Hallo,

(DeepL-enabled somewhat readable English translation under the French version below)
💬💬💬 Expressions idiomatiques 💬💬💬
  • Mediawijsheid : terme hollandais désignant l'éducation aux médias (Source: Mediawijze)
  • "Audible command filtering" : a technique to prevent Alexa from waking up “as part of a broadcast watched by a large population (such as during a popular sporting event),” annoying customers and overloading Amazon’s servers with millions of simultaneous requests (Source: Bloomberg)
  • Keyword void : néologisme décrivant l'absence de résultats pertinent lors d'une requête sur un moteur de recherche, et la présence de contenu uniquement produit de manière orientée (commercialement ou politiquement) (Source: kottke)
  • Skill-squatting : une technique d'utilisation des mots homophones verbalisées par les utilisateurs d'un assistant vocal comme Alexia fin de faire appel à des applications aux noms très proches, et qui peuvent déboucher sur le piratage à distance des données (Source: Ars Technica).
🤖🤖🤖 Le coin des algorithmes 🤖🤖🤖

🔪 Le projet "Anatomy of AI" porté par Kate Crawford and Vladan Joler et publié par SHARElab vient d'être mis à jour. Réalisé pour une exposition à venir au V&A Museum de Londres, il consiste en un diagramme qui propose rien de moins qu'un "anatomical case study of the Amazon echo as a artificial intelligence system made of human labor", accompagné d'un essai décrivant les implications d'un tel décorticage. Un bon exemple du recours croissant au design comme moyen d'enquête ou de restitution des conclusions (qui fait écho à ce séminaire à Londres sur le thème de l'explication de l'automatisation et qui indiquer "The project combines academic research with design methods to investigate how to make data-based, automated decision-making systems understandable to people"). Cela me semble intéressant de constater que c'est bien souvent le fait d'aborder de tels sujets complexes qui amène des chercheurs en sciences humaines et sociales à avoir recours aux designers.

🌶️ Mathilde Buenerd a mis en ligne plusieurs principes tirés de son projet de diplôme mené au sein du Master Media Design de la HEAD – Genève où j'enseigne. Il est intitulé "Chili Design", en référence à la proposition de Frédéric Kaplan il y a quelques années de concevoir des interfaces "piquantes", c'est à dire stimulantes, surprenantes et provoquant la surprise chez les utilisateurs alentours. Ce travail de Mathilde a débouché ensuite sur la conception d'une app, Jouska, qui collecte et analyse les échanges de messages textuels et lui donne un feedback émotionel: "The app allows users to have a better understanding of the way they talk, and in the end, of one of the others. This way, they can develop a form of immunity at targeted solicitations from advertisers, politicians, or noxious relations.". Voir cette vidéo pour plus d'information.
 
🌱🌱🌱 Fragments 🌱🌱🌱

👀 Onglets ouverts dans le navigateur : cet article à propos la circulation des emoji/stickers/emoticons sur les médias sociaux chinois, un essai sur la mélancolie occidentale envers le futur, cette page web expliquant le propos et l'importance des "discard studies", un entretien avec Jérôme Denis sur son dernier bouquin qui concerne les multiples opérations souvent invisisbles réalisées sur les données, un autre entretien sur l'évolution de la voix humaine, un article académique sur l'analyse des promesses technologiques et qui aborde le cas du OLPC1.
💬💬💬 Digital Parlance 💬💬💬
  • Mediawijsheid : dutch term referring to their own perspective on "media literacy" (Source: Mediawijze)
  • "Audible command filtering" : a technique to prevent Alexa from waking up “as part of a broadcast watched by a large population (such as during a popular sporting event),” annoying customers and overloading Amazon’s servers with millions of simultaneous requests (Source: Bloomberg)
  • Keyword void: "a situation where searching for answers about a keyword returns an absence of authoritative, reliable results, in favor of “content produced by a niche group with a particular agenda." (Source: kottke)
  • Skill-squatting: "the potential to exploit some of the idiosyncrasies of voice-recognition machine-learning systems for malicious purposes has grown as well. (...) Thanks to the way Alexa handles requests for new "skills"—the cloud applications that register with Amazon—it's possible to create malicious skills that are named with homophones for existing legitimate applications. Amazon made all skills in its library available by voice command by default in 2017, and skills can be "installed" into a customer's library by voice. "Either way, there's a voice-only attack for people who are selectively registering skill names" (Source: Ars Technica)
🤖🤖🤖 Algo corner🤖🤖🤖

🔪 "Anatomy of AI" is a project led by Kate Crawford and Vladan Joler and published by SHARElab. its authors have just been released good content on the website. Produced for a forthcoming exhibition at the V&A Museum in London, it consists of a diagram that proposes nothing less than an "anatomical case study of the Amazon echo as a artificial intelligence system made of human labor", along with an essay describing the implications of such shelling. Beyond the analysis of algorithms, it's a good example of the increasing use of design as a means of investigation or restitution of conclusions (which echoes this seminar in London too, on the theme of explaining automation and which indicates "The project combines academic research with design methods to investigate how to make data-based, automated decision-making systems understandable to people"). It seems interesting to me that it is often the fact of dealing with such complex subjects that leads researchers in the humanities and social sciences to resort to designers.

🌶️ Mathilde Buenerd published design principles from her diploma project she conducted in the Media Design master at the Geneva School of Art and Design where I teach. It's called "Chili design", in reference to Frédéric Kaplan's proposal from few years ago about "chili" technologies – stimulating, disrupting, surprising for users. Mathilde's work led to the design of a an app called Jouska, that collects and analyses textual exhcanges in order to gives users an emotional feedback: "The app allows users to have a better understanding of the way they talk, and in the end, of one of the others. This way, they can develop a form of immunity at targeted solicitations from advertisers, politicians, or noxious relations.". See the video for more information.
 
🌱🌱🌱 Fragments 🌱🌱🌱

👀 Tabs currently opened in my web browser : this article about the circulation of emoji/stickers/emoticons on chinese social media, an essay about Westerners' melancholia towards the future, this webpage that explains the what and why of "discard studies", an interview with Jérôme Denis about his latest book on the multiple invisible operations necessary for Big Data usage, another interview about the evolution of human voice, an academic article about "charismatic technologies" using the case of OLPC1.

++
– nicolas
EOF