wrappers, pgrx, pgdog, ParadeDB, Neon (Postgres in S3)

It's a cheap version for liquid glass with backdrop-filter: blur(10px) brightness(120%); :D

I didn't read it but look after the strategies: sequences with serial IDs, for update skip locked

So, it's a matter of your requirements. If you need massive scale and throughput (> 100k messages/sec), Kafka will be hard to beat. But if you're not operating at that scale, Postgres is hard to beat.

So, it's a matter of your requirements. If you need massive scale and throughput (> 100k messages/sec), Kafka will be hard to beat. But if you're not operating at that scale, Postgres is hard to beat.

No event listeners. No state tracking. No ARIA updates handled in JavaScript. I focused the button, and the tooltip appeared. I pressed the Esc key, and it disappeared.

• popovertarget connects the button to a popover element
• popovertargetaction specifies what should happen (show, hide, toggle as default)
• popover declares an HTML element can be targeted by popovertarget (ou .showPopover())

Note Floating UI already solves many of these issues. CSS anchor positioning should solve most of the layout stuff

A decision tree: https://res.cloudinary.com/indysigner/image/fetch/f_auto,q_80/w_1200/https://files.smashing.media/articles/getting-started-popover-api/5-popover-api-browser-support.png

ul.notes li {
    list-style-type: "Note: ";
    list-style-position: inside;
}

The ::marker pseudo-selector can be used to customize the rest. Generating content for markers is supported by Chromium and Firefox, but not by WebKit.

A list of symbols can be used with symbols(). The browser support is not great though. @counter-style can be used instead. It's Baseline Newly Available since 2023.

There is also the old ::before trick to set custom contents.

As summary:

Passer les PNG/JPEG qualité 90 à AVIF qualité 50 permet d'économiser au moins 75% de bande passante.

L'idée plus innovante est de compresser au préalable les ressources avant qu'elles soient utilisées.

[Précompresser avant de déployer] veut dire qu’on peut les compresser une seule fois, avec le niveau maximum, et demander à nginx de servir directement les fichiers pré-compressés. Zéro CPU à chaque requête, mais surtout un meilleur ratio au final, car on peut compresser plus fort.

En outre, Zopfli permet de compresser en .zip avec 3 à 8% d'efficacité en plus.

# Serve pre-compressed files generated at build time
gzip_static on;
brotli_static on;  # nécessite libnginx-mod-http-brotli-static

# Fallback pour les contenus non pré-compressés
gzip on;
gzip_vary on;
gzip_min_length 1024;
gzip_types text/plain text/css text/xml text/javascript
           application/javascript application/json
           application/xml image/svg+xml;

La compression Brotli permet de compresser à hauteur de 81% le HTML. La score des Web Core Vitals est passé de 70-85 à 99%.

The reset:

legend {
  padding: 0;
  display: table;
}
fieldset {
  border: 0;
  padding: 0.01em 0 0 0;
  margin: 0;
  min-width: 0;
}
body:not(:-moz-handler-blocked) fieldset {
  display: table-cell;
}

Great utils

1. A simple queue.json
2. Batching with group commit
3. Use a brokered group commit to eliminate contention over the queue object
4. HA brokered group commit to handle unfinished job or broker machine die

Similarly to the job system we built at work, it guarantees at-least-once delivery.

I don't know if the pattern becomes too complex to be viable.

A deep dive into ASCII rendering.

It seems great but I don't have time for it.

Mots-clés lié à display: flex

• auto : L'ancien roi. Contextuel, imprévisible parfois, il laisse le navigateur calculer la taille selon le modèle de boîte standard (display).
• min-content : "Je veux être aussi petit que possible." Le navigateur va essayer de réduire la boîte jusqu'à ce que le mot le plus long ou l'élément le plus large force la largeur minimale. C'est le "soft wrapping" ultime.
• max-content : "Je prends toute la place dont j'ai besoin." La boîte s'élargit pour contenir tout le texte sans aucun retour à la ligne, quitte à provoquer un scroll horizontal (à utiliser avec prudence !).
• fit-content : Le compromis parfait. C'est mathématiquement équivalent à min(max-content, available-space). La boîte s'adapte au contenu, mais s'arrête poliment si elle touche le bord du conteneur parent.
• stretch : L'élément s'étire pour remplir l'axe disponible. C'est souvent le comportement par défaut des flex-items ou des grid-items.

Math functions such as clamp(), min(), max().

Le pattern Repeat Auto Minmax (RAM) est usuellement donné aux snippets qui génère une grille responsive sans aucun breakpoint grid-template-columns: minmax(300px, 1fr)

The author creates a project to gather statistics about the top most starred projects on Github or the most downloaded packages on NPM.

1. 9-27% of JS/TS projects declare themselves to be ES Modules
2. Less than 6% of JS/TS files declare that they are ES Modules via the .mjs, .cjs or .mts file extensions.

Some ideas:

1. kill .mjs, .cjs and .mts should be replaced by the type="module" in package.json. Let's stick to .js, .jsx, .ts and .tsx
2. Make type="module" the default and warns when the type is not set to module.
3. We should upgrade the most common libraries used by the community to ES Modules
4. The NPM registry can require an explicit module field on new packages, making it clear when a package intentionally uses CommonJS.
5. NodeJS can officially drop support for require and module.exports in a future version, creating a bit more pressure to migrate.

appearance: base-select, so it's too early yet.

Here's the thing: 99% of companies don't need them. The top 1% have tens of millions of users and a large engineering team to match.

The fun thing about Postgres is there is already an extension for that: PostGIS, Full-text search, JSONB, TimescaleDB, pgvectorm, and many for AI

Each database add hidden costs: backup strategy, monitoring dashboards, seceurity patches, on-call runbooks, failover testing.

SLA math: Three systems at 99.9% uptime each = 99.7% combined

compared to 25M+ with cloud services.

So in order to set up one, the data center needs power, cooling, servers and software. The post goes in-depth about it. The list of softwares available is a great inspiration.

All of our storage arrays use mkv. The main array is 3PB of non-redundant storage hosting our driving data we train on. We can read from this array at ~1TB/s, which means we can train directly on the raw data without caching. Redundancy is not needed since no specific data is critical.