for the new psychology, and he excluded from study those with mental deficiencies, children, and animals</p>
</blockquote>
American Psychological Association (APA) in 1892</p>
Titchner first student → women named Margaret Floy Washburn, who became first woman to earn a PhD and 2st woman president of APA.</p>

Striking a balance between the science and practice of psychology continues to this day.</p>
</blockquote>
William James → functionalism : the utility of focus</p>
influeced by Darwin's theory of evolution.</p>
opened the way to animal an comparative psychology.</p>
W. James wrote: Principles of Psychology</em>:</p>

James proposed that consciousness is ongoing and continuous; it cannot be isolated and reduced to elements. For James, consciousness helped us adapt to our environment in such ways as allowing us to make choices and have personal responsibility over those choices.</p>
</blockquote>
G. Stanley Hall:</p>

created the first journal of psychology in America, American Journal of Psychology.</li>
researched the process of adaptation and human development.</li>
mentored first black man who got a PhD in psychology, Francis Cecil Sumner, who become to be commonly known as the “Father of Black Psychology„</li>
</ul>
James McKeen Cattell:</p>

student of Wundt</li>
researched assessment of individual differences → Ways in which people differ in terms of their behavior, emotion, cognition, and development.</li>
promoted eugenics ... outch</li>
</ul>

What about Phenotype's enviromental parameters ? When was that observed ?</p>
Also, what about determinism</strong> and constructivism</strong> ?</p>
</aside>

believed that mental abilities such as intelligence were inherited and could be measured using mental tests.</li>
</ul>
The Growth of Psychology</h4>
Gestalt Psychology</strong>
Began in Germany.</p>
Max Wertheimer</p>
In opposition to Wundt's laboratory approach.</p>

[They] believed that studying the whole of any experience was richer than studying individual aspects of that experience. The saying “the whole is greater than the sum of its parts” is a Gestalt perspective.</p>
</blockquote>
Hypothesis: the mind processes information simultaneously rather than sequentially.
Using Gestalt Principles, they explored the nature of learning and thinking.</p>
Most early academics of this school of thought were Jewish, and fled Germany.</p>
In America, Gestalt Psychology served as a precursor to cognitive Psychology (The study of mental processes</em>).</p>
Behaviorism</strong> early 20th John B. Watson (1878–1958) and B. F. Skinner (1904–1990)</p>
Behaviorism ≠ mind → overt and observable behavior as proper subject.
→ hope to derive laws of learning → promote prediction + behavior control
Ivan Pavlov influenced the early behaviorism in America → work on conditioned learning</strong> = learning + behavior = conditioned by environment and events outside of mind/consciousness.</p>
Behaviorism dominated the American Psychology before 60s, until recognition of it's inability to fully explain human behavior, neglecting mental processes.</p>
But turn to cognitive psy. was not new :</p>

30s → British psychologist Frederic C. Bartlett (1886–1969): the constructive mind</strong> → people use past experiences to construct frameworks to understand new experiences.</li>
</ul>
major pioneers in American cognitive psychology:
Jerome Bruner (1915–), Roger Brown (1925–1997), and George Miller (1920–2012).</p>

Bruner:

cognitive aspects of sensation and perception.</li>
flashbulb memory : high detail vivid memory of emotionally significant event.</li>
tip-of-the-tongue phenomenon</li>
</ul>
</li>
Miller:

working memory</li>
The Magic Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information</em>, in 1956, one of most highly cited papers in Psychology.</li>
popular interpretation: number of bits of information an average human can hold in memory is 7 ± 2.</li>
</ul>

</li>
</ul>
That is a bit cybernetic stuff. This also seems overly simplistic.
I wonder how well this held out overtime.</p>
</aside>

same time as computer science was growing + used as analogy to mind.</li>
</ul>
Applied Psychology in America</h4>
America → !! application of psychology to everyday life.
!!! Mental testing (?)</p>
  

How very utilitarian.</p>
</aside>
Intelligence testing → Alfred Binet →</p>

goal was to develop a test that would identify schoolchildren in need of educational support.</p>
</blockquote>
Introduced by Henry Goddard and standardized by Lewis Terman at Stanford Uni.</p>
HUGE debate on nature/nurture</strong> → the relative contributions of heredity and environment in determining intelligence</p>
employee selection, eyewitness testimony, and psychotherapy
psychology of advertising and marketing
industrial psychology and engineering psychology
time and motion studies to improve efficiency in industry
psychology of efficiency
Clinical psychology</p>
WWII = necessity for professional psychologists</p>
scientist-practitioner model = both researcher and clinical study.
scholar-practitioner = emphasis on clinical training. more common</strong></p>
Psychology and Society</h4>
SPSSI (Society for the Psychological Study of Social Issues</em>)</p>
Helen Thompson Woolley → examined the assumption that women were overemotional compared to men and found that emotion did not influence women’s decisions any more than it did men’s.</p>
Leta S. Hollingworth → found that menstruation did not negatively impact women’s cognitive or motor abilities.</p>
Mamie Phipps Clark (1917–1983) and her husband Kenneth Clark → studied the psychology of race and demonstrated the ways in which school segregation negatively impacted the self-esteem of African American children.</p>
Brown v. Board of Education, which ended school segregation</p>
Association of Black Psychologists (ABPsi)</p>
Evelyn Hooker → “The Adjustment of the Male Overt Homosexual,” → No significant differences in psychological adjustment between homosexual and heterosexual men → helped to de-pathologize homosexuality → contributed to the decision by the American Psychiatric Association to remove homosexuality from the Diagnostic and Statistical Manual of Mental Disorders.</p>
 

What about homosexual women?</p>
</aside>
Text 2 → Research Designs</h3>


RPG GMing
2025-03-06T00:00:00+00:00
Types of questions for players, and when to use them</h2>

Closed</strong> = yes or no</li>
Open</strong> = generates more detail</li>
Reading</strong> = emphasis one solution</li>
Inverted</strong> = outcome is set, lots of creative limitation there.</li>
Redirect</strong> = motivate inter-player interactions</li>
Affective</strong> = character internal monologue, feelings and emotions. Deep & rich inner life.</li>
</ul>
When to use the questions, based on outcome/information required</h3>
</p>

The six question types before neatly fit into the three functional categories. If you want to:</p>

Clarify or learn about a character or the world then you can ask open questions to elicit detail or closed questions to clarify established facts.</li>
Tempt players or incite quick action then you can ask leading questions to prime them for a tense situation or an inverted questions to have them roleplay or explain what happened.</li>
Learn about a character’s state or adjust the pace then you can ask affective questions to learn about a character’s emotions and thoughts on the scene or redirect questions to gently check-in with a player or have them comment on the actions of another character.</li>
</ul>
</blockquote>
Source</strong>: How to improve player engagement in roleplaying games using questions -- The Bardic Inquiry</a></p>


Good & Bad Product Design
2025-03-05T00:00:00+00:00

Good designs expose systematic structure; they lean on their users’ ability to understand this structure and apply it to new situations. We were born for this.</p>
Bad designs paper over the structure with superficial labels that hide the underlying system, inhibiting their users’ ability to actually build a clear model in their heads.</p>
</blockquote>
from Andrea diSessa's Changing Minds</a></p>
via [Avoid the nightmare bicycle - Geoffrey Litt]</p>


Organisations
2025-03-05T00:00:00+00:00
A list of orgs of interest.</p>


test
2025-03-05T00:00:00+00:00



Design Tools
2025-03-04T00:00:00+00:00
-hue.tools</a> a tool for colors. Better then the W3school one.</p>


Firefox
2025-03-04T00:00:00+00:00
Usage</h2>
Keybindings</h3>
Keywords</h3>


*</code> → Search bookmarks</p>
</li>

%</code> → Search open tabs</p>
</li>

^</code> → Search history</p>
</li>

about:addons</code></p>
</li>

about:config</code></p>
</li>

about:preferences</code></p>
</li>

about:profiles</code></p>
</li>

about:profiling</code></p>
</li>

about:processes</code></p>
</li>
</ul>
Configuration</h2>
Run firefox with prime-run on nvidia+intel system</h3>
My blog post</a> about that procedure and
how it solved some crashes of firefox.</p>
Memory usage</h3>
about:memory</code> and about:processes</code></p>
https://robots.net/tech/how-to-limit-firefox-ram-usage/
https://gbtimes.com/how-to-make-firefox-use-less-memory/
https://superuser.com/questions/1740578/how-to-reduce-the-firefox-memory-usage-as-much-as-possible
https://support.mozilla.org/en-US/kb/firefox-uses-too-much-memory-or-cpu-resources
https://support.mozilla.org/en-US/kb/firefox-uses-too-much-memory-or-cpu-resources#w_firefox-tools
https://support.mozilla.org/en-US/kb/task-manager-tabs-or-extensions-are-slowing-firefox</p>
How do I find out which extensions take a lot of ram.</p>
How do I run AutoDiscardTab more agressively ? !!!!</h2>
The about:</code> keyword in the searchbar enables access to divers tools and
configuration pages. There is autocomplete on, so if you can't find a page, it
either doesn't exist, or you are misspelling the words.</p>
Unloading Tabs</h2>
about:unloads</code></p>

It displays information about your open tabs, such as the order of priority
for unloading tabs, the last time you accessed each tab and the memory each
tab is consuming.</p>
</blockquote>
Source</strong>:</p>

Unload Inactive Tabs - Mozilla</a></li>
</ul>
Sidebar Tabs</h3>

about:config</code></li>
type sidebar</code></li>
Scroll down until you see sidebar.verticalTabs</code></li>
Set sidebar.verticalTabs</code>, sidebar.revamp.round-content-area</code> and
sidebar.revamp</code> to true</code>.</li>
</ol>
Fix gesture swiping tabs not working correctly on linux (laggy, jagged slow
movement):</strong></p>

same as above</li>
apz.gtk.pangesture.delta_mode</code> set to 2</code></li>
</ol>


Programming Languages
2025-03-04T00:00:00+00:00
Zig vs Rust</h2>

Zig or Rust - Reddit.com</a></li>
Zig and Rust - Matklad</a></li>
</ul>


Notes on Jacob Geller's Video Essay Method
2025-03-03T00:00:00+00:00
Here are my notes on Jacob Geller's method to create a video essay, as told in this presentation on Nebula</a></p>
Ambient Research</h2>


Ambient research</strong>: what you absorb from media you consume. It's like picoré everywhere. When you find the idea, you have a encyclopedia in your head from which you can pull links to the subject.</p>
</li>

Read people that write better than you, to improve your own writing.</p>
</li>
</ul>
Finding Ideas</h2>
Start from a detail, not a whole. This detail is the spark.</p>

Why that moment triggered a reaction in me ?</p>
</blockquote>
For essays, also think about the question your essay is posing.</p>

Because this is what brings people in.</strong></p>
</blockquote>

I have a rule for myself which is, if I feel like I already have the answer to whatever question that I have posed, I as a viewer would not be interested in it, because I would think «I already know where this is going». I have to assume that</p>
</blockquote>

There should be some ambiguity about why it does make me feel the way that I do. Investigating that is hopefully where I find the rest of the essay.</p>
</blockquote>
Already at the beginning you should be formulating a title ? That is the core of the question you are asking.</p>
Research (true)</h2>
University skill → use research paper's bibliography as a map to where I want to go with the essay.</p>
Outlines and Intros</h2>
Outlines</h3>
Useful as a writing tool to scope out the breadth of your essay. It is less of a strict guideline. See it as a creative limitation which is good.</p>
Going on tangents is the enemy of good pacing.</p>
Intro</h3>
Here Jacob mentions the attention span metric for YouTube Video as an indicator of if the intro is good or not.</p>

I wonder what good examples I can find in different media and ways to tell a story, of telling a smaller story that leads to the overarching one.</p>
</aside>
Tell a smaller story that ends at the start of the larger essay. It's a way to lead in the viewer/reader/player etc.</p>

Pull the viewer in with a short anecdote that proves I know how to tell a story [...] but it does not require them to buy into the entire premise of the essay.</p>
</blockquote>
You need them to trust you, by giving them something right away.</p>
Writing</h2>


Your own voice, who you are speaking as ? personal perspective or not ? For whome do you speak to ?</p>
</li>

Are you writing about something or about you experiencing that something ? Meaning, are you a part of the story ?</strong></p>
</li>
</ul>
Two perspectives are mentioned here:</p>


2000 m view, pure facts</p>
</li>

your process getting to the 2000m view. How hard was it to find the information ?</p>
</li>

who you are writing for ? do the people understand (technical writing) or not (layman) what you are writing about ? What's the level of understanding ?</p>
</li>
</ul>
Jacob:</p>

[...] What I want to do generally is let the hardcore committed fans know that I know what depth there is, I know all the nuances that are potentially there on the topic, but try to talk about it without using the terminology that can become so obvious when you are in that community.</p>
</blockquote>
The danger of excluding less knowledgable people. Sometimes you need to assume people are starting at level 0.</p>
Think about:</p>

pacing (super important and the hardest to nail in video essays for Jacob)</li>
tone</li>
how it is going to be delivered ? This question is very relevant when it is not just text. Jacob mentions for his videos, the case where is he is filmed entirely, or when he is a talking head, or even a unseen narrator. All those cases and more are written differently.</li>
how emotional you are able to get with your delivery ? Pivots to certain emotions ? This comes down to the viewer considering you as a person or not.</li>
</ul>

My essays often end up being about their conclusions.</p>
</blockquote>
Use the conclusion to reframe the information stated in the essay, in a new perspective or context.</p>

Make the end statement declarative in some way. [...] How do you want your essay to live on in your audience's memory.</p>
</blockquote>
Reward those that read to the end.</p>
Polishing</h2>
Whever or not you plan on reading out loud your essay for the final delivery, Jacob states that you should try reading it out loud multiple times.</p>
If you get out of breath in the middle of a sentence, then you should go back to it and improve it by cutting up, shorten it, transform it so that it is more fluid.</p>
The polishing stage is also a good time to check if you have gone out of scope, if your idea got too big!</p>
If you get bored in a section of your own sections/paragraphes, etc, then it has to change or go. You are your best (and worst) critic.</p>

I'm starting to wonder if all of this could also be applied to RPG storytelling.</p>
</aside>
You want your essay to pull you towards you next point.</p>
Take a couple of days away from your essay after you have written it and return to it with new eyes.</p>
Final stage for Jacob, what music is going to accompany the piece.</p>

Music for me, is an incredibly important tool both telling the audience how to feel and, once again, figuring out how the essay is paced. Audiences can't see paragraphs when you are reading them, and so that's how I use music: I will use it to seperat different ideas.</p>
</blockquote>

Words tell people what to think, music tells them what to feel.</p>
</blockquote>

Finding the music is [...] helpful for me [to find out if I'm conveying the correct emotion].</p>
</blockquote>
Preparing for Video</h2>
1st Challenge: Speaking out loud</strong></p>
It took Jacob year's to learn to talk in a microphone.
Most helpful examples are from the world of radio.</p>
Your voice can be a tool to make you found less monotonous.
Jacob finds himself sometimes exagerating his vocal tones, saying some things faster, some slower, some things more or less emotional.</p>

It's a way to remind the audience that you are not a robot and the things they are watching are thoughts coming out of someone's head.</p>
</blockquote>
He records his voice while listening to the music that is going to play. He makes the full track with each music piece first. Then, while listening to this first music track, he records himself. The first track informs him of the emotions to convey.</p>
For recording the images (in video games and whatnot), he uses the script as a shot list. When he describes something, he knows then to go and record that in the video game.</p>
alternatively, ask someone who has record a let's play. They are sometimes very gracious and let you take their footage.</p>
Video editing</h2>
Jacob: it's as hard as you want to make it. Then two examples of video editing timelines:</p>

Hideo Kojima (very complexe and LAYERED)</li>
Jacob Geller (2 audio tracks, 3 video tracks.)</li>
</ul>
Trick for editing: video should follow speech patterns. You finish a sentence, you finish a video clip at the same time, and change clips for a new sentence ?</p>
Speaking rythm first, video second in terms of priority. Slow down the footage, take another longer footage if need be.</p>
IMPORTANT: When words come on screen, we automatically start reading them. Be mindful your use of subtitles etc.</p>
Sound quality is also much more important than video quality !!
Music louder than voice, big NO NO</em></p>
The fundamental question you want to answer with editing is simply do you want the audience to notice it</p>

The fundamental question you want to answer with editing is simply do you want the audience to notice it. Do you want them to think of it as an edited video or not. [...] Remember that you are telling a story and whatever you do, it should further this goal.</p>
</blockquote>

Thumbnail + title work together to tell a story</p>
</blockquote>


Terrific CGI
2025-03-03T00:00:00+00:00

AmbientCG</a></li>
Polyhaven</a></li>
</ul>
Blender Addons</h2>

Baga Pie | Architecture & Vegetation, Rocks, Trees</a></li>
</ul>


CSS
2025-02-28T00:00:00+00:00
Cleanup CSS</h2>

PurgeCSS - Remove unused CSS PurgeCSS</a></li>
leeoniyadropcss An exceptionally fast, thorough and tiny unused-CSS cleaner</a></li>
uncss Remove unused styles from CSS</a></li>
</ul>
specific stuff</h2>

You can use text-wrap: balance; on icons</a></li>
</ul>
Text Alignment</h2>

vertical-align</code></a></li>
</ul>
Animations</h2>

view transitions</a>, via Terence Eden</a>

chrome dev demo</a></li>
Old JQuery version</a></li>
</ul>
</li>
</ul>


SingleLink #4
2025-02-27T00:00:00+00:00
Hi there.</p>
Today's single link is :</p>
Kagi Small Web</a></h1>
It's an aggregator that serves a random page from the small</a> web</a>.</p>
It enabled me to discover nice handwritten human curated information and content. While I'm unsure I would recommend the search engine Kagi (a paid product), I find this initiative nice.</p>
There at least two others like it that I know of:</p>

marginalia search</a></li>
wiby.org</a></li>
</ul>

I'm thinking of making the singlelink</strong> into a link</a>blog</a>.</p>


GLSL
2025-02-27T00:00:00+00:00

Learn Shader Programming with Rick and Morty</a></li>
</ul>


git
2025-02-27T00:00:00+00:00
Resources</h2>

Oh Shit, Git!?!</a> (&
Dangit, Git!?!</a> for those that don't like swear
words.)</li>
multiple-repository-and-identities-git-configuration.md</a></li>
</ul>
Learning git</code></h3>

Pro Git</a></li>
Git for Beginners: Zero to Hero 🐙</a></li>
GitFlow considered harmful | End of Line Blog</a></li>
OneFlow – a Git branching model and workflow | End of Line Blog</a></li>
branch - How do I fetch all Git branches? - Stack Overflow</a></li>
</ul>
git checkout</code></h4>

stackoverflow</a></li>
</ul>
Setup upstream and merge with fork/main</h3>
When you make a fork, you might want to track and merge commits from the
upstream repo into your fork's main branch. To do so:</p>
  # add the upstream/main as remote upstream</span></span>
  git</span> add remote upstream git@the_repo</span></span>
</span>
  # fetch upstream</span></span>
  git</span> fetch upstream</span></span>
</span>
  # merge upstream into main</span></span>
  git</span> merge upstream/main</span></span>
</span>
  # resolve conflicts</span></span>
  # then push to your own origin/main</span></span>
  git</span> push origin main</span></span></code></pre>Rebase and squash</h3>
If you are using a forked repo, don't forget to update to latest upstream</a></p>
# After updating your main branch</span></span>
# use the interactive rebase command</span></span>
git</span> rebase</span> -i</span> main</span></span></code></pre>
This opens a todo file for the rebase in progress, in your EDITOR</code>. You then
have to pick</code> the first commit into which you want to squash the other commits.</p>
add</span> af32ff1b12a1096f</span> #your commit message</span></span>
add</span> 88d6c9bb26ffb381</span> #your second commit message</span></span>
# ...</span></span></code></pre>
The file usually has instructions at the bottom on keywords and the associated
action.</p>
You can pick</code> the commit into which you will squash</code> the other commits.</p>
pick</span> af32ff1b12a1096f</span> #your commit message</span></span>
squash</span> 88d6c9bb26ffb381</span> #your second commit message</span></span>
# ...</span></span></code></pre>
Commit messages are then added in order, and can be interactively</p>
Commits can be renamed with edit</code> or after the rebase has been finishe, with:</p>
git</span> commit</span> --amend</span></span></code></pre>branching and merging - How to change current branch in git from master to main - Stack Overflow</h3>


rename your local branch:</p>
git</span> branch</span> -m</span> master main</span></span></code></pre></li>

change the tracked branch</p>
git</span> fetch</span> -p</span> origin</span></span>
git</span> branch</span> -u</span> origin/master main</span></span></code></pre></li>

change the main local branch</p>
git</span> remote set-head origin</span> -a</span></span></code></pre></li>

optionally, remove the master branch, local and remotely:</p>
git</span> branch</span> -D</span> master</span></span>
git</span> push origin :master</span></span></code></pre></li>
</ol>
Drop commit</a></h2>
Careful:</strong> git reset --hard</code> WILL DELETE YOUR WORKING DIRECTORY CHANGES</strong>.
Be sure to stash any local changes you want to keep</strong> before running this
command.</p>
Assuming you are sitting on that commit, then this command will wack it...</p>
git</span> reset</span> --hard</span> HEAD~1</span></span></code></pre>
The HEAD~1</code> means the commit before head.</p>
Or, you could look at the output of git log</code>, find the commit id of the commit
you want to back up to, and then do this:</p>
git</span> reset</span> --hard</span> <</span>sha1-commit-i</span>d</span>></span></span></code></pre>

If you already pushed it, you will need to do a force push to get rid of it...</p>
git</span> push origin HEAD</span> --force</span></span></code></pre>
However</strong>, if others may have pulled it, then you would be better off starting
a new branch. Because when they pull, it will just merge it into their work, and
you will get it pushed back up again.</p>
If you already pushed, it may be better to use git revert</code>, to create a "mirror
image" commit that will undo the changes. However, both commits will be in the
log.</p>

FYI: git reset --hard HEAD</code> is great if you want to get rid of WORK IN
PROGRESS.It will reset you back to the most recent commit, and erase all the
changes in your working tree and index.</p>
git stash</code> does the same except you can restore it later if you need</strong>,
versus permanently delete with reset hard mode. Check your stashes by using
git stash list</code> and git stash show 'stash@123'</code></p>

Lastly, if you need to find a commit that you "deleted", it is typically present
in git reflog</code> unless you have garbage collected your repository.</p>

Git Branch Management and History Cleanup Guide</h3>
Overview</h3>
This guide covers common Git workflows for managing branches, cleaning up commit
history, and preparing branches for pull requests. These techniques help
maintain clean, organized version control history.</p>
Separating Mixed Commits into Different Branches</h3>
When you have a single branch containing commits for multiple problems that
should be separated:</p>
Method 1: Interactive Rebase (Recommended)</h4>
# Create new branches from your mixed branch</span></span>
git</span> checkout your-mixed-branch</span></span>
git</span> checkout</span> -b</span> problem-1-branch</span></span>
git</span> checkout</span> -b</span> problem-2-branch</span></span>
</span>
# Use interactive rebase to filter commits for each branch</span></span>
git</span> checkout problem-1-branch</span></span>
git</span> rebase</span> -i</span> HEAD~n</span>  # where n is the number of commits to review</span></span>
# Delete lines for commits that belong to problem 2</span></span>
</span>
git</span> checkout problem-2-branch</span></span>
git</span> rebase</span> -i</span> HEAD~n</span></span>
# Delete lines for commits that belong to problem 1</span></span></code></pre>Method 2: Cherry-picking</h4>
# Create clean branches from before mixed work</span></span>
git</span> checkout</span> <</span>commit-before-mixed-wor</span>k</span>></span></span>
git</span> checkout</span> -b</span> problem-1-branch</span></span>
git</span> checkout</span> -b</span> problem-2-branch</span></span>
</span>
# Cherry-pick relevant commits to each branch</span></span>
git</span> checkout problem-1-branch</span></span>
git</span> cherry-pick</span> <</span>commit-hash-</span>1> <</span>commit-hash-</span>3></span></span>
</span>
git</span> checkout problem-2-branch</span></span>
git</span> cherry-pick</span> <</span>commit-hash-</span>2> <</span>commit-hash-</span>4></span></span></code></pre>Resetting Master Branch to Origin</h3>
When your local master branch has unwanted commits:</p>
Basic Reset Process</h4>
# Backup current state (safety first)</span></span>
git</span> checkout master</span></span>
git</span> branch backup-master</span></span>
</span>
# Fetch latest from origin</span></span>
git</span> fetch origin</span></span>
</span>
# Reset to match origin/master</span></span>
git</span> checkout master</span></span>
git</span> reset</span> --hard</span> origin/master</span></span>
</span>
# Verify the reset</span></span>
git</span> status</span></span>
git</span> log</span> --oneline -10</span></span></code></pre>If Origin Also Has Bad Commits</h4>
# Find the last good commit</span></span>
git</span> log</span> --oneline</span></span>
</span>
# Reset to that specific commit</span></span>
git</span> reset</span> --hard</span> <</span>good-commit-has</span>h</span>></span></span>
</span>
# Force push to update origin</span></span>
git</span> push origin master</span> --force-with-lease</span></span></code></pre>Selective Commit Squashing</h3>
Use interactive rebase to selectively squash commits while ignoring others:</p>
# Start interactive rebase</span></span>
git</span> rebase</span> -i</span> HEAD~n</span>  # or git rebase -i <commit-hash>^</span></span></code></pre>Available Commands in Interactive Rebase</h4>

pick</code></strong> - Keep the commit as-is (ignore/don't touch)</li>
squash</code></strong> or s</code></strong> - Squash this commit into the previous one</li>
fixup</code></strong> or f</code></strong> - Like squash, but discard this commit's message</li>
drop</code></strong> or d</code></strong> - Completely remove this commit</li>
edit</code></strong> - Pause to modify this commit</li>
</ul>
Example: Selective Squashing</h4>
pick abc1234 Commit message 1      # keep as-is</span></span>
pick def5678 Commit message 2      # this will be the base</span></span>
squash ghi9012 Commit message 3    # squash into commit 2</span></span>
pick jkl3456 Commit message 4      # keep as-is</span></span>
pick mno7890 Commit message 5      # keep as-is</span></span></code></pre>Alternative: Cherry-pick with Squashing</h4>
# Create new branch from clean point</span></span>
git</span> checkout</span> -b</span> cleaned-branch</span> <</span>base-commi</span>t</span>></span></span>
</span>
# Cherry-pick and combine commits</span></span>
git</span> cherry-pick</span> <</span>commit</span>1></span></span>
git</span> cherry-pick</span> --no-commit</span> <</span>commit</span>2></span>  # stages without committing</span></span>
git</span> cherry-pick</span> --no-commit</span> <</span>commit</span>3></span>  # stages without committing</span></span>
git</span> commit</span> -m</span> "</span>Combined commit message</span>"</span></span>
</span>
# Cherry-pick other commits normally</span></span>
git</span> cherry-pick</span> <</span>commit</span>4></span></span></code></pre>Rebasing Branch on Updated Master</h3>
Before creating a pull request, rebase your feature branch on the latest master:</p>
Standard Rebase Workflow</h4>
# Update master</span></span>
git</span> checkout master</span></span>
git</span> pull origin master</span></span>
</span>
# Switch to feature branch and rebase</span></span>
git</span> checkout your-feature-branch</span></span>
git</span> rebase master</span></span>
</span>
# Force push the rebased branch</span></span>
git</span> push origin your-feature-branch</span> --force-with-lease</span></span></code></pre>Handling Rebase Conflicts</h4>
# When conflicts occur during rebase:</span></span>
# 1. Fix conflicts in your editor</span></span>
# 2. Stage the resolved files</span></span>
git</span> add .</span></span>
</span>
# 3. Continue the rebase</span></span>
git</span> rebase</span> --continue</span></span>
</span>
# If you need to abort the rebase</span></span>
git</span> rebase</span> --abort</span></span></code></pre>Interactive Rebase for Cleanup</h4>
# Rebase with ability to modify commits</span></span>
git</span> rebase</span> -i</span> master</span></span></code></pre>Best Practices</h3>

Always backup before destructive operations</strong>: Create backup branches
before using reset --hard</code> or force pushing</li>
Use --force-with-lease</code> instead of --force</code></strong>: Safer when force pushing
as it won't overwrite others' changes</li>
Keep commits focused</strong>: Each commit should represent a single logical
change</li>
Rebase before creating PRs</strong>: Ensures clean, linear history on top of
latest master</li>
Test after rebasing</strong>: Verify that your changes still work correctly after
rebase operations</li>
</ol>
Safety Notes</h3>

Interactive rebase and force pushing rewrite Git history</li>
Always coordinate with team members before force pushing to shared branches</li>
Use backup branches when performing complex history modifications</li>
--force-with-lease</code> is safer than --force</code> for force pushing</li>
</ul>

Git Pull: Skipped Cherry-Pick Commits</h2>
Issue Description</h3>
When executing git pull</code>, users may encounter a warning message about skipped
commits that were previously applied through cherry-picking operations.</p>
Example Output</h3>
git pull</span></span>
warning: skipped previously applied commit 3a5eaa6</span></span>
hint: use --reapply-cherry-picks to include skipped commits</span></span>
hint: Disable this message with "git config set advice.skippedCherryPicks false"</span></span>
Successfully rebased and updated refs/heads/fix_url_parsing.</span></span></code></pre>Explanation</h3>
Warning Message</h4>

"warning: skipped previously applied commit [hash]"</strong>: Git detected a commit
that had already been applied to the current branch and automatically skipped
it to prevent duplication</li>
This commonly occurs when commits have been cherry-picked from other branches</li>
</ul>
Hints Provided</h4>

--reapply-cherry-picks</code> flag</strong>: Optional parameter to force reapplication
of skipped commits (rarely needed)</li>
Configuration option</strong>: git config set advice.skippedCherryPicks false</code> to
disable this specific warning message</li>
</ul>
Success Confirmation</h4>

"Successfully rebased and updated refs/heads/[branch-name]"</strong>: Indicates the
pull operation completed successfully using a rebase strategy</li>
The branch has been updated with the latest changes from the remote repository</li>
</ul>
Resolution</h3>
This warning is normal behavior</strong> and indicates successful operation. No
action is required unless you specifically need to reapply the skipped commits.</p>
Related Commands</h3>

git pull --reapply-cherry-picks</code>: Force reapplication of previously
cherry-picked commits</li>
git config set advice.skippedCherryPicks false</code>: Disable the warning message
permanently.</li>
</ul>
git diff</h2>
Diff latest vs last commit</h3>

As the previous commit can be accessed with HEAD^, I think that you are
looking for something like:</p>
git</span> diff HEAD^ HEAD</span></span></code></pre>
That also can be applied for a :commithash</p>
git</span> diff</span> $commithash</span>^</span> $commithash</span></span></code></pre>
As of Git 1.8.5, @</code> is an alias for HEAD</code>, so you can use:</p>
git</span> diff @~..@</span></span></code></pre>
The following will also work:</p>
git</span> show</span></span></code></pre>
If you want to know the diff between head and any commit you can use:</p>
git</span> diff commit_id HEAD</span></span></code></pre>
And this will launch your visual diff tool (if configured):</p>
git</span> difftool HEAD^ HEAD</span></span></code></pre>
Since comparison to HEAD is default you can omit it (as pointed out by
Orient</a>):</p>
git</span> diff @^</span></span>
git</span> diff HEAD^</span></span>
git</span> diff commit_id</span></span></code></pre>Warnings</h2>

@ScottF and @Panzercrisis explain in the comments that on Windows the ~</code>
character must be used instead of ^</code>.</li>
</ul>
</blockquote>
Source:
https://stackoverflow.com/questions/9903541/finding-diff-between-current-and-last-version</a></p>

.gitignore exclude specific file</h3>
First, set that file's path in the .gitignore</code>:</p>
path/to/file.txt</span></span></code></pre>
Then exclude the file from the repo.</p>

If you want to to keep that file locally, use  git rm --cached path/to/file.txt</span></span></code></pre></li>
Else just use the regular git rm</code> cmd.</li>
</ul>
Source: StackOverflow</a></p>


Markup Systems
2025-02-25T00:00:00+00:00
Scribble</h2>

Scribble is a collection of tools for creating prose documents—papers, books, library documentation, etc.—in HTML or PDF (via Latex) form. More generally, Scribble helps you write programs that are rich in textual content, whether the content is prose to be typeset or any other form of text to be generated programmatically.</p>
</blockquote>
It is a markup language integrated or at least adjacent to Racket, a lisp dialect.</p>

Scribble</a></li>
</ul>
Nota</h2>

Nota is a language for writing documents, like academic papers and blog posts. The goal of Nota is to bring documents into the 21st century.</p>
</blockquote>

Nota</a></li>
</ul>

Bot Nota</strong> and Bene</strong> are authored by Will P.L. Crichton</a></p>
</aside>
Bene</h2>

Bene is a reading system for documents written in the EPUB file format.</p>
</blockquote>

Bene - Github</a></li>
Portable-Epubs</a></li>
</ul>
Typst</h2>

a new open-source typesetting software that aims to effectively replace LaTeX in its niche and also bring a modern, free tool to a broader set of use cases.</p>
</blockquote>

Typst.app</a></li>
Typst - Github</a></li>
</ul>
What people are saying about Typst</h3>

Using Pandoc and Typst to Produce PDFs - imaginarytext.ca</a></li>
More on Typst - imaginarytext.ca</a></li>
</ul>


fclones
2025-02-24T00:00:00+00:00

Github</a></li>
</ul>
Installation</h2>
  sudo</span> pacman</span> -S</span> fclones</span></span></code></pre>Usage</h2>
Find duplicates</h3>
  fclones</span> group .</span> ></span> dupes.txt</span></span></code></pre>Remove duplicates</h3>
Don't forget to review the dupes.txt</code> BEFORE making any changes.</p>
  fclones</span> remove</span> <</span> dupes.txt</span></span></code></pre>


NOTES while going through DCIC
2025-02-24T00:00:00+00:00
A Data-Centric Introduction to Computing</a></strong></p>
1. → 2. Intro</h1>
Values</h2>

Software is made to run and be maintenained and read by others.</li>
Programmers are responsible for their software.</li>
Programms should be predictable in what they do.</li>
</ol>
Performance, reliability and predictability</strong></p>
Perspective on data</h1>

data-driven programming (same as data-oriented game dev --> Nic Barker</a>?)</li>
real data makes more sense for learners</li>
data science is limited by data, and computer science start after thos limitations</li>
</ul>

In particular, the structure of data serve as a point of departure for thinking about and achieving some of the values above—performance, reliability, and predictability—using the many tools of computer science.</p>
</blockquote>
data-centric curriculum</strong> as :</p>

data centric = data science + data structures</p>
</blockquote>
In the order of :</p>

ideas from data science</li>
classical ideas from data structures</li>
the rest from computer science</li>
</ol>
notional machines</strong>: abstractions of program behavior meant to help students understand how programs work</p>

[...] programs are not only objects that run, but also objects that we reason about.</p>
</blockquote>
socially-responsible computing</p>
3.1 Getting Started</h1>
flag observations:</p>

they are horizontal rectangles</li>
there are between 2 and 5 colors</li>
there are strips, circles and rarely, more complexe images.</li>
sometimes there is only one type of decoration, some have both.</li>
</ul>
questions about flags and how to make them:</p>

how does I draw them on screen ?</li>
how can I store the data for easy replication</li>
are there sometimes other shapes ? Why is it just rectangles ?</li>
</ul>
questions i MIGHT have about the flag problem :</p>

Do I need to be able to draw these images by hand?</p>
Will we be able to generate different sized flags from the same code?</p>
What if we have a non-rectangular flag?</p>
We might want to compute the heights of the stripes from the overall flag dimensions (we’ll write programs using numbers)</p>
We need a way to describe colors to our program (we’ll learn strings)</p>
We need a way to creat</p>
</blockquote>
3.1.2 Numbers</h2>

A software or web-application in which you write and run programs is called a programming environment</strong>.</p>
</blockquote>

We will use the term expression</em> to refer a computation written in a format that Pyret can understand and evaluate to an answer.</p>
</blockquote>

Expression: a computation written in the formal notation of a programming language
Value: an expression that can’t be computed further (it is its own result)
Program: a sequence of expressions that you want to run</p>
</blockquote>
Note: a literal is a value stated in the program itself. It has not bee computed.</em></p>
functions and expression composing</h3>
3.1.6.1</p>

“functions” is the term more commonly used in computing, whereas your math classes likely used “operations”</p>
</blockquote>

Every function produces a value, which can be used as input to another function. We build up expressions by using values and the outputs of functions as inputs to other functions.
This idea of using the output of one function as input to another is known as composition</strong>.</p>
</blockquote>
Types and Contracts</h3>
3.1.7.1</p>

In programming, values are organized into types</strong> (e.g., number, string, image). These types</strong> are used in turn to describe what kind of inputs and results (a.k.a., outputs) a function works with.</p>
</blockquote>

the term contract</em> refers to the required types of inputs and promised types of outputs when using a specific function.</p>
</blockquote>

Whenever you compose smaller expressions into more complex expressions, the types produced by the smaller expressions have to match the types required by the function you are using to compose them.</p>
</blockquote>

A contract also summarizes how many inputs a function expects.</p>
</blockquote>
Types of errors in programming</h3>

type error</li>
input error</li>
syntax punctuation</del> error</li>
</ul>

In programming, we use the term syntax to refer to the rules of writing proper expressions (we explicitly didn’t say “rules of punctuation” because the rules go beyond what you think of as punctuation, but that’s a fair place to start).</p>
</blockquote>
3.1.7.2</p>
"evaluated" errors, for type errors or incorrect number of inputs.</p>
space around operators
close parentheses
space after comma
quotes around strings</p>

Every programming language comes with documentation, which is where you find out the various operations and functions that are available, and your options for configuring their parameters.</p>
</blockquote>
3.2 Naming Values</h1>
unkown words at this point when typing puppy :</p>

unbound</li>
not previously defined</li>
</ul>

“unbound identifier” when an expression contains a name that has not been associated with (or bound to) a value.</p>
</blockquote>
So far, we've seen words</strong> used for :</p>

data within strings</li>
names for values (also called identifiers</strong>)</li>
</ul>
Main differences:</p>

"string" vs name</li>
string contain spaces, names cannot</li>
</ul>
Expressions vs Statements</p>

Definitions and expressions are two useful aspects of programs, each with their own role. Definitions tell Pyret to associate names with values. Expressions tell Pyret to perform a computation and return the result.</p>
</blockquote>

In programming, we distinguish expressions</em>, which yield values, from statements</em>, which don’t yield values but instead give some other kind of instruction to the language. So far, definitions are the only kinds of statements we’ve seen.</p>
</blockquote>

The program directory is an essential part of how programs evaluate.</p>
</blockquote>
Might need to delete later, if substitution is defined</em></p>

How does Pyret evaluate (width * 3)? Since width is a word (not a string), Pyret looks up its value in the directory. Pyret substitutes that value for the name in the expression, resulting in 30 * 3, which then evaluates to 90. After running these two expressions, the directory looks like:</p>
Directory</p>
width →  30</span></span>
</span>
height →  90</span></span></code></pre>
Note that the entry for height in the directory has the result of width * 3, not the expression. This will become important as we use named values to prevent us from doing the same computation more than once.</p>

In practice, programmers don’t name every individual image or expression result when creating more complex expressions. They name ones that will get used more than once, or ones that have particular significance for understanding their program. We’ll have more to say about naming as our programs get more complicated.</p>
</blockquote>
3.3 From Repeated Expressions to Functions</a></h2>

To make this program work, then, we need the ability to add our own operators</strong> (henceforth called functions</strong>) to Pyret.</p>
</blockquote>
Functions</h3>

In programming, a function takes one or more (configuration) parameters and uses them to produce a result.</p>
</blockquote>
Strategy: Creating Functions From Expressions</strong></p>

Write down at least two expressions showing the desired computation.</li>
Identify wich part is fixed and which is changing.</li>
For each changing part, give it a name, which will be the parameter that stands for that part.</li>
Rewrite the examples to be in terms of these parameters.</li>
Name the function something suggestive.</li>
Write the syntax for functions around the expression where the expression is called the body of the function.</li>
</ul>
Programmers often use angle brackets to say “replace with something appropriate”; the brackets themselves aren’t part of the notation.</em></p>

When we provide values for the parameters of a function to get a result, we say that we are calling</em> the function. We use the term call</em> for expressions of this form.</p>
</blockquote>

Substitution restores that expression, while still allowing the programmer to write the shorthand (func name).</p>
</blockquote>

annotating the parameters with information about the expected type of value for each parameter.</p>
</blockquote>
 fun three-stripe-flag(top :: String,</span></span>
      middle :: String,</span></span>
      bottom :: String):</span></span>
  frame(</span></span>
    above(rectangle(120, 30, "solid", top),</span></span>
      above(rectangle(120, 30, "solid", middle),</span></span>
        rectangle(120, 30, "solid", bottom))))</span></span>
end</span></span></code></pre>
Is this setting a type ?</em></p>

It is also common practice to add a type annotation</strong> that captures the type of the function’s output. That annotation goes after the list of parameters:</p>
</blockquote>

Note that all of these type annotations are optional. Pyret will run your program whether or not you include them. You can put type annotations on some parameters and not others; you can include the output type but not any of the parameter types. Different programming languages have different rules about types.</p>
We will think of types as playing two roles: giving Pyret information that it can use to focus error messages more accurately, and guiding human readers of programs as to the proper use of user-defined functions.</p>
</blockquote>

Programmers also annotate a function with a docstring, a short, human-language description of what the function does.</p>
</blockquote>
Python has docstrigs, java has javadoc, go has godoc, etc.</p>
the DRY principle</a>, where DRY means "don’t repeat yourself".</p>
Writing good code annotation is hard. Try to be as concise and straight to the point as possible.</p>

In each of the functions above, we’ve started with some examples of what we wanted to compute, generalized from there to a generic formula, turned this into a function, and then used the function in place of the original expressions.</p>
</blockquote>
Keep examples to use as tests.</p>

Software Evolves. Over time, any program that has any use will change and grow, and as a result may end up producing different values than it did initially.</p>
</blockquote>

Therefore, it’s always useful to keep those examples around for future reference, so you can immediately be alerted if the function deviates from the examples it was supposed to generalize.</p>
</blockquote>

In fact, this is so valuable in professional software development that good programmers always write down large collections of examples—called tests—to make sure their programs are behaving as they expect.</p>
</blockquote>

For our purposes, we are writing examples as part of the process of making sure we understand the problem. It’s always a good idea to make sure you understand the question before you start writing code to solve a problem. Examples are a nice intermediate point: you can sketch out the relevant computation on concrete values first, then worry about turning it into a function. If you can’t write the examples, chances are you won’t be able to write the function either.</mark> Examples break down the programming process into smaller, manageable steps</p>
</blockquote>

we also want to include special yet valid cases that the function might have to handle, such as an empty message.</p>
</blockquote>
Always create the test first with your expected returned values, before running the function itself. Don't use the function to provide the returned values, as this might poison your tests with erroneous examples, if you borked the function.</p>
3.4 Conditionals and Booleans</a></h2>
when choosing examples, use edge cases to fine-tune tests;</p>
proposed examples from the course follow this strategy: define the boundary, just over the boundary, both natural and real(decimal) numbers floats and ints</em>, and use both types of static value that is being used in the computation.</p>
add-shipping expression is different from other functions used so far in that there are two computed values that appear only inside the function, and change based on the input value.</p>

This calls for being able to ask questions about inputs within our programs.</p>
</blockquote>
Conditionals: Computations with Decisions</h3>
new expression: if expression (or statement) that can ask a questions that produce true or false (Booleans).</p>
else marks the false case(in the example)</p>
end instructs pyret that we're stopping the interrogation</p>
Booleans</h3>
The values true and false</p>
names after George Boole</p>
Many built-in functions return boolean</p>
x == x</p>
==</code> check if two values are equal</p>
Strings are case-sensitive!
Compare strings to determine their alphabetical order → this means that a</code> comes before b</code>, thus a < b</code>; a</code> is "smaller than b</code>"</p>
alphabetical order = ascending order</strong></p>
With strings, keep in mind of ASCII and what's outside of ASCII. → Falsehoods programmers believe about languages</a></p>
Notably, ordering depends on location (geography).</p>
In general, you can compare almost any two values for equality.</p>
There are specific operator to compare values of a specific kind (or type?)
Those operators are only accept the type, and only the type. It returns a useful error because it does type-checking(?)</del> not work on another type. Free debugging!!</p>

Then, Pyret will signal an error if you go wrong, instead of blindly returning an answer (false) which lets your program continue to compute a nonsensical value.</p>
</blockquote>

In fact, just about every kind of data will have some Boolean-valued operators to enable comparisons.</p>
</blockquote>
Combining Booleans</h2>
Make decisions on more than one boolean.</p>

Pyret offers three main operations: and</code>, or</code>, and not</code>.</p>
</blockquote>

Explain why numbers and strings are not good ways to express the answer to a true/false question.</p>
</blockquote>
well, I'm guessing its something to do with how operators determines if something is true or false?</del> I believe that numbers or strings don't necessarily describe the complexity of the question? Or maybe they might not have any relation to the question and thus are not easily understandable?</p>
Asking multiple questions</p>
else if</code> → evaluates each question expression in order</strong>, until it returns the result in case the answer that returns true.</p>
In the exercise, there is a implicit between</code> expression: (x >= 30)</code> after a if statement for (x <= 10)</code>
but there is also the if (x > 10) and (x <= 30)</code> to explicit the between 10 and 30</code></p>
The second option improves readability, especially for future readers. It makes the if statement error proof against an earlier question that would be erroneous, since we are evaluating a specific range. It would also hold up much better in case of modifications of the earlier statement too.</p>
Exercise</strong>:</p>
fun show-ad(haircolor :: String, age :: Number) -> Boolean:</span></span>
  doc: "show an ad if the haircolor is pink/purple and the age is between 9-18"</span></span>
  if ((haircolor == "pink") or (haircolor == "purple")) and ((age >= 9) and (age <= 18)):</span></span>
    true</span></span>
  else:</span></span>
    false</span></span>
  end</span></span>
where: </span></span>
  show-ad("purple",9) is true</span></span>
  show-ad("purple",18) is true</span></span>
  show-ad("purple",7) is false</span></span>
  show-ad("green",9) is false</span></span>
  show-ad("pink",18) is true</span></span>
  show-ad("pink",19) is false</span></span>
end  </span></span></code></pre>
Responsible</strong>:</p>

Assumptions about users get encoded in even the simplest functions.</p>
</blockquote>

What data about individuals should be used to represent them for processing by programs and what stereotypes might those data encode. [...] Decisions based on those predictions can be inaccurate and hence harmful.</p>
</blockquote>
Function Composition</h2>
two examples:</p>
fun1(fun2 (arg), arg)</span></span></code></pre>
and</p>
output_fun2 = fun2(arg) </span></span>
fun1(output_fun2, arg)</span></span></code></pre>
The second option commits the output_fun2</code> to "memory" as an extra step and is more verbose, while the first option directly substitutes the fun2</code> output as an argument for fun1</code>.</p>
Two styles, each with pros and cons.
There seems to be an importance in how programs evaluate. My intuition would be that it's the programming language who is responsible.</p>

Directory entries made within a function are local (private) to the function body</p>
</blockquote>
It's the whole global var</code> vs var declared inside a function</code> as in local var</code>.</p>

For some reason, the course material keeps mentioning a directory that is created for each program, where data values are store. Is this a reference to memory and memory management?</p>

Nested Conditionals</h3>


wine
2025-02-16T00:00:00+00:00
Instalation</h2>
Install wine</code> and winetricks</code>.</p>
On Archlinux, wine</code> is in the multilib</code> repo, which is not enables by default. You need to enable it first</a>.</p>
Then install :</p>
  sudo</span> pacman</span> -S</span> wine winetricks</span></span></code></pre>Minimal Wine Setup for Gaming</h2>
Install the following</p>
  winetricks</span> vcrun2022</span></span>
  winetricks</span> dxvk</span></span>
  winetricks</span> vkd3d</span></span>
  winetricks</span> corefonts</span></span>
  winetricks</span> xna40</span></span></code></pre>
Then ensure only one sound device is selected in winecfg</code> and that you have the corresponding audio backend installed.</p>
In my case, although I usually have pipewire installed, I've had better result with alsa.</p>
  winetricks sound=alsa</span></span></code></pre>
This command changes the verb</code> sound to use alsa instead of pulse.</p>
Run apps or games</h2>
To run software with wine</code>, just prefix it with the wine</code> command.</p>
  wine</span> my_app.exe</span></span></code></pre>Sources</h2>

Wine - Archlinux.org</a></li>
Wine - Wiki sebsauvage.net</a></li>
</ul>


A great explanation of the Y2K bug
2025-02-12T00:00:00+00:00
Nic Baker</h2>
I started watching all of Nic Baker's videos</a> after reading Simon Willison's post</a> on the Clay UI library that Nic Barker built.
In his post, Simon Willison mentions the intro video</a> that Nic Barker made to this library.</p>

His intro video to the library is outstanding: I learned a ton about how UI layout works from this, and the animated visual explanations are clear, tasteful and really helped land the different concepts.</p>
John Willison
I was also blown away, not just by the results of Clay, but by Nic Barker's quality as presenter and teaching.</p>
</blockquote>

I use Grayjay to follow people on Twitch, Youtube and Peertube. I highly recommended it.</p>
</aside>
This I added all his videos in the Watch Later playlist in Grayjay</a>.</p>
Y2K</h2>
Today I final start watching the video Nic Barker made about the Y2K bug</a>.</p>
Not only was it very funny, I finaly understood how base2 works !!
Thx Nic :D</p>


f2
2025-02-12T00:00:00+00:00
Source</strong> : github.com/f2</a></p>


helix
2025-02-11T00:00:00+00:00
Tutor</h2>
The tutor, aka Tutorial, is great to learn how to use Helix.</p>
Go thought it once at least. Don't hesitate to go back to it if you forget how
keybindings work.</p>
To open the tutor, run :</p>
hx</span> --tutor</span></span>
# or, if you don't have the alias hx = helix</span></span>
helix</span> --tutor</span></span></code></pre>Open file as sudo</h2>
Instead of running :</p>
sudo</span> helix myfile.conf</span></span></code></pre>
use sudoedit</code>:</p>
sudoedit</span> myfile.conf</span></span></code></pre>
This will use your $EDITOR, open a temporary file in /tmp</code> and then save it to
the correct location.</p>
Don't forget to export the EDITOR variable in your .bashrc</code> or .zshrc</code>.</p>
echo</span> '</span>export EDITOR=helix</span>' ></span> ~/.bashrc</span></span></code></pre>Reference</h3>

How to write to files that require super user rights? #4251 - github.com/helix</a></li>
man sudoedit</a></li>
</ul>
Terminal view</h2>
Seems to be in works. For now, it is recommended to run programms or builds in a
seperate terminal, a terminal multiplexer or a terminal (like kitty) that
supports spliting.</p>
References</h3>

How to run code in Helix - reddit.com</a></li>
</ul>
Write without formating</h2>
No current straightforward way to do this. But:</p>

I don't think it's necessary to add commands for this since there's a
workaround: you can :set auto-format false</code>, :write</code>, and then
:set auto-format true</code> when you want auto-formatting again.</p>
</blockquote>
This can then be bound to a key. Apparently there is a way to bind a sequence of
commands to one key.</p>


test this sequence thing.</li>
</ul>
Sources</h3>

PR - Add :write-without-auto-format commands #4909</a></li>
</ul>
Syntax Highlighting, LSP, Formaters & Linters</h2>
GO TO my lsp configuration wiki page</a>.</p>
Yank a diagnostic</h3>
:yank-diagnostic</code></p>
Select all occurence of regex in a document</h3>
Use %</code> to select the whole document, then use s</code> to select all occurence of
the regex.</p>
You can be more specific by being more selective before s</code>.</p>
Workspace</h2>
.helix</code></h3>
Disabling Autoformatting for JavaScript in Helix</h4>
To stop Helix from autoformatting .js</code> files, add this block to your
.helix/languages.toml</code>:</p>
[[</span>language</span>]]</span></span>
name</span> = "</span>javascript</span>"</span></span>
auto-format</span> =</span> false</span></span></code></pre>
This overrides the default JavaScript config so files are no longer formatted
automatically on save. You can still run :format</code> manually if needed.</p>


emacs
2025-02-07T00:00:00+00:00
Run Emacs in a terminal</h2>
well, I haven't found it again yet.</p>
There seems to be a emacs daemon mode, that acts like a server. I can then just call emacsclient</code> in a terminal ? Like foot terminal ?</p>


Weekly Single Link #3
2025-02-05T00:00:00+00:00
This week's Single Link is about AI.</p>
Specifically, the web crawler bots employed by AI companies & startups to siphon up content from the internet. I've been reading quite a few posts (blogs and social media) about people observing those web crawlers either scraping entire websites while thoroughly ignoring robots.txt files, or even DDoSing forums</a> and self-hosted services.</p>
This is what was happening with Xe Iaso</a>'s self-hosted Gitea instance. The solution they came up with is Block AI scrapers with Anubis</a>. This is the week's link !!</p>
Anubis is a middleware that forces any browser or bots to solve a puzzle with JavaScript. It's pretty neat, and it has a cool illustration that you'll see if you visit the test page</a>.</p>
It is the first of its kind. Other</a> methods</a> always seem to be a poison attack, akin to tarpits</a> that serve bad content. Anubis is maybe what I would call a naïve captcha ? Much less problematic version of Cloudflares in terms of usability, but probably more costly for the hoster.</p>
PS; I know this is not just one link. But I need to set up the context. It's fine. The objective is to reduce flooding readers with links. I'm mostly keeping to that rule.</p>


I'm a brocoli
2025-02-05T00:00:00+00:00
Apperently ?</p>
</p>
Discovered at What vegetable are you?</a> via Sylvia's about page</a>.</p>


Caddy Static File Server Configuration
2025-02-05T00:00:00+00:00
Compression</h2>
To add gzip encoding support :</p>
:</span>80</span> {</span></span>
  root * /srv/www</span></span>
  file_server</span></span>
  encode zstd, gzip</span> </span></span>
}</span></span></code></pre>Basic Site Configuration Approaches</h2>
1. Domain-Based Approach</h3>
site.example.com </span>{</span></span>
    # Configuration here</span></span>
}</span></span></code></pre>

Automatically handles HTTP/HTTPS</li>
Cleaner syntax</li>
Automatic SSL certificate management</li>
Best for simple, single-domain setups</li>
</ul>
2. Port-Based Approach</h3>
:</span>80</span> {</span></span>
    @site1 host site.example.com</span></span>
    handle @site1 {</span></span>
        # Configuration here</span></span>
    }</span></span>
}</span></span></code></pre>

More flexible for complex setups</li>
Better for multiple domains with shared configurations</li>
Explicit port control</li>
Requires separate HTTPS configuration</li>
</ul>
Static vs General File Handling</h2>
Static Files</h3>

Files that don't change frequently (CSS, JS, images, fonts)</li>
Suitable for aggressive caching</li>
Served directly from disk</li>
Typically matched by file extensions</li>
Example extensions: .css</code>, .js</code>, .jpg</code>, .png</code>, .woff2</code></li>
</ul>
General Files</h3>

Any type of file</li>
May be dynamic</li>
Might need processing</li>
Less aggressive caching</li>
No specific extension matching</li>
</ul>
Caching Configuration</h2>
Client-Side Caching</h3>
header @static </span>{</span></span>
    Cache-Control</span> "</span>public, max-age=604800, must-revalidate</span>"</span></span>
}</span></span></code></pre>

Controlled via Cache-Control headers</li>
Tells browsers how long to cache content</li>
Common durations:

1 week: 604800 seconds</li>
1 month: 2592000 seconds</li>
</ul>
</li>
</ul>
Server-Side Caching</h3>

Optional in Caddy</li>
Used for reducing server load</li>
Not necessary for basic static file serving</li>
Beneficial for dynamic content</li>
</ul>
Complete Configuration Example</h2>
:</span>80</span> {</span></span>
    # Global compression</span></span>
    encode zstd gzip</span></span>
</span>
    # Security headers</span></span>
    header {</span></span>
        Content-Security-Policy</span> "</span>default-src 'self'; script-src 'self'; style-src 'self'; img-src 'self'; font-src 'self'; object-src 'none'</span>"</span></span>
    }</span></span>
</span>
    # Static file handling</span></span>
    @static </span>{</span></span>
        file</span></span>
        path *.css *.js *.jpg *.jpeg *.png *.gif *.ico *.svg *.woff *.woff2</span></span>
    }</span></span>
    handle @static </span>{</span></span>
        header Cache-Control</span> "</span>public, max-age=604800, must-revalidate</span>"</span></span>
    }</span></span>
</span>
    # Site configurations</span></span>
    @site</span>1</span> host site</span>1</span>.example.com</span></span>
    handle @site</span>1</span> {</span></span>
        root * /path/to/site1</span></span>
    }</span></span>
</span>
    # Global file server</span></span>
    file_server</span></span>
}</span></span></code></pre>Best Practices</h2>


Header Order</strong></p>

Place global headers before handling directives</li>
Security headers should be defined early in the configuration</li>
</ul>
</li>

File Serving</strong></p>

Use a single global file_server</code> directive when possible</li>
Only specify roots in site-specific handlers</li>
</ul>
</li>

Compression</strong></p>

Apply compression globally using encode zstd gzip</code></li>
Let Caddy handle compression automatically</li>
</ul>
</li>

Security</strong></p>

Always include Content Security Policy (CSP) headers</li>
Use appropriate Cache-Control</code> directives</li>
Consider additional security headers as needed</li>
</ul>
</li>

Caching Strategy</strong></p>

Use client-side caching for static assets</li>
Consider server-side caching only if needed</li>
Include must-revalidate</code> for better cache control</li>
</ul>
</li>
</ol>
Common Issues</h2>


CSS/JS Not Loading</strong></p>

Check CSP headers</li>
Verify file permissions</li>
Confirm correct file paths</li>
Ensure style-src</code> and script-src</code> are properly configured</li>
</ul>
</li>

Caching Issues</strong></p>

Verify Cache-Control header syntax</li>
Check browser developer tools</li>
Clear browser cache during testing</li>
</ul>
</li>

Multiple Site Configuration</strong></p>

Use appropriate matching based on hosts</li>
Maintain correct root paths</li>
Keep shared configurations at the global level</li>
</ul>
</li>
</ol>
Sources</h2>

header - Caddy Documentation</a></li>
header - Caddy Documentation</a></li>
</ul>

Interesting setups involving caddy</code>:</p>

Static Sites with Python, uv, Caddy, and Docker</a></li>
</ul>


robots.txt
2025-02-05T00:00:00+00:00
robots.txt are</p>

robots.txt is the filename used for implementing the Robots Exclusion
Protocol, a standard used by websites to indicate to visiting web crawlers and
other web robots which portions of the website they are allowed to visit.</p>
robots.txt - wikipedia</a></p>
</blockquote>
AI bot crawlers, how to handle them</h2>
There are a couple of projects that aim at helping against ai companies bots
that crawl and scrape our content. Amongst them:</p>

robots.txt/ai</a></li>
ai-robots.txt</a></li>
</ul>
I'm going to setup in my Justfile a update setting for my robots.txt file that
is in my static folder.</p>
https://github.com/ai-robots-txt/ai.robots.txt/releases/latest/download/robots.txt</span></span></code></pre>curl</span> --url</span> https://github.com/ai-robots-txt/ai.robots.txt/releases/latest/download/robots.txt</span> -o</span> static/robots.txt</span></span></code></pre>
About robots.txt</code> to block crawlers from indexing stuff.</p>
https://searchengineland.com/robots-txt-new-meta-tag-llm-ai-429510</p>
https://www.robotstxt.org/robotstxt.html</p>
https://developers.google.com/search/docs/crawling-indexing/robots/intro?visit_id=638273818678687427-847406478&rd=2#:~:text=The%20instructions%20in%20robots.txt%20files%20cannot%20enforce%20crawler%20behavior%20to%20your%20site%3B%20it%27s%20up%20to%20the%20crawler%20to%20obey%20them.</p>
https://developers.google.com/search/docs/crawling-indexing/block-indexing</p>


aerc
2025-02-04T00:00:00+00:00
aerc, or A</p>


dotfiles
2025-02-04T00:00:00+00:00
stow + git repo</h2>
Install stow</code> and git</code>. Create an empty git repo.</p>
Stow is used to symlink folders and files to specific places.</p>
For example, if a folder called git_repo_name/helix/.config/helix/config.toml</code>,
and I run stow helix</code> from the root of the git repo, it whill symlink
git_repo_name/helix/.config/helix/config.toml</code> into
~/.config/helix/config.toml</code>.</p>
When I modify my config files, I'm essentially modifying the files in my repo
directly.</p>
I have another page that about GNU Stow</a>.</p>
If I ever make my dotfiles public again, I will link them here.</p>
Chezmoi</h2>
https://chezmoi.io</a></p>
Init new dotfiles:</p>
chezmoi</span> init</span></span></code></pre>
Add files or folders to chezmoi dotfiles:</p>
chezmoi</span> add ~/.bashrc</span></span></code></pre>
Note:</em> Add with the whole path ?</p>


GNU Stow
2025-02-04T00:00:00+00:00
Sources</h2>
I use GNU stow</a>.</p>
I followed the this
tutorial by Brandon Invergo</a>,
linked in the official GNU Stow page</p>
Process</h2>
Stow enables you to store config files in a folder and symlink them in the
proper places.</p>
In the case of my dotfiles are stored in a folder called dotefiles that is also
a git repo.</p>
The structure is as follows, with helix</a> and sway
sway</a> as examples:</p>
dotfiles</span></span>
├─ helix</span></span>
├─── .config</span></span>
├────── helix</span></span>
├──────── config.toml</span></span>
├──────── languages.toml</span></span>
├─ sway</span></span>
├─── .config</span></span>
├────── sway</span></span>
├──────── config</span></span>
├──────── styles.css</span></span>
├─ other programs ...</span></span></code></pre>
When running stow folder_name</code> it will symlink the folder or it's content (if
already preexisting folder) to the location.</p>


sway
2025-02-04T00:00:00+00:00
sway is currently the most stable wayland compositor I have ever used. It's a
drop-in replacement to i3</a>.</p>
Setup</h2>
Software</h3>
Here is the list of software I use with sway to run a functionning desktop:</p>

sway</li>
swaylock</li>
foot</li>
waybar</li>
fuzzel</li>
mako</li>
micro</li>
yazi</li>
aerc</a></li>
</ul>
Quick install</h3>
Here is an opinionated barebones list of basic programs you will need to run
sway. For a full overview of tools and software I use, there is a
list</a> in the dotfiles entry of the wiki.</p>
sudo pacman -S sway swayidle swaylock swaybg \</span></span>
foot waybar fuzzel mako wlsunset wl-clipboard \</span></span></code></pre>Sources</h2>

archlinux</li>
gentoo wiki</li>
</ul>
Swayidle Configuration Troubleshooting</h1>
Common Issues with Screen Locking and Power Management</h2>
When configuring swayidle for automatic screen locking and display power
management, certain syntax errors can prevent proper functionality.</p>
Problematic Configuration</h3>
exec</span> swayidle</span> -w</span> \</span></span>
    timeout</span> 600</span> '</span>waylock -fork-on-lock</span>'\ </span>      </span></span>
    timeout</span> 900</span> '</span>swaymsg "output * power off"</span>'\</span></span>
    resume</span> '</span>swaymsg "output * dpms on"</span>'\</span></span>
    before-sleep</span> '</span>waylock</span>'</span></span></code></pre>
Issues:</strong></p>

waylock -fork-on-lock</code> uses an invalid flag that doesn't exist</li>
Mixing power off</code> with dpms on</code> creates inconsistent commands</li>
</ul>
Corrected Configuration</h3>
exec</span> swayidle</span> -w</span> \</span></span>
    timeout</span> 600</span> '</span>waylock</span>' \</span></span>
    timeout</span> 900</span> '</span>swaymsg "output * power off"</span>' \</span></span>
    resume</span> '</span>swaymsg "output * power on"</span>' \</span></span>
    before-sleep</span> '</span>waylock</span>'</span></span></code></pre>Power vs DPMS Commands</h3>
Modern approach (recommended):</strong></p>

swaymsg "output * power off/on"</code> - Uses current Wayland protocols</li>
</ul>
Legacy approach (still works):</strong></p>

swaymsg "output * dpms off/on"</code> - Based on older X11 DPMS standards</li>
</ul>
Both achieve the same result, but power</code> commands are preferred for new
configurations as they use modern Wayland-native protocols and are more
future-proof.</p>


micro editor
2025-02-01T00:00:00+00:00
Custom scripts</h1>
textfilter</code> does not parse arguments to a shell that would expand $HOME</code> so you could try binding with command:textfilter sh -c \"~/path_to_script\"</code>.</p>
You could also move the script to a directory in $PATH</code> (default ~/.profile</code> may add ~/.local/bin</code>) so that you'll just have to bind command:textfilter script_name</code></p>
Programs in $PATH</code> can be run anywhere, so I personally add a prefix like micro-</code> in the filename or avoid adding program-specific parts in scripts.</p>


Fediverse
2025-01-29T00:00:00+00:00
Ressources</h2>

FediDevs</a></li>
</ul>


ffmpeg
2025-01-29T00:00:00+00:00
List Video file's codec</h3>
	ffprobe -v error -select_streams v:0 -show_entries stream=codec_name -of default=noprint_wrappers=1:nokey=1 video.mkv</span></span></code></pre>
Source</strong>: How to determine video codec of a file with FFmpeg -- StackOverflow</a></p>
Ressources</h2>

FFmpeg by Example</a></li>
</ul>


TUI Timer
2025-01-28T00:00:00+00:00
Tools used:</p>

clock-tui</a></li>
pipewire</li>
pipewire-pulse</li>
notify-send</li>
</ul>
with pplay</h3>
tclock</span> timer</span> -d</span> 1m</span> -e</span> "</span>notify-send </span>"</span>Times up</span>"</span>; paplay ~/Music/Tibetan_Singing_Bowl_Set_Sound_Bowl_Meditation_Bowl-sPOTaJZ7cbw.opus</span>"</span></span></code></pre>with pw-play</h2>
tclock</span> timer</span> -d</span> 2s</span> -e</span> "</span>notify-send </span>"</span>Times up</span>"</span>; pw-play ~/Music/Tibetan_Singing_Bowl_Set_Sound_Bowl_Meditation_Bowl-sPOTaJZ7cbw.opus</span>"</span></span></code></pre>Sources</h2>

Play single mp3 through specific Pipewire device via shell command</a></li>
</li>
</ul>


yt-dlp
2025-01-28T00:00:00+00:00
Video</h2>
Audio</h2>
Download and extract best audio</h3>
yt-dlp</span> -f</span> bestaudio</span> --extract-audio</span> URL</span></span></code></pre>Automation</h2>
Stream a video directly in a video player without saving to file</h3>
yt-dlp</span> -o</span> - URL</span> |</span> mpv</span> -</span></span>
# or vlc, or another video player</span></span></code></pre>Download from a list inside a .txt file</h3>
# -a/--batch-file</span></span>
yt-dlp</span> -a</span> links.txt</span></span></code></pre>


Single Weekly Link #2
2025-01-27T00:00:00+00:00
Here is the Single Weekly Link</strong>!</p>
Man this was hard for me. I had an extraordinary amount of things I would have loved to share with you since last week. But I've allowed myself one link.</p>
This week's link is Simulating worlds on the GPU: Four billion years in four minutes - David A Roberts</a>.</p>
PS: I think I'll make a shortcode element to make the single link pop more on the page.</p>
Anywho, tata :D</p>


Science is late to the party
2025-01-27T00:00:00+00:00
A list of articles that illustrate things people know are observed, or how science is late to the party.</p>


CSS
2025-01-26T00:00:00+00:00
CSS resources</p>


Random Secret Key Generation
2025-01-26T00:00:00+00:00
To generate a random string of a set size, on linux, run :</p>
# for hex string</span></span>
</span>
openssl</span> rand</span> -hex</span> number</span></span>
</span>
# output:</span></span>
</span>
35d9bbe3576a0866335eea99ba7b6a248f89d07d51c2b30022a8b06f7ac587da</span></span>
</span>
# or for base64 string</span></span>
</span>
openssl</span> rand</span> -base64</span> number</span></span>
</span>
# output:</span></span>
</span>
8SGmlygNlJFRPtsvcwzI17VYksi2sblO2AD1I2BPLxs</span>=</span></span></code></pre># hex of length 16</span></span>
openssl</span> rand</span> -hex</span> 16</span></span>
</span>
# ... 32</span></span>
openssl</span> rand</span> -hex</span> 32</span></span>
</span>
# ... 64</span></span>
openssl</span> rand</span> -hex</span> 64</span></span></code></pre>
For improved compatibility and reliability, I use the hexadecimal (hex)</strong> version of character sets to generate secret keys and database (DB) passwords for my Docker Compose containers.</p>


Flutter
2025-01-25T00:00:00+00:00
Installation (Archlinux)</h2>

Install Flutter on Arch Linux</a></li>
</ul>

Install flutter</li>
</ol>
yay</span> -S</span> flutter</span></span></code></pre>

You then need to adjust the permissions of the binary and install folder so that we can run stuff as our user:</li>
</ol>
sudo</span> groupadd flutterusers</span></span>
sudo</span> gpasswd</span> -a</span> $USER</span> flutterusers</span></span>
sudo</span> chown</span> -R</span> :flutterusers /usr/lib/flutter</span></span>
sudo</span> chmod</span> -R</span> g+w /usr/lib/flutter</span></span>
sudo</span> chown</span> -R</span> :flutterusers /bin/flutter</span></span>
sudo</span> chmod</span> -R</span> g+w /bin/flutter</span></span></code></pre>

Android -> Install the necessary packages:</li>
</ol>
yay</span> -S</span> android-sdk android-sdk-platform-tools android-sdk-build-tools</span></span>
yay</span> -S</span> android-platform</span></span></code></pre>

And also fix the permissions</li>
</ol>
sudo</span> groupadd android-sdk</span></span>
sudo</span> gpasswd</span> -a</span> $USER</span> android-sdk</span></span>
sudo</span> setfacl</span> -R -m</span> g:android-sdk:rwx /opt/android-sdk</span></span>
sudo</span> setfacl</span> -d -m</span> g:android-sdk:rwX /opt/android-sdk</span></span></code></pre>

You’ll probably also have to set up environment variables for ANDROID_SDK_ROOT</code> and JAVA_HOME</code>.</li>
</ol>
The Java version is under /usr/lib/jvm</code>, in the case of version 8, it’s this:</p>
export</span> JAVA_HOME</span>='</span>/usr/lib/jvm/java-8-openjdk</span>'</span></span></code></pre>Usage</h2>
Check installation</h3>
To check if flutter is correctly installed:</p>
flutter</span> doctor</span></span></code></pre>Run the app locally</h3>
To run the app on linux:</p>
flutter</span> run</span></span></code></pre>
to use Hot Reload while running the app, you need to focus the terminal running the app and press r</code>.</p>
Alternatively, use dashmon</a></p>
To install it, run :</p>
flutter</span> pub global activate dashmon</span></span></code></pre>
and substitute flutter run</code> with dashmon</code>.</p>
More information in the package page.</p>
Projects</h2>
Project structure</h3>
Configuration</h3>

pubspec.yaml</code> is the configuration file for the app you are making.</li>
analysis_options.yaml</code> is the configuration file for code analysis.</li>
</ul>
main.dart</h3>


Every widget defines a build() method that's automatically called every time the widget's circumstances change so that the widget is always up to date.</p>
</li>

Every build method must return a widget or (more typically) a nested tree of widgets. In this case, the top-level widget is Scaffold. You aren't going to work with Scaffold in this codelab, but it's a helpful widget and is found in the vast majority of real-world Flutter apps.</p>
</li>

Having separate widgets for separate logical parts of your UI is an important way of managing complexity in Flutter.</p>
</li>

safe area: takes into account
expanded, takes the whole space</p>
</li>
</ul>


Single Weekly Link #1
2025-01-21T00:00:00+00:00
I'm going to try out a new practice. I'm calling it Single Weekly Link</strong>.</p>
Some blogs I follow (through RSS) have link dumps. A link dump is a blog post that lists links the author has found/read/stumbled upon.</p>
I used to be follow many of those blogs. Some I still do. An inherent problem I have this is that it's very time consuming. When you subscribe to multiple blogsn you very quickly end up with dozens of open tabs. In my case, I regularly achieved the hundreds of tabs in one sitting. Thankfully, the computers I usually have come equipped with more than 16 Go of RAM. This is sufficient to handle Firefox with a whole lot of tabs opened. The challenge was then to triage and read those new webpages before a new batch of link dumps arrived. I struggle to do so.</p>
So here is a format more respectful of our time: I'll post every week a new post entitled Single Weekly Link, containing a link that I want to share with the readers of this blog.</p>
This weeks link is  A 2-ply minimax chess engine in 84,688 regular expressions, by Nicholas Carlini</a>.</p>


Zola
2025-01-20T00:00:00+00:00

Breadcrumbs</a></li>
blogging section inspiration</a></li>
list section inspiration</a></li>
wiki section inspiration</a></li>
</ul>
Shortcodes don't work in templates, only in content files. Use macros</code> in templates/macros/</code> to make custom components and functions.</p>
Tera (the Rust Jinja-inspired templating engined) has a integer → human readable bitsize function.</p>


Publish Online
2025-01-11T00:00:00+00:00
A list of ressources to publish content online. This is both a list of tools to make websites, and places or tools to take that website online.</p>
Tools</h2>

zola</a></li>
haunt</a></li>
quartz</a>, for obsidian -> internet</li>
</ul>
Platform</h2>

github pages</li>
gitlab pages</li>
codeberg pages</li>
</ul>


Art
2025-01-08T00:00:00+00:00

DOOM: The Gallery Experience</a></li>
</ul>


Toolbox
2025-01-08T00:00:00+00:00
This is a list of tools I keep coming back to. Across GUI apps, CLI and TUI, web
services, entire desktop environments, to tools in my EDC, Every Day Carry</em>
set, I hope this nudges someone towards awesome software or hardware.</p>
Software</h2>

linux-wifi-hotspot</li>
blueman</li>
btop</li>
micro</li>
oh-my-bash/oh-my-zsh</li>
sway</li>
firefox</li>
zed</li>
obsidian</li>
foot</li>
stremio</li>
bitwarden</li>
zathura</li>
mpv</li>
zola</li>
</ul>
Online Tools</h2>

SQLite File Format Viewer</a></li>
vert.sh</a> → that one has not worked yet.</li>
BreezePDF</a></li>
Atkinson Dithering</a></li>
</ul>


Programming languages
2024-12-30T00:00:00+00:00
Learn Programming</h2>

A Data-Centric Introduction to Computing (DCIC)</a></li>
</ul>
Weird</h2>

Piet (Mondrian)</a></li>
</ul>


Wikipedia Reading List
2024-12-26T00:00:00+00:00



Web Dev Ressources
2024-11-19T00:00:00+00:00
HTML</h2>

Every HTML Element</a></li>
HTML: HyperText Markup Language - MDN</a></li>
</ul>
CSS</h2>

My other list of CSS ressources</a></li>
</ul>
Fonts</h2>
Font-size</h3>

Fluid Type Scale Calculator</a>

Creating a Fluid Type Scale with CSS Clamp</a></li>
https://typescale.com/</a></li>
https://utopia.fyi/type/calculator/</a></li>
Consistent, Fluidly Scaling Type and Spacing</a></li>
Modern fluid typography editor</a></li>
</ul>
</li>
</ul>
Interactivity</h2>
Dark light toggle</h3>

Dark mode in css switcher example</a></li>
</ul>


Artist Websites
2024-10-21T00:00:00+00:00



Rapid Photo Downloader Setup
2024-10-09T00:00:00+00:00
My PHOTOS file name template:</p>
<Image date (YYYY)>/<Image date (MM)>/<Image date (YYYY_MM_DD)>_<Image date (HH-MM-SS)>_<Camera model (UPPERCASE)></span></span></code></pre>
My VIDEOS file name template:</p>
<Video date (YYYY)>/<Video date (MM)>/</span></span></code></pre>


Using greetd as login manager
2024-09-26T00:00:00+00:00
I changed my login manager from sddm</a> to
greetd</a>. I am able to login into sway,
hyprland and riverwm.</p>
I decide to change because it felt that greetd might be more lightweight then
sddm.</p>
It is in fact, with greetd using about 3.6M and the worker launching the
selected session at 8.6M. Compared to sddm at 21M, it's slightly smaller in
memory use.</p>
I held back since Hyprland documentation seemed to indicate that it would not
work with greetd. Since then, I moved back to sway due to weird gpu crashes when
using hyprland. I decided today to jump ship.</p>
I used regreet as a greeter. I'm already thinking of changing to using qtgreet
instead.</p>
To install on archlinux:</p>
sudo</span> pacman</span> -S</span> greetd regreet qtgreet</span></span></code></pre>
Everything happens in /etc/greetd/</code> where:</p>

you configure greetd with config.toml</code></li>
the greeter with regreet.toml</code></li>
the config for the compositor (here sway</code>) used to display the greeter</li>
the entries in environment</code></li>
</ul>
My config is as follows:</p>
config.toml</code></h2>
[</span>terminal</span>]</span></span>
# The VT to run the greeter on. Can be "next", "current" or a number</span></span>
# designating the VT.</span></span>
vt</span> =</span> 1</span></span>
</span>
[</span>default_session</span>]</span></span>
command</span> = "</span>sway --config /etc/greetd/sway-config</span>"</span></span>
user</span> = "</span>greeter</span>"</span></span></code></pre>
greetd starts on tty1</p>
it needs a compositor to start. This is way sway is launched first, I believe...</p>
sway-config</code></h2>
exec "regreet; swaymsg exit"</span></span>
include /etc/sway/config.d/*</span></span>
</span>
input * {</span></span>
	xkb_layout fr</span></span>
}</span></span></code></pre>
and the environment:</p>
sway</span></span>
hyprland</span></span>
river</span></span>
bash</span></span>
zsh</span></span></code></pre>
I also needed to modify PAM so that my keyring is opened on login. Nextcloud
Desktop Sync app notably asks for a password or just does not login and needs to
reauthencate on each login. Which is very annoying.</p>
#%PAM-1.0</span></span>
auth       required     pam_securetty.so</span></span>
auth       requisite    pam_nologin.so</span></span>
auth       include      system-local-login</span></span>
auth       optional     pam_gnome_keyring.so  # Add this line</span></span>
account    include      system-local-login</span></span>
session    include      system-local-login</span></span>
session    optional     pam_gnome_keyring.so auto_start  # Add this line</span></span></code></pre>


Awesome Lists
2024-08-09T00:00:00+00:00
Awesome</strong> Lists sont des listes crée par des utilisateurs de Github. Chaque
liste regroupe des awesome</em> ressources, liens, projets, en rapport à un sujet
particulier.</p>
Voici ma liste de Awesome Lists</strong>:</p>

Awesome Selfhosted</a></li>
Awesome FOSS apps to analyze and learn from</a></li>
Awesome Board games</a></li>
Awesome RPG</a></li>
Awesome Godot</a></li>
</ul>
Self-hosting lists</h2>

https://awesome-selfhosted.net</a></li>
</ul>
General Purpose Wikis</h2>

https://www.baeldung.com/</a></li>
</ul>
Self-Hosting Wikis</h2>

https://selfhoster.dev/</a></li>
https://libreserver.org/</a></li>
</ul>
Web Servers</h2>

https://selfhoster.dev/simple-docker-compose/</a></li>
https://tildegit.org/solderpunk/shizaru/</a></li>
Use Traefik 2 with Nginx, Apache, or CaddyServer to Serve Static Files</a></li>
</ul>
Container Image Providers</h2>

https://www.linuxserver.io/</a></li>
https://hotio.dev/</a></li>
</ul>
Server OS</h2>

https://homelabos.com/</a></li>
https://umbrel.com/</a></li>
</li>
</ul>


Fabulous TTRPG
2024-08-09T00:00:00+00:00
Systems</h2>
General Mastodons of the TTRPG space</h3>

Apocalypse World</a></li>
Books & Bullets</a> Cthulhu micro
trpg.</li>
</ul>
Open Source</h3>

Roll for Shoes</a></li>
Cairn SRD</a></li>
Primeval 2D6</a></li>
Flexspiel</a></li>
GIST!</a></li>
Saga</a></li>
</ul>
Solo RPGs</h3>

One Page Solo Engine</a></li>
</ul>
Untested</h3>

Star Crossed</a> →
official store with actual play</a></li>
Mythic Bastionland</a> - In Mythic Bastionland you begin as a young knight errant, seeking the glory to lead warbands, earn a place in court, and rule your own domain. Glory is found in hunting the Myths of this world, manifesting them into reality while protecting the Realm from their strange influence.
Each player is one of 72 Knights, with personal equipment, a unique ability, and a passion that fuels their spirit. As your glory grows, you might eventually embark on the legendary City Quest, an impossible challenge to find The City itself, a shining metropolis seen only in dreams.</em></li>
The GLOG</a>

Glog West TTRPG</a></li>
</ul>
</li>
Vaults of Vaarn</a>,fragments of a blue ruin</em></li>
</ul>
Creators</h2>

John Harper</a> Blades in the Dark</em>, Lasers &
Feelings</em>, World of Dungeons</em>, ...</li>
The Bardic Inquiry</a> → Solo RPGs</li>
</ul>
Tools</h2>
Cartography</h3>

hex-kit</a> - Hex Kit is a multi-platform desktop application for building hex maps to be used with table top role playing games. Rather than being bogged down by features and complications, Hex Kit is built to be intuitive and quick to use with an emphasis on art.</em></li>
</ul>
Music</h3>
Desktop/Web</strong></p>

Kenku FM</a></li>
Tabletopy  Sounds for tabletop role-playing games, RPG sounds</a></li>
Tabletop Audio - Ambiences and Music for Tabletop Role Playing Games</a></li>
Sound Effects & Ambient Music for RPGs  Syrinscape</a></li>
dScryb  Descriptions, Sounds and Maps for Roleplaying Games</a></li>
RPG Sound Made Easy</a></li>
</ul>
Phone App</strong></p>

Audio Forge - Transform Your TTRPG Soundscapes  Slashpaf</a></li>
Tabletop RPG Soundboard & Audio App  TableTone</a></li>
</ul>
DM</h3>

The Oracle</a></li>
Combat Cards</a>
→ I want to use this with
Fantasy Statblock</a>.</li>
UNE</a>
Universal NPC Emulator</em></li>
</ul>
Obsidian Plugins</strong></p>

Javalent's Plugins</a> → Includes
Fantasy Statblocks</a>, Leaflet map
stuff, intitiative tracker,</li>
</ul>
Gameplay Mecanics</strong></p>

Dungeon World's Fronts</a></li>
</ul>
Social Moderation Tools & Methods</h3>

X-Card</a>
by John Stavropoulos.</li>
Script Change RPG Toolbox</a></li>
</ul>
Ressources</h2>
Online Platforms</h3>

Roll20</a></li>
Foundry Virtual Tabletop</a></li>
Let's Role</a></li>
Owlbear.rodeo</a></li>
PlayRole</a></li>
BeyondTabletop</a></li>
Tableplop</a></li>
FreeBeeGee</a></li>
</ul>
Notes</h3>

Critical Notes</a></li>
</ul>
Whiteboards</h3>

Excalidraw.com</a></li>
</ul>
Graphic Design</h3>

https://game-icons.net/</a></li>
</ul>
Other Online lists</h2>

Awesome-tabletop-rpg</a></li>
https://nagoshiashumari.github.io/Rpg-Awesome/</a></li>
</ul>


Markiplier going strategist
2024-08-06T00:00:00+00:00
I was watching, then listening to Markiplier's Series of Videos on The Joy of Creation:</p>
First Video</h2>

TGM

Fabulous Brussels

Go

Thinkpad

Cameron n’est qu’une incarnation de plus de la prédation du capitalisme, cynique au point de nous faire passer un monde synthétique comme un geste politique

Nix

Useful Links</h2> Package Search & configuration options</h3> Packages</strong></p>

Package Search & configuration options</h3> Packages</strong></p>

Installation</h2>

System-wide/userspace-wide</h3> There are two ways to install nix to use as your system package manager etc:</p>

Specifics</h3>

Fripes à Bruxelles

Bonne question pour savoir comment l'autre personne va

Python

Fabulous Hack

Thunderbird

Troubleshooting</h2>

Fabulous Manuals

Android

Flour Glue

Euro Cloud Providers

Forward Proxy

ShadowDark

Fabulous Rust

Tools</h2>

Authors of note</h2> This is a listing of people that work on rust, whose work is too extensive to list in other sections. I however do not consider those people better then the devs working on projects solo etc.</p>

Maps

Virtual Table Top

Online VTT</h2>

Roll20</h3>

Audio in TTRPG

Original Soundtracks</h2>

Video Games</h3> all souls games</strong></p>

Fabulous Communication

Fabulous Board Games

Fabulous Web

My setup on linux

General</h2> Those couple of pages have pretty good recommendation for tools to use on your system.</p>

Sway</h2>

Media</h3> avizo</a> - a simple notification daemon, mainly intended to be used for multimedia keys for example with Sway.</p> </blockquote> </li> </ul>

date

Open a file with today's date:</h2> # Your editor + the date expression + the rest of the file name + file extension </span></span> hx</span> $(</span>date</span> '</span>+%Y-%m-%d</span>')</span>_something.md</span></span></code></pre>

Créer un ficher avec la date

Créer un fichier avec la commande date</h1> Il est courant d'utiliser la commande [[date]] pour générer dynamiquement un nom de fichier basé sur la date ou l'heure actuelle. Cette technique est utile pour organiser des fichiers ou des journaux.</p>

Ressources</h2> Consultez man date</code> pour explorer les options de format.</p> </li> Explorez des scripts avancés pour des tâches automatisées avec des dates.</p> </li> </ul>

Rust Syntax

date

DIY Sacs de vélo et autres matos</h1>

sac

Stuff sack</h3>

Workspace

Shell tricks

Fuzzy-Finding and Opening Files in Helix (fzf / skim)</h3> Open file with fzf:</p>

Find and remove file</h3> rm</span> $(</span>sk</span>)</span></span></code></pre>

Maps & Routing

General Maps found online (desktop)</h2>

Site de Rando et Balades

Bretagne</h2>

Fabulous Static Site Generators

Bretagne

cargo

rustup

Mini PCs

License & Copyright

Handling Third-Party Code and License Notices</h2> When incorporating external code (directly or indirectly) into a multi licensed or open source project, the following rules could apply:</p>

Rclone

LLM serpent

State Surveillance, Arch and Psych

Fabulous Audio

LLM

Prompt Engineering</h2>

Context Engineering</h2>

Interface</h2> → Ollama and LM Studio provide and inteface</p>

Rust

rustup</h3> rustup</code> packages all things rust, including rustc</code>,cargo</code>,rust-analyzer</code>, and the std</code> lib.</p>

rust-analyzer</h3> Syntax</h2> fn main()</code> is always the first function to run in a rust program (probably true for most programming languages).</p> Most lines of Rust code end with a semicolon.</strong></p>

Syntax</h2> fn main()</code> is always the first function to run in a rust program (probably true for most programming languages).</p> Most lines of Rust code end with a semicolon.</strong></p>

Dependencies</h2>

Compiling</h2>

Types of builds</h3> --dev</code> builds are optimized at level 0, which means it's unoptimized, but is also faster to compile.</p>

Authors of note</h2>
This is a listing of people that work on rust, whose work is too extensive to list in other sections. I however do not consider those people better then the devs working on projects solo etc.</p>

Armin Ronacher / mitsuhiko</a></li> </ul>

Video Games</h3>
all souls games</strong></p>
Dark Souls 1</a></li>
Dark Souls 3</a></li>
Bloodborn</a></li>
Elden Ring</a></li> </ul>
Other Games</strong></p>
The Witcher 3: The Wild Hunt</a></li> </ul>
Films</h3>

King Arthur</a></li> </ul>
TV Shows</h3>

Carnival Row</a></li> </ul>

`Open a file with today's date:</h2> # Your editor + the date expression + the rest of the file name + file extension </span></span> hx</span> $(</span>date</span> '</span>+%Y-%m-%d</span>')</span>_something.md</span></span></code></pre>`

Créer un fichier avec la commande date</h1>
Il est courant d'utiliser la commande [[date]] pour générer dynamiquement un nom de fichier basé sur la date ou l'heure actuelle. Cette technique est utile pour organiser des fichiers ou des journaux.</p>

Ressources</h2>

Consultez `man date</code> pour explorer les options de format.</p> </li>`
`Explorez des scripts avancés pour des tâches automatisées avec des dates.</p> </li> </ul>`

Find and remove file</h3>
`rm</span> $(</span>sk</span>)</span></span></code></pre>`

Handling Third-Party Code and License Notices</h2>
When incorporating external code (directly or indirectly) into a multi licensed or open source project, the following rules could apply:</p>

rustup</h3>
`rustup</code> packages all things rust, including rustc</code>,cargo</code>,rust-analyzer</code>, and the std</code> lib.</p>`