The 199th Block: Media regulation, search engine quality, and hyperrealism
Scroll to the end to see why NYT sucks
This week…
Your reading time is about 5 minutes. Let’s start.
Last week’s edition was long so let’s keep it short this week: There is a much more modern, and much less tragic story behind another seed bank (as opposed to last week’s story), even if it’s sometimes dubbed a ‘Doomsday’ vault, the Svalbard Global Seed Vault in Norway. Jennifer Duggan wrote an exceptional piece for Time Magazine in 2017 about it, which you can read here.
So far, there have only been two withdrawals made from the vault: In 2015, as a result of the Syrian civil war, the International Centre for Agricultural Research in the Dry Areas (ICARDA), which was unable to maintain its genebank in Syria, withdrew some of the backup samples it had previously stored to regenerate them. They also made another withdrawal in 2017. Those seeds were planted in the fields of Lebanon and Morocco, multiplied, and some were returned to the vault, while the rest were added to the ICARDA’s genebanks for conservation and distribution.1
Since only a small number of relevant people are allowed inside the vault, take a virtual tour of the vault here, if you are curious.
And now, a selection of top stories on my radar, a few personal recommendations, and the chart of the week.
ICYMI: The Previous Block informed you of that time when Lamarckism was favoured over genetics, causing the death of tens of millions of people. CORRECTION NOTICE: None notified.MEDIA REGULATION
Indonesia issues regulations requiring digital platforms to pay media for content
Ananda Teresia for Reuters:
The regulation posted on the government's website suggests cooperation between digital platforms and media companies could be in the form of paying licences or sharing data of news users.
A committee would be formed to ensure digital platforms fulfil their responsibilities to the media companies, it said.
The regulation, which takes effect in six months, would not harm content creators as it applied only to digital platforms, Jokowi said.
Hmm, haven’t we heard this song before? Loosely linked:
Meta believes it is not required to pay for Indonesia news content posted voluntarily by Stanley Widianto for Reuters.
X takes down accounts that media say are linked to India farmers protests by Shivam Patel for Reuters.
X briefly suspends account of Alexei Navalny’s widow by Sean Seddon for BBC.
Milei, the media, and the market by Jon Allsop for CJR.
Demoted, deleted, and denied: There’s more than just shadowbanning on Instagram by Natasha Uzcátegui-Liggett and Tomas Apodaca for The Markup.
SEARCH ENGINE AND WEB INDEXING
Google’s retiring of Internet archiving tool draws ire of China researchers
Erin Hale for Al Jazeera:
There are alternatives to Google’s cached pages, namely the non-profit Internet Archive’s Wayback Machine.
But Google’s removal of cached links makes it harder to know what is missing in the first place, said Dakota Cary, a non-resident fellow at the Atlantic Council’s Global China Hub.
“We’re not going to know how much we are missing because we can’t measure what was lost, because it’s not something we can see any more,” Cary told Al Jazeera.
Even dead links in Google’s search results could give researchers pointers or show how a website had been changed, he said.
Loosely linked:
Google tests removing the News tab from search results by Sarah Scire for Nieman Lab.
Google cut a deal with Reddit for AI training data by Emma Roth for The Verge.
Does anyone even want an AI search engine? by Ryan Broderick for Fast Company.
Indexing the information age by Monica Westin for Aeon.
HYPERREALISM
Behind a secretive global network of non-consensual deepfake pornography
Kolina Koltai for Bellingcat:
Perhaps the strangest discovery through this entire saga was that an AI blogging site hosted exclusive interviews with both the “CEO” of Clothoff and the “Founder” of Undress.
The interviews themselves read like ChatGPT responses to prompts or poorly written prose that no executive would consider a marketing coup. Even more interesting is that the headshots of these people appear to be AI-generated. This was particularly odd since the other interviews on the website appear to feature real people.
Websites these days all look alike but this investigation showed how several domains are tied together. Loosely linked:
Deepfake democracy: Behind the AI trickery shaping India’s 2024 election by Yashraj Sharma for Al Jazeera.
AI deepfakes come of age as billions prepare to vote in a bumper year of elections by Jonathan Yerushalmy for The Guardian.
Google pauses Gemini’s ability to generate AI images of people after diversity errors by Tom Warren for The Verge.
OpenAI’s new video generation tool could learn a lot from babies by John Naughton for The Guardian.
What I read, listen, and watch…
I’m reading Calling Bullshit: The Art of Scepticism in a Data-Driven World (2020) by Carl Bergstrom and Jevin D. West — the same professors who taught the university course with the same name.
I’m listening to Tech Won’t Save Us, hosted by Paris Marx. In this episode, Nastasia Hadjadji explained why France’s start-up nation Is a neoliberal hell.
I’m watching an episode of The Listening Post hosted by Richard Gizbert on how the Western media chooses who they champion.
Other curious links, including en español et français:
A new study looks at the positive things that can happen when journalism and comedy intersect by Hanaa’ Tameez for Nieman Lab.
Why African filmmakers aren’t producing nature documentaries by Marin K.N. Siele for Semafor.
‘Disillusioned about China’, more Chinese aim for US via risky Darien Gap by Peter Yeung for Al Jazeera.
Are evidence-based medicine and public health incompatible? by Michael Schulson for Undark.
La tecnoeuforia es una falacia. Ha llegado la hora de recuperar el control de la tecnología por Marta del Amo en Retina.
Locura en España por la criptomoneda del jefe de ChatGPT: casi el 1% de la población ha dado su iris por ella por Carlos del Castillo en elDiario.es.
Quand « du coup » devient « fait que » : comment se font les transferts linguistiques entre les Français et les Québécois par Nadège Fournier dans The Conversation.
La liberté d’expression encadrée ou entravée ? par Cyrille Frank dans Mediaculture.
Chart of the week
For Reuters Institute, Richard Fletcher detailed the proportion of top news websites in ten countries that block OpenAI’s and Google’s AI crawlers.
And one more thing
Does the NYT still do A/B tests on headlines? Why do they still suck?
More on NYT’s A/B headlines tests here by Tom Cleveland on TJCX from 2021.
Westengen, Ola T., et al. ‘Safeguarding a Global Seed Heritage from Syria to Svalbard.’ Nature News, Nature Publishing Group, 9 Nov. 2020, www.nature.com/articles/s41477-020-00802-z. Accessed 25 Feb. 2024.