Sarvatra Bridge System Gives Indian Army Rapid River-Crossing Capability

Water bodies have historically posed major challenges for military operations worldwide, often determining the outcome of battles. India's 1971 war highlighted the importance of river-crossing tactics, including innovative heliborne assaults. The indigenously developed Sarvatra bridge system addresses this challenge by enabling armoured formations to quickly cross water obstacles. The system is designed to ensure swift and uninterrupted movement of heavy military units during operations. Its development marks a significant step in India's self-reliant defence capability for battlefield mobility.

Developer Shares Nine-Year Struggle to Run Digital Signage Software on Raspberry Pi

A software developer has documented nearly a decade of attempts to make his open-source digital signage application, garlic-player, run smoothly on Raspberry Pi hardware. Early efforts on the Pi 3 around 2017–2018 were hampered by the lack of zero-copy display output in Qt, causing video stuttering even at low CPU loads, despite partial workarounds using a GPU-based tool called PiOmxTextures. A breakthrough for smooth HD playback was eventually achieved in 2018, but it broke WebView and multi-zone layout support — two key features of the application — with no useful error output to diagnose the cause. The 2019 release of the Raspberry Pi 4 brought hardware improvements including dual HDMI ports, but hardware-accelerated video decoding was only available on the 32-bit OS, while the more capable 64-bit Raspbian remained in beta without acceleration. 4K video support also proved unreliable, working only in VLC under specific conditions, leaving the developer repeatedly forced to choose between a modern operating system and functional hardware decoding.

Serial killer’s preferred poison revealed: How Chhattisgarh man left trail of bodies

Open-Source Tool CauterRule v0.3.1 Fixes Zero-Pass Corpora With Data Repairs, Not Model Changes

CauterRule, an open-source sidecar that converts repeated AI agent failures into reusable rules, released version 0.3.1 after investigating two corpora that scored zero passes in the previous cycle. The 'adapters' corpus failed entirely because the reference pool contained no framework-specific trajectories, meaning the model was producing valid candidates that had nothing to be compared against. Similarly, the 'raw/ci' corpus suffered from four compounding data bugs, including a script that captured log headers instead of actual failure text and a reference pool padded with irrelevant trajectories. After fixing the underlying data issues — adding domain-matched reference signatures and correcting the log collector — the adapters corpus improved from 0/60 to 60/60, and raw/ci rose from 0/110 to roughly 21–26 out of 47 valid entries. The key finding is that a zero-percent corpus signals a data quality problem to investigate, not necessarily a model limitation to address.

DEV Community Launches Weekly Meme Monday Thread

DEV Community hosts a recurring weekly feature called Meme Monday, inviting members to share humor-based content. The cover image for the latest edition was sourced from a previous thread submission. DEV emphasizes that it operates as an inclusive platform for all its members. To maintain community standards, moderators will downvote any humor deemed to be in poor taste.

How Startups Can Survive AWS Regional Outages Without Doubling Cloud Costs

Full multi-region infrastructure redundancy is the standard recommendation for surviving AWS outages, but it can double cloud bills and is impractical for most startups and mid-sized businesses. A hybrid approach combining intra-region Multi-AZ automation with a lightweight cross-region backup strategy offers a more affordable alternative. The method relies on managed services like AWS Fargate, Lambda, and RDS Multi-AZ for automated local failover, while storing only data — via daily RDS snapshots and S3 Cross-Region Replication — in a secondary region rather than running live infrastructure there. Infrastructure-as-code tools like Terraform or CloudFormation are used to define the backup environment, with compute resources spun up only when a real disaster occurs, avoiding the cost of idle servers. This trade-off accepts a short recovery delay during a worst-case regional failure in exchange for significantly lower ongoing costs, with cold backup storage estimated at around $5 per month for a 100 GB database.

How to Build an AI Agent Harness From Scratch in Your Code Repository

A developer tutorial on DEV Community walks through building an AI agent harness — a structured set of files, rules, and conventions stored directly in a code repository. The harness gives an AI agent everything it needs to operate autonomously, including how to start, what constraints to follow, how to verify task completion, and how to hand off work to the next session. The guide uses a simple command-line ledger project as a sandbox, constructing key files such as CLAUDE.md for agent instructions and ARCHITECTURE.md for system design details. A startup script is also added to handle dependency installation and verification in a single step, forming the foundation of the agent's operating loop. The core principle is that keeping task criteria inside the repository — rather than in ephemeral chat messages — ensures consistent decision-making across agents and developers alike.

How a Provider Policy Linter Can Catch Silent AI Gateway Failures in CI

Multi-provider AI gateways can fail quietly when model aliases, routing rules, or pricing data change without triggering obvious errors in the SDK or dashboard. A provider policy linter is a lightweight CI gate designed to catch these issues before a release ships, by validating assumptions around models, endpoints, pricing, and billing groups. The approach involves five core checks: confirming pricing source freshness with dated references, verifying model route existence, enforcing explicit endpoint compatibility declarations, clarifying billing-group context in examples, and requiring named fallback policies. Reference pricing data for this guide was sourced from AIWave's public endpoints on September 14, 2026, reflecting a 64-model snapshot dated September 10, 2026. The linter is intended to complement, not replace, live monitoring, and its policy configuration file should be version-controlled alongside SDK code.

Adding One String Key to a PHP Array Can Silently Consume 25 MB Extra Memory

A developer refactoring legacy PHP code discovered a significant memory cost tied to how PHP internally handles arrays. A packed array of one million integers uses about 16.8 MB in PHP 8.4, but adding even a single string key forces PHP to convert it into a full hash table, jumping memory usage to nearly 42 MB. This conversion adds roughly 25 MB of overhead and is irreversible — unsetting the key does not reclaim the memory. The same penalty applies when filling an array with sequential integer keys in reverse order, as PHP treats out-of-order insertion as a hash table rather than a packed list. The findings highlight how PHP's internal array optimizations can silently break under common coding patterns, with real consequences for memory-heavy workloads like processing large database result sets.

BCB recalls Hasan Mahmud from Kent stint ahead of packed international schedule

The Bangladesh Cricket Board has withdrawn fast bowler Hasan Mahmud from his county cricket stint with Kent. The decision was made to ensure the pacer is well-rested before a demanding international winter for Bangladesh. Kent are currently competing for promotion, meaning the withdrawal comes at a significant moment for the county side. The BCB prioritised national duty commitments over Mahmud's continued involvement in the county campaign.

HTMP Library Brings React-Like Reactivity to a Single HTML File at 2.9 kB

A lightweight library called HTMP enables reactive UI behavior — including counters, todo lists, and live form updates — directly inside a single HTML file with no build tools or package managers required. The zero-dependency engine weighs just 2.9 kB gzipped and works in any browser, including via a local file:// URL. HTMP is built on three core concepts: an HTML template parsed by the browser's native DOMParser, a reactive state object powered by JavaScript's Proxy, and plain event-handler functions. It offers a lifecycle API with mount, unmount, remount, and destroy methods, making it suitable for single-page apps and embeddable widgets. The library is aimed primarily at PHP, WordPress, and legacy-stack developers who need client-side interactivity without adopting a full JavaScript framework.

Largest known Roman mosaic, beneath Baths of Trajan, opens to the public

Article URL: https://www.theartnewspaper.com/2026/09/14/largest-roman-mosaic-opens-to-the-public Comments URL: https://news.ycombinator.com/item?id=49696132 Points: 4 # Comments: 0

Step-by-Step Lab Guide: Build an Active Directory Domain with Hyper-V

A hands-on IT lab project walks users through setting up a Windows Server Active Directory Domain Services environment using Hyper-V virtualization on a Windows 10 or 11 machine. The project covers creating and configuring a domain controller, promoting it, and adding a domain member server using two Windows Server 2022 Evaluation Edition virtual machines. Users need the Professional or Enterprise edition of Windows 10/11 and at least 16 GB of RAM on the host machine to run the setup. The lab is designed to simulate a real-world scenario, such as a company establishing centralized authentication for a new branch office. The entire exercise is estimated to take approximately 45 minutes to complete.

Guide Outlines Best Practices for Writing Effective Software Design Documents

A new guide published on Refactoring English offers practical advice for software developers on writing effective design documents. The article focuses on structuring design docs in a way that clearly communicates technical decisions and project scope. It aims to help engineers improve collaboration and reduce ambiguity during the software development process. The resource was shared on Hacker News, where it attracted initial reader interest.

GitHub Copilot Billing Gap Sparks Proposal for Project-Level AI Cost Tracking

Enterprises using GitHub Copilot face a significant challenge in accurately attributing AI usage costs across multiple projects, customers, and internal units, since current billing only aggregates data at the user level. This makes project-based cost accounting, customer chargebacks, and departmental budgeting difficult or impractical. A proposal raised in GitHub Community Discussion #206110 suggests allowing Copilot clients to attach a configurable billing reference — defined via a small file at the workspace or project root — to AI usage based on the active project context. The identifier would be organization-defined, enabling alignment with existing internal accounting structures such as cost centers or project codes. If adopted, the feature would expose this reference as an additional dimension in AI usage reporting and APIs, giving technical and financial leaders clearer visibility into the true ROI of AI investments per project.

How a Caffeine Tracker Uses Decay Math to Show What's Still in Your System

A developer built a manual caffeine tracker that estimates residual caffeine in the body using an exponential decay formula based on a default five-hour half-life. Unlike a simple daily total, the tool sums the decayed residuals of every logged drink, meaning a late-night coffee from the previous day can still show up in the current estimate. Records are kept for 48 hours rather than reset at midnight, preventing the misleading scenario where a low daily total masks significant leftover caffeine. The tracker includes a Chart.js projection of residual levels over the next 24 hours, sampled at 30-minute intervals, with a configurable 50 mg sleep threshold line and a separate 400 mg daily reference. The project's key design principle is keeping dose, time, half-life, and display thresholds as distinct variables rather than collapsing them into a single opaque figure.

Developer releases open-source CeX UK price tool after bug inflated £30 case value to £136

A developer building an eBay pricing tool discovered a bug that incorrectly attributed a £136 CeX cash offer — meant for a Raspberry Pi 500+ — to a £30 Argon ONE Pi case. The error stemmed from CeX's Algolia-powered search returning fuzzy, mismatched results rather than exact product matches. To address this, the developer built and published a standalone Python library called cex-uk-prices on GitHub under an MIT licence, which scores search results for relevance and discards weak matches below a 0.6 threshold. The tool queries CeX's Algolia index directly, since the main website returns 403 errors to server-side requests, and retrieves sell prices, cash and voucher trade-in values, and stock availability. The developer cautions that the library relies on an unofficial endpoint that may break if CeX updates its front end.

Microsoft releases 37-page AI code of conduct amid growing safety concerns

Microsoft has published a 37-page 'humanist AI code of conduct' asserting that people matter more than AI. The document states that AI models are not conscious and should not be designed to imitate consciousness. The release comes amid broader industry concerns that AI development may be outpacing humanity's ability to safely deploy and control increasingly complex systems. Anthropic CEO Dario Amodei recently called for a coordinated slowdown in AI development over the weekend. Microsoft's code also rejects the notion of granting AI legal personhood.

3,500-year-old British timber monument is drying out underground

Rahul Gandhi to Hold Student Outreach Event 'Chhatron Ki Goonj' in Indore on Sept 19

Congress leader Rahul Gandhi has invited students to participate in 'Chhatron Ki Goonj', a dialogue event scheduled for September 19 in Indore. The initiative is designed to give students a platform to raise concerns about the education system. Gandhi has previously conducted similar sessions in cities including Kota, Dehradun, Prayagraj, and Pune. The announced visit has triggered a political exchange between Congress and BJP leaders. Gandhi stated that the event aims to collectively push for meaningful reforms in the education sector.