@fyrfli Yes, it would appear populist homophobia has kept many politicians in power.
https://www.fordfoundation.org/news-and-stories/videos/queer-african-history-must-be-told-and-documented-featuring-selly-thiam-of-none-on-record/
Posts in English
âBy donning a second-hand hat, a 74-year-old man is about to become monarch of a small island kingdomâ
As expected, the best reporting on todayâs coronation comes from @thecontinent
While many mentioned atrocities are well-known, this part of erased African history is rarely talked about:
âWhile England has records of who lived in what village in 1068, its empire was quick to destroy the histories of those it colonised. These could then be overwritten with Protestant orthodoxy and white exceptionalism, banning things like queer love, once part of Africas cultural fabric.â
https://www.thecontinent.org/
#Coronation
@troed Or run in parallell for a while to evaluate đ
Why I love Akkoma. People are asking more often why I prefer Akkoma over Mastodon, so here is a list.
Note that Iâm running Akkoma as a single-user instance.
âą Significantly less resource-demanding than Mastodon if you have many followers and/or follow many people. This seems to apply to memory, processing power as well as storage. I used to run Mastodon and it slowed down my server significantly. Doesnât need sidekiq for example and doesnât store âexternalâ media locally. Fewer dependencies.
âą More community-driven development. (Mastodon is much more of a one-man show.)
âą Greater flexibility (more file formats, linking of text, formatting of text). Can do markdown, HTML, BBCode and even Misskey Markdown.
âą Better handling and visibility of comments in threads. I see the answers in a hierarchy and I see - usually - everyone who answered in a thread. Mastodon is notoriously bad at this.
⹠Has a built-in free-text search engine⊠which can also be extended with Meilisearch or Elasticsearch - but of course those would make everything more memory-intensive
âą More freedom (e.g. number of characters in posts/profile, and control of front-end appearance, can make lists even with people I donât follow, etc.)
âą Have had âquote-boostâ since forever (not that I really use it).
âą Is a fork of Pleroma that was released some time before Mastodon. The underlying philosophy (more like a forum and with a local community mindset) is a bit different than Mastodon I feel.
These aspects were top of mind. Iâll add to this thread if I think of more.
Behöver du UX-kompetens? Kolla hÀr!
Det finns ett gÀng fantastiska mÀnniskor som jag vill ge mina allra varmaste rekommendationer. Ett gÀng som precis gÄtt klart utbildningen till UX-Designer pÄ Nackademin.
HÀr Àr fem skÀl till att du ska anstÀlla dem innan nÄgon annan hinner före:
1. I tvÄ Är har de fördjupat sig i allt som marknaden behöver.
2. Har nu bredd och expertis i allt frÄn research till grÀnssnitt.
3. Mer ambitiösa, drivna och nyfikna individer fÄr man leta lÀnge efter.
4. VÄgar ifrÄgasÀtta, utmana och tÀnka nytt för att nÄ bÀsta möjliga resultat.
5. Ăr helt fantastiska pĂ„ inkluderande design och tillgĂ€nglighet.
Och det skadar förstÄs inte att de Àr galet trevliga!
Du kan alltsÄ sluta fiska nu.
Kontakta Mikael Boman pÄ Nackademin sÄ förmedlar han kontakt med klassen.
AlltsÄ gör det typ nu. mikael.boman@nackademin.se
Kan ju bara bli en like pÄ den!
â ïž re: Ice cream blind problems. NSFW.
@weirdwriter
Absolutely epic storytelling Robert. Thank you for the laugh. And also just that insight of forgetting how far sighted people can see was such a great nugget of wisdom.
@ojala Akkoma is just a much better piece of software in many ways. It started with me caring about performance for a 1-person instance, but there are other benefits as well.
I should do a writeup đ
@ojala Ah thanks! I always forget that Mastodon and Akkoma don't play well together with hashtags when there are no other links in the post. I'll remove it, as a solution for now.
Oh, so you've been producing online content for a quarter of a century? Yeah we sucked up that content to build our tool. Now please be a good fool and pay a monthly fee to use our tool that we built on top of your labor.
Technology is improving. Technology is becoming cheaper. The porn industry is growing in all manners. Everyone is an actor.
Itâs now possible to generate pornographic films at a low cost starring any person on Instagram who has a number of photos published. Heaps of films with celebritiesâ faces are available for as little as five dollars. Those who want a customized video with any person on Instagram with less than 2 million followers, a so-called âpersonal girl,â will pay $65.
Yes, itâs fake, but itâs also abuse. And actually, those faces really belong to someone. Desperate young women are now being made aware of nude pictures and videos with people who look just like them. Itâs hard to imagine what this does to their stress levels, wellbeing, mental health and self-image â and what further consequences this may lead to.
While the techbros in Silicon Valley garner attention with their wild theories about computers taking over the world, people are already suffering now, today, for real from these technological leaps.
Media needs to shift focus away from the white menâs fictional fear of being oppressed and focus on those who already are.
https://www.nbcnews.com/tech/internet/deepfake-porn-ai-mr-deep-fake-economy-google-visa-mastercard-download-rcna75071
#AIHype #DigitalEthics #AIEthics
Efter förmiddagens varma, mÀnskliga och kurativa samtal bokar jag genast in nÀsta veckas ljudevent. Jag Àr vÀrd för prestigelösa diskussioner om makt, vÀlbefinnande och inkludering i tech-branschen. Du vÀljer sjÀlv om du bara lyssnar eller rÀcker upp handen och deltar.
Jag kommer senare idag skriva en liten guide för dig som vill vara vÀrd för ditt eget ljudevent pÄ LinkedIn.
Daniel Motaung's battles against Meta should be made into a movie. And here is some awesome news.
« âI never thought, when I started the Alliance in 2019, we would be here todayâwith moderators from every major social media giant forming the first African moderators union,â Motaung said in a statement. âThere have never been more of us. Our cause is right, our way is just, and we shall prevail. I couldnât be more proud of todayâs decision to register the Content Moderators Union.â »
Wow I would love to feature an interview with Daniel in my upcoming book.
https://time.com/6275995/chatgpt-facebook-african-workers-union/
#DigitalEthics
So yeah I get the sentiment of protecting protestors from identification, but they did not think this through. If images of human rights abuse are to be AI-generated, how is there possibly any trust left?
https://www.theguardian.com/world/2023/may/02/amnesty-international-ai-generated-images-criticism
#AIHype #AIEthics #DigitalEthics
It always bothered me that âEvil geniusâ is never considered to be an oxymoron.
The phrase says a lot about a dominant rationalist view of intelligence. And an apparently reasoned dismissal of compassion as evidence of wisdom.
Anyway. Carry on. The sheer number of smart devices in my home is making me question why they havenât gotten rid of me yet.
#AIHype
Somewhere in Silicon Valley there is a playbook that says to launch your product with some glaring privacy issues.
After 2 months of objection fix an issue with the code you've already written for it. You will get good press and this keeps the media and activists busy - so they won't be talking as much about much bigger problems with your product.
When this time has passed the news cycle will start leaving you alone and move on to someone else. Likely someone with a glaring privacy issue.
Efficient.
Inefficient.
Fast.
Slow.
These are not competing concepts. We are always in a dance involving both. Listening. Adapting. Adjusting. Taking action. So when inhuman efficiency becomes the only North Star we miss all the signs, proceeding to crash and burn.
Do > Do > Do > Do > đ±đ„
Do > Observe > Think > đ„°đ
Inefficiency is a truly misleading word. In many cases it can be replaced with thoroughness.
@complexical
Thank you Rachel! I agree many of us are aware but that we also tend to forget to regularly reason around the many implications of this situation. Iâve found that when I produce illustrative models I also encourage more people understand the inherent risks. Iâm imagining, for example, that many users of LinkedIn donât consider how they are âtalkingâ to the algorithm by liking, commenting, watching or even just âhoveringâ(!) over someone elseâs post.
I also use them in teaching of course, and itâs always fun to see how models trigger creative output when discussing them in educational settings.
@petermedvind
Hej Péter, brukar inte vara sÄ hÀr pÄflugen men nÀr jag ser nÄgon jobba med litterÀr översÀttning kan jag inte lÄta bli att dela min C-uppsats frÄn 1994. Mest för att tala om att det Àr ett intresse jag har :)
https://axbom.se/dom/
Had an idea and am now working on a Communication Model for Algorithm-Reinforced Feedback in Digital Social Networks.
Because what else would I be doing on a Saturday morning? đ
My hope is to illustrate (in a write-up) how much the receiver of a message is in fact communicating with an algorithm, and how the sender of a message becomes incentivized to encourage human-to-computer communication rather than human-to-human communication.
Image description:
Communication Model for Algorithm-Reinforced Feedback in Digital Social Networks âą Per Axbom âą Work in progress
Model with a sender composing a message through an encoder (user interface) that passes through some sort of entropy (disorder/randomness) to then be decoded (also in a user interface) and delivered to a receiver who will interpret the message. The receiver provides feedback to the sender but also to an algorithm. The algorithm is then used to instruct the machine which in turn contributes to the aforementioned entropy.
#DigitalEthics #CommunicationScience #SocialNetworking