photog.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A place for your photos and banter. Photog first is our motto Please refer to the site rules before posting.

Administered by:

Server stats:

245
active users

#diffusion

1 post1 participant0 posts today

Voyons voir si les #travailleureusesdelart arrivent à s'organiser en dehors d'IG 😜 . Existe-il une cartographie des lieux formels et (surtout) informels de #diffusion, d' #exposition, de #concert, de #spectacle , de salons (ex : micro-édition), de #festivals, de #residence , de #discussion ,... ? Lieux publics, privés, #associations , chez l'habitant, #mobilitecroisee (échange d'ateliers-logements), etc... ? Sinon mettons ça en place ! #entraide #cooperation #cartographie #art

#introduction
Docteure en #biologie et ex-Ens de Lyon, je suis impliquée dans la #diffusion et le #partage de la #culturescientifique.

Au boulot, je suis cheffe de projets d’événements culturels et scientifiques à l’#UniversitéParisSaclay, où j’essaie d’éclairer et de créer du débat sur les enjeux socio-politico-environnementaux pour la communauté étudiante et le « grand public ». Je suis notamment commissaire de l’expo À la limite – Innover à la mesure du monde – 2055 (lumen.universite-paris-saclay.) (avec @gblanc05 dans le conseil scientifique.)

Côté perso, je suis rédac’chef de Papier-Mâché, un site de #vulgarisation par les pairs de publications scientifiques. L’objectif : que chaque personne puisse comprendre les résultats de la recherche, contextualisés.

Plus globalement, j'essaie de faire ma part pour que la société prenne le chemin d'un avenir basé sur le respect des #limitesplanétaires et sur le bien-être de toutes les populations vivantes humaines et non-humaines (et pas juste de quelques hommes blancs riches etc.).

Lumen · À la limite - Innover à la mesure du monde - 2055Une catastrophe climatique et écologique est en cours. L’innovation – au sens d’innovation technologique – est brandie comme solution miracle, omettant un facteur fondamental : tout est lié. Le climat, l’utilisation de nos sols, les ressources disponibles, chaque espèce vivante, nos modes de vie, la répartition des richesses, etc. Innover en trouvant une solution pour l’un qui ne prenne pas en compte les autres est un non-sens scientifique. L’exposition nous projette en 2055, sur une planète Terre où les environnements et les populations se portent bien. Cet avenir, fictif, se veut « désirable » selon les critères d’équité, de justice sociale, de respect et de bien-être de toutes les populations humaines et non-humaines. Et, au-delà du réchauffement climatique, ce futur s'inscrit à l’intérieur des limites du système Terre. Dans cet avenir encore possible, mais de moins en moins probable, que veut dire innover ? Quelles innovations, sociales et technologiques, y auraient leur place ? En particulier, quelles questions se posent aujourd’hui pour les métiers de la recherche, des techniques et de l'ingénierie ? Prenant appui sur des œuvres de science-fiction, l'exposition propose un espace de débat pour tenter de construire collectivement un avenir qui donne envie, avec l’ambition d’être source d’optimisme. Le parcours commence par dresser un portrait actuel de l’innovation, en montre les coûts et conséquences, et présente le modèle des limites planétaires. Puis, une rupture temporelle permet d’interroger des exemples vertueux pour et dans le futur esquissé, tout en nous confrontant à nos choix. Des dispositifs interactifs et ludiques jalonnent le voyage, mis en valeur par deux partenariats avec l’École Estienne pour la production d’illustrations scientifiques. Visites guidées sur demande à eleonore.peres[at]universite-paris-saclay.fr Animations :- mur participatif "Raconte-nous ton Université désirée de 2055" : tous les jours dans le Forum- création participative de nano-fictions : tous les jours dans le Hall d'exposition Informations pratiquesExposition présentée du 02 janvier au 16 février 2025. Adresse : Lumen - Université Paris-Saclay, 8, avenue des Sciences 91190 Gif-sur-YvetteHoraires : Du lundi au vendredi de 8h30 à 22h30 et le weekend de 10h à 19h Une exposition produite par la Direction des Bibliothèques, de l’Information et de la Science Ouverte (DiBISO) en partenariat avec la Diagonale · Université Paris-Saclay et le Design Spot. Avec le soutien de CentraleSupélec, de l’ENS Paris-Saclay et de l’Université Paris-Saclay.
Continued thread

#WritersCoffeeClub Ch 12 Nbr 15 FEV— Have you used any form of AI to help in creating your work? Do you have a red line? Where is it? Full Essay Version

I've been around long enough that I remember when the #Altair 8080 (the first commercial personal computer) got a full page ad in Scientific American. My BFF and I talked about it after school (11th grade, I think?) for hours. I had my own slimline phone on my own extension. I was a privileged kid. The #computer club had a teletype connection to a computer that ran #BASIC. We also had access to an #IBM mainframe assembler using #MIS (pencil in) punch cards. A friend built an #IMSAI 8080 with 256 bytes of memory and it played songs we could listen to by tuning to part of the FM band influenced by the frequency of bits passing through the memory buss.

Why do I bring this up? The first things we programmed (after ping-pong on the toggle switch lights, Star Trek, and later adventure games) where types of artificial intelligence. We've been "teaching" computers to "think" for us since we could rip them away from the greedy money grubbers who wanted to do things like accounting and payroll. Current concepts of AI rely on pattern matching against databases, in very simplified terms. Early #AI were procedural, and if it solved your math problem, spelled your word correctly, or found data in the noise, it was intelligent. It is inescapable that unless we write longhand or type on a typewriter, that we have used what someone terms AI.

Anybody remember Clippy?

The key to this question, and how I am going to take it, is to focus on the word "creating." To me, that highlights generating something. Text. Graphics. Though this implies #genAI (the current thing, something like #Diffusion or #ChatGPT). It also means graphic tools that I use to remove and add image elements to photos, or otherwise intelligently improve them in ways I could imagine but was previously incapable due to lack of talent, practice, or understanding of the unenhanced tools.

As for text, I remember once writing a chatbot psychologist. IF-THEN-ELSE. In BASIC, back in high school or it may have been in an equivalent language on a DEC-10 at university. It did not pass the Turing Test, but it could lift one's mood, which could help with me "creating" work!

I've no interest in using something like chatGPT to build a story from prompts. It would be the program's story, not mine. It would be a Cliff Notes version of an unwritten story, summarized, never provided a soul, before even being written. Because these things work by pattern matching and predicting what would be the next word (based on an average of everybody's writing or specific genres or authors), it would be pure statistical fabrication—not even fabulation, which implies creative fantasy. It would be a story with none of the emotion, feeling, or meaning I'd put into it. It would be phantom chatter masquerading as wit through statistical randomness. No enlightenment possible, except by accident.

As an author, why would I even try to replace myself, or create the tools for others to try to do the same? I suspect people that do, do so as I wrote above because they are "incapable due to lack of talent, practice, or understanding of the tools." I'll add, also laziness, though I tar myself with the same brush as I wrote above.

Never discount greed. Authors cost money. Computers cost once.

In any case, the AI tools built into the software I do use barely save me time as it is. I can't start a dictation with the 1st person personal pronoun on any of my systems. Statistically, more people start dictation with "Hi" then "I," so I always have to correct that. Thanks to auto-incorrect, even the words I spell correctly (few enough as it is) get changed. In this very article, the tool insisted I had a "teletype connection to a commuter!" Not only do I have to be aware of my own tendency toward grammar issues and typos, I have to proof for AI typos, including chaining typos since the tools keep track of some meaning and can spawn a hellacious brew of misconception and embarrassment should it slip out to readers. I know what I mean to write. Sometimes I find myself overruled by my own tools.

Have I used AI? Not genAI. Procedural tools, yes. Not for willingly generating text.

Red line? Yes. No generating story or text. What's the point?

[Author retains copyright (c)2024 R.S.]

#BoostingIsSharing and #CommentingIsCool

#fiction #romance #fantasy #sf #sff #sciencefiction #writing #writer #writers #author #writingcommunity #writersOfMastodon
#RSdiscussion #theComingLaborApocalypse

I'm humbled and awed to be in the company of so many accomplished artists, researchers, practitioners, guardians, technologists and designers as part of the Fantastic Futures Conference #FF24 Key Speakers lineup - presented by #NFSA on behalf of the #AI4LAM community.

Peter-Lucas Jones, and the work of #TeHiku Media in preserving and protecting Indigenous speech data of #TeReo is an effort I have long admired. Associate Professor Kirsten Thorpe's work at the Jumbunna Institute at University of Technology Sydney also centres on data sovereignty for Indigenous data, from an archival perspective.

Eryk Salvaggio - @CyberneticForests - also an alum of the #ANU School of #Cybernetics, like myself does incredible work in interrogating social and cultural impacts of #AI, especially in #ComputerVision.

Kartini Ludwig's use of creativity at Kopi Su Studio to empower artists runs counter to the prevailing norm of scraping the internet to build #LLMs and #diffusion models.

And Associate Professor Sydney Shep's trans-disciplinary work in book history and print culture, as well as her practice as a letterpress printer and bookbinder, is a fascinating exploration of how our cultural histories shape our futures.

Together, we hope to convene conversations that help shape the future of #AI and #ML within the #GLAM sector.

Huge thank you to Keir Winesmith and @ingridbmason for the opportunity, and to Ashlinn H. who I know is doing incredible work behind the scenes.

nfsa.gov.au/fantastic-futures-

Composite photography, statistical correlations and eugenics were all created by the same man — the British sociologist Francis Dalton - in the 19th Century. AI Images Class 10 examines the legacy of this common root as it has reconvened in image synthesis — and the principles of categorization and representation they enact. m.youtube.com/watch?v=0sK_jAOm #aiimages #aiethics #diffusion