Looking for something?
A place, an idea, a question. Find the articles that interest you.
Articles to discover
Data breach: keep up the follow-upThe CNIL offers a timeline of checks to carry out after receiving a notification.
Travel and overnight stays: two measuresTourism figures for 2025 published by Eurostat in October distinguish between residents’ trips and overnight stays recorded by accommodation establishments. Luxembourg illustrates the difference.
Cosmetics: reading the open-jar symbolThe ‘12M’ marking starts from the date the product is opened. Keep track of it alongside the instructions for use to avoid confusing the time limits.
Antibiotics: preserving access to care tooThe WHO’s equity guide links prevention, diagnosis and access to treatment. Reducing inappropriate use is not the whole answer to antimicrobial resistance.
Dakar: a second chance on the tatamiThe repechage format gives young judoka several routes through the competition.
Biodiversity: counting species is not enoughA review of 423 studies examines the functions of biodiversity.
Organic: agreement on European rulesThe 20 October compromise distinguishes between the sale of imported organic products and the use of the European logo. Formal adoption is still pending.
Businesses: lasting five yearsLuxembourg’s industry has a higher five-year survival rate in STATEC’s 2026 profile. This finding calls for a distinction between the cohorts studied.
Arlon: what the posters tell usAt the Musée Gaspar, more than forty posters shed light on the messages addressed to the town. The exhibition can also be read through its printers, its intended audiences and the traces it leaves behind.
A correction must be visibleAP clarified its rules on AI in July. Its guidance, along with Reuters’ rules, also helps explain what a correction should give readers.
Wool: read before washingA jumper’s composition alone is not enough to choose how to wash it. Temperature, mechanical action and drying should each be checked against its care instructions.
Speech recognition: what errors should we count?The latest Common Voice release distinguishes between read speech and spontaneous responses. This difference sheds light on the error rates displayed by transcription tools.