Artificial Intelligence On 8 May 2026, Thariq Shihipar, an engineer on the Claude Code team at Anthropic, published a manifesto that’s been circulating ever since: drop markdown, switch to HTML for your specs, implementation plans and reports. The piece is clever, well written, and structurally an act of strategic marketing dressed up as personal experience. On a narrow subset (throwaway editors, interactive playgrounds, prototypes with sliders), the author is right; on everything else, his blanket enthusiasm conceals six blind spots. Token cost waved away, semantic HTML flipped into presentational HTML, loss of reviewability, the maintenance paradox that traps the user inside the production loop, reading that gets skimmed rather than scrutinised, an attack surface opened up by indirect prompt injection. An analysis of a poorly disguised strategic signal, plus a pragmatic decision matrix for sorting out where HTML deserves to be adopted and where markdown remains, by design, the right pivot format.
Artificial Intelligence Every session, Claude Code reopens your codebase like a visitor who has never set foot in it. It unfolds the README, fires off its greps, opens whole files to reconstruct an architecture it had already pieced together the day before. Understand-Anything, an open-source project released in March 2026, offers another path: index the codebase as a local knowledge graph, versioned like a lock-file, queried on demand by the agent instead of being re-read on every turn. This isn’t a token optimisation, it’s an architectural shift that extends the “RAG is dead, long live the Agent” thesis right into the code itself. Multi-agent pipeline, comparative economics, the limits of a project just six months old: what this break changes for Claude Code and for your workflow.
Softwares For twenty years, WordPress has let every plugin name its options and tables however it likes. The add_option() function sets autoload = ‘yes’ by default, with no warning, no documentation urging caution. The register_uninstall_hook() hook exists but is mandatory for no one. The uninstall.php file is optional. And the API provides no discovery mechanism letting the administrator find out, after the fact, which plugin a given option or table belonged to.
Hardware For more than a decade on Sony A7R bodies, one observation has imposed itself with irritating regularity: skin tones drift towards green, and no amount of post-production correction really sets things right. This Sony green, as the photo community calls it, isn’t the whim of a Canon nostalgic. It’s a technical reality written into the colour matrix that Sony has been supplying to RAW-conversion software for years. The most paradoxical part? Sony bought Konica Minolta in 2006, inheriting along the way decades of expertise in the natural rendering of skin, a heritage Minolta had spent fifty years building, and that Sony seems to have filed away in a cupboard. While Fujifilm built its entire marketing around its Film Simulations and Canon kept flattering flesh tones by default, Sony was optimising its matrix for DxOMark test benches. Laboratory truth rather than human truth. Mathematical excellence comes at a price, and it’s your portraits that pay it.
Artificial Intelligence Converting a technical PDF into RAG-ready markdown isn’t just a matter of running LlamaParse over it. On a 51-page chapter of the Funktionsrahmen MSS60, the Siemens documentation for the engine control unit of the BMW M3 E92 and M5 E60, I discovered that the interesting work begins after the automatic parsing. The semantic decisions that really condition retrieval quality, preserving the source language rather than translating, breaking dense tables into individual subsections, supplementing diagrams with French prose and pseudo-code, no parser will make them for you. This LLM-assisted work is done only once per document, its marginal cost per query is zero, and its upfront investment pays for itself in user-hours from the very first week of use. Field notes from two evenings of conversion that turned my understanding of what really makes a RAG pipeline worthwhile on its head.
E-invoicing FICOBA in January, ANTS in April, Mentor at the end of April, impots.gouv flagged in early May: four sovereign vaults gutted in ninety days. On 1 September 2026, the State will force ten million businesses to push the entirety of their invoicing flows through one hundred and fifteen private platforms it has accredited. Sold as a fight against fraud, the reform in reality builds a complete graph of the private French economy, hosted at around a hundred operators, over which the State decides everything and answers for nothing. It’s the French invention of an unprecedented mechanism: responsibility without culpability, in which political risk is transferred to private operational risk with no transfer of the decision. We haven’t multiplied the targets: we’ve merged them.
Artificial Intelligence We spent two years debating the model. GPT-5 versus Claude 4.5, Gemini versus Mistral, million-token context versus prefix cache. Meanwhile, the elephant in the room stayed invisible: 90% of production RAG pipelines are swallowing a mush of badly extracted pixels, then act surprised when the LLM hallucinates. The problem isn’t in the brain. It’s in […]
Artificial Intelligence In 2026, generating a photorealistic image costs a cent, indistinguishable from a real photograph. Four professional markets are collapsing in silence: stock photography, e-commerce packshots, editorial illustration, routine retouching. Beyond the markets, it’s the social institution of photographic proof, a hundred and eighty years old, that is cracking: courts, insurance, journalism, elections, privacy. A cold mapping of the models that matter, of copyright laundering act II, and of what regains value when every image becomes suspect by default.
Hardware Linky, Itron and Landis+Gyr meters equipped with a TIC output let you integrate your real-time electricity consumption data into Home Assistant. Two technical solutions face off: the ESP32-PoE-ISO with the EthTInfo shield favors the robustness of a wired Ethernet PoE link, while LiXee’s ZLinky v2 bets on simplicity with self-powered Zigbee communication. This technical guide compares the architecture, installation, budget and reliability of the two approaches to help you choose the solution suited to your home-automation setup. Whether you have one or several meters to connect, find out which method best matches your priorities between maximum performance and ease of deployment.
Softwares After four years of patience in the face of the promise of a “modern PrestaShop”, I migrated my entire client fleet to WooCommerce. The trigger didn’t come from a catastrophic bug, but from the loss of confidence in the project’s trajectory: governance blur between “Project” and “Edition”, the absurd coexistence of two template engines (Smarty and Twig), and updates that turn every minor maintenance into a risky operation. When a product-page change takes 4 hours on PrestaShop versus 1.5 hours on WooCommerce, the opportunity cost for the client becomes unsustainable. I developed a WooCommerce Multisite + API-first architecture (via my StockSync tool) that offers more robustness than PrestaShop’s native multistore, with no dependence on direct SQL writes. PrestaShop may have gained Symfony, but it has lost its field ambassadors.