Numérique Responsable

Yellow Lab Tools is a free online web performance analyzer. It audits a webpage for performance and front-end quality issues. And it's open-source!

Tumblr and WordPress are reportedly set to strike deals to sell user data to artificial intelligence companies OpenAI and Midjourney. The platforms’ parent company, Automattic, is nearing completion of an agreement to provide data to help train the AI companies’ models.

Towards environmentally sustainable computational science

Dead Simple Sites curates the most minimal sites on the web.

Les meilleures pratiques en matière de performances web sont consensuelles et largement acceptées, mais comment les valoriser d'un point de vue commercial ?J...

Due to their effects on performance, it's critical to understand how memory leaks in your Node.js apps occur, how to debug them, and how to prevent them.

Details of the publication

Comparing two identically designed websites, their weight, amount of HTML and CSS, rendering speed, and best practices.

Quels sont les enjeux d'une IA éthique ? L'intelligence artificielle interroge la notion de confiance et pose la question des usages.

Low numeracy affects half of adults in the UK. Many live with the limitations of not understanding numbers and the impact that has on everyday life.

Documentation for the Helsinki Design System

Écoconception Date de mise à jour : 13/02/2024 En lançant le projet de refonte de son site https://www.sentinelles-climat.org/ en 2023, […]

24 jours de web : Le calendrier de l'avent des gens qui font le web d'après.

Suite de l’article l’accompagnement des schémas d’accessibilité en intelligence collective (partie 1 sur 2). Sommaire : En quoi sont modifiés les processus de production numérique et la gestion …

L'intelligence collective est un levier pour déployer la démarche d'écoconception numérique dans une organisation : on vous explique pourquoi

L’écoconception numérique continue à se développer en France, notamment portée par le nouveau réglement général d’écoconception de services numériques (RGESN) développé au sein de l’État. Cette pratique consiste à se demander quoi faire pour réduire la charge environnementale (émissions de gaz à effet de serre, consommation d’énergie, d’eau, etc.) d’un site web et de son usage. Nous avons maintenant une certaine expérience du sujet puisque nous explorons à la fois en théorie et en pratique depuis 5 ans toutes ces questions. La refonte de Telecoop s’inscrit d’ailleurs dans la continuité de celle pour Commown avec Derek Salmon. Dans cette étude de cas, nous allons tâcher d’appliquer ce fameux RGESN qui se distingue par 8 grands axes : la stratégie, les spécifications, l’architecture, l’UX/UI, les contenus, le front-end, le back-end, l’hébergement. Nous allons reprendre point par point chacun de ces éléments.

Bonjour 👋 Bienvenue dans l’article #14 de Design 4 Change ! Nous sommes désormais 820 dans cette newsletter depuis 6 mois. Que tu me suives depuis le commencement ou que tu viennes de me découvrir, un grand merci de me lire ❤️.

Découvrez la conférence de Christophe Clouzeau, qui présente l'importance d'une écoconception et de l'impact environnemental du numérique.

Tired of manually checking spreadsheets, messy handovers, and expensive retrofitting? The all-new Stark Suite is a powerful combination of integrated tools that help you streamline your accessibility workflow. Now, you can create and test accessible software in record time. Whether you’re buildi...

Contrast lets you quickly run a AA or AAA type contrast report on all text across all your artboards at once. Simply select a few layers and run the command to see your report. See the code or file a bug on GitHub

Test the contrast of two colors to ensure the highest measure of readability using the Advanced Perception of Color Algorithm (APCA) and the classic ratio.

The APCA Accessible Perceptual Contrast Algorithim BASIC VERSION.

Achieving the UN Sustainable Development Goals (SDGs) demands a shift by industry, governments, society, and individuals to reach adequate levels of a…

Le SPN se nourrit des collaborations au sein de son réseau. Fin 2022, Game Partners évoquait le projet de construire un jeu sérieux sur le Numérique Responsable.

Smart coders can see the system is broken

Learn about optimizing your Azure carbon emissions.

Deep Learning (DL) frameworks such as PyTorch and TensorFlow include runtime infrastructures responsible for executing trained models on target hardware, managing memory, data transfers, and multi-accelerator execution, if applicable. Additionally, it is a common practice to deploy pre-trained models on environments distinct from their native development settings. This led to the introduction of interchange formats such as ONNX, which includes its runtime infrastructure, and ONNX Runtime, which work as standard formats that can be used across diverse DL frameworks and languages. Even though these runtime infrastructures have a great impact on inference performance, no previous paper has investigated their energy efficiency. In this study, we monitor the energy consumption and inference time in the runtime infrastructures of three well-known DL frameworks as well as ONNX, using three various DL models. To have nuance in our investigation, we also examine the impact of using different execution providers. We find out that the performance and energy efficiency of DL are difficult to predict. One framework, MXNet, outperforms both PyTorch and TensorFlow for the computer vision models using batch size 1, due to efficient GPU usage and thus low CPU usage. However, batch size 64 makes PyTorch and MXNet practically indistinguishable, while TensorFlow is outperformed consistently. For BERT, PyTorch exhibits the best performance. Converting the models to ONNX usually yields significant performance improvements but the ONNX converted ResNet model with batch size 64 consumes approximately 10% more energy and time than the original PyTorch model.

Today, we’re releasing a new guide explaining how to compel companies employing deceptive design to dodge data protection requests—like those for data deletion or access—to comply. It encapsulates five years...