Around the Web

The war in Tigray, the effort of resposible AI, digital gardens, dead internet, aesthetics of NFTs and why «My body, my choice» feels out of date.

As I spent a good deal of my week lying in bed, I had too much time at hand to read the internet. This issue is going to be on the longer end. If you only want to read two things, make it the article about the use of AI by The Trevor Project and this article about digital gardening, a blooming subculture in a wonderful niche of the Web. The rest is good too, though.

On a meta level: I’ve written an article which offers a glimpse into the architecture and infrastructure of this very post and how it reaches your mailbox or feed-reader.

The war we don’t talk about

The war in Ukraine still dominates European headlines, diverting attentions all too easily from other wars going on.

In Ethiopia the government has declared a truce in their attacks against Tigray. Tigray is a region in the north of country, home of the Tigray’s People Liberation Front (TPLF). The war is ongoing for the last sixteen months, were government troops, helped by local militias and forces from Eritrea, try to break the resistance of the TPLF. The war is deeply tied to the country’s history. If you want to get a better understand, the New York Times has published an explainer.

Millions of humans in Tigray are mostly cut off from access to food, as famine looms. The truce might make it possible to deliver humanitarian aid to the region, something that has been all but impossible for the last months. But, as Voice of Africa reports, the declaration of the truce might not spell the end of the suffering.

However, Hassan Khannenje, the head of the Horn Institute for Strategic Studies, does not believe the government and the Tigray People's Liberation Front, or TPLF, will give aid groups a free hand.

Tek from Tgaht but it even more bluntly in a YouTube video, calling the truce a lie, designed to skirt looming sanctions.

This ain’t intelligence

AI is often touted as the solution for a broad range of problems, including in medicine. So far, those solutions failed to materialise. Protocol wrote in-depth about the use of AI by The Trevor Project, a nonprofit in the USA catering to LGBTQ+ teenagers in the USA.

The project took immense care how it uses AI, specifically large language models such as GPT. But still, they are aware of the fact that AI can train humans, but not replace them. As such their models are not used in care, but only in training and even there they need to be recalibrated regularly.

While he said the persona models are relatively stable, Fichter said the organization may need to re-train them with new data as the casual language used by kids and teens evolves to incorporate new acronyms, and as current events such as a new law in Texas defining gender-affirming medical care as “child abuse” becomes a topic of conversation, he said.

The whole piece is really worth a read, and a great example of the lengths companies have to go to build products on top of present AI capabilities if they do not want these to be hurtful.

Algorithmenethik, a Bertelsmann Stiftung affiliated group in Germany, posted a review in similar manner, exploring the ways and preqrequisites to use alorithmic systems in the support of women who have experienced domest violence.

But what to do if you found that an algorithm hasn’t treated you fairly? Currently there’s no real legal basis on which you can appeal. AlogorithmWatch wrote down some demands to adapt the German anti-discrimination law.

I’m still closing old tabs (I have a lot of them). A while back Wired reported on AcccesiBe. AccessiBle offers an «accessibility overlay». Overlays are snake oil products that promise to use some obscure mixture of Artificial Unintelligence and public relation promises to make your sites accessible.

Accessibility practitioners agree that overlays do not work, at times make your site harder to use for people with accessibility needs, and are an all-in-all superbad idea. Adrian Roselli has posted an in-depth look into yet another overlay company, their claims, and their failures.

The social and the media

Two years after Facebook has been accused of amplifying hate against the Rohingya minority in Myanmar, hate is still rampant on its platform.

Platforms were mostly locked out of Russia. One notable exception being TikTok. They decided to practise self-censorship, making it impossible for users outside of Russia to view content out of Russia and vice versa.

Nonetheless, an investigation found that TikTok pushes misinformation to new users almost immediately after sign-up. False information is rampant on TikTok, and it’s designed to make it hard to see.

Its core features prime it for remixing media, allowing users to upload videos and sound clips without attributing their origins, the paper said, which makes it difficult to contextualize and factcheck videos. This has created a digital atmosphere in which “it is difficult – even for seasoned journalists and researchers – to discern truth from rumor, parody and fabrication”, researchers added.

Officially because it’s not used to disseminate what the Kreml calls lies, WhatsApp is another service exempt from being blocked in Russia.

Meanwhile pressure to regulate content in its service grows on Telegram. The app was close to being blocked in Brazil, the block was renounced shortly thereafter as Telegram reacted to the demands of Brazil’s highest court.

World Wide Web

Most of us use a version of the internet that is controlled by algorithms, and non-stop feeds of informational overload. But underneath the concrete surface, a small movement of digital gardeners is planting their seeds. Reading this piece about digital gardening gave a sense of calm I seldom experienced when reading about the Web recently.

A garden is a collection of evolving ideas that aren't strictly organised by their publication date. They're inherently exploratory – notes are linked through contextual associations. They aren't refined or complete - notes are published as half-finished thoughts that will grow and evolve over time. They're less rigid, less performative, and less perfect than the personal websites we're used to seeing.

A tweet by Claire Evans reminded of one of my favourite articles about conspiracy theories ever. There’s a thing called Dead Internet Theory. It stipulates that the internet died some years ago, and is a barren wasteland full of bots. Which totally feels true, even if it’s nonsense. Still: I believe.

Thankfully, if all of this starts to bother you, you don’t have to rely on a wacky conspiracy theory for mental comfort. You can just look for evidence of life: The best proof I have that the internet isn’t dead is that I wandered onto some weird website and found an absurd rant about how the internet is so, so dead.

The road to hell is paved with crypto intentions

I really loved the article Why do all NFTs look the same by Max Kohler, as it does not try to do lazy argument «This is not art, it’s just a computer», but ties NFTs into a larger picture of virtual effects in movies, as well as the reproducibility of all art, which Walter Benjamin already observed.

People who make [NFTs] recog­nise it’s dif­fi­cult to ar­gue that a dig­i­tal im­age can be “original” on any ma­te­r­ial level, so they sug­gest a kind of au­then­tic­ity-by-proxy: Buy an NFT and you get a unique en­try in our spe­cial data­base say­ing you own the im­age. That data­base en­try has ef­fec­tively the same func­tion as those fancy art his­to­ri­ans and copy­right lawyers: Establish au­thor­ship, keep track of prove­nance, au­tho­rise de­riv­a­tive works, me­di­ate roy­alty pay­ments, and so on.

Last week, TIME published a longer portrait of Vitalik Buterin, the creator of the Ethereum blockchain. The interview paints a sympathetic picture. A picture I do not want to disagree with. It makes it obvious, though, that Ethereal (and crypto at large) is yet another problem created by privileged white men, who did not need to think about the real-life consequences their «pure» and intellectually challenging project might have.

Vice visited SXSW and witnessed the takeover by crypto mediocrity and a version of the future that is disappointingly blunt, there’s no fun, nowhere.

And yet, despite all the talk I heard about ushering in a new era of diversity and inclusion, it was hard to not notice that every room felt largely the same: mobs of white wealthy men who quickly volunteered that they worked in finance, tech, marketing, or some buzzy fusion of the three.

I decided to cut down on reading crypto news. I still follow web3 is going great, for the lulz, but will dedicate way less time on this topic, especially on the fraud part.

What a lapses

At the beginning of the week hacking group Lapsus$ made the news when they were able to compromise Okta. Okta is an identity broker, a service large companies use to handle identities of their employees and manage capability such as Single Sign On.

Lapsus$ quickly rose from relative obscurity to hacking stardom through multiple high profile breaches over the last months.

It didn't last too long. On Wednesday Bloomberg reported that they have identified one core member of Lapsus$, a teenager from Oxford. On Thursday, the Police of London announced seven arrests in connection with the hacking group.

The episode Dirty Coms of Darknet Diaries profiled a contemporary hacking culture, of wich Lapsus$ appears to be a part, painting a picture of a hacking culture mostly revolving around pranks and money.

Doom is mine but I will share it

Climate. What a bummer. The direst predictions of scientists are coming true fast and even more dire than predicted.

This week, the disappearance of underwater permafrost made the news. Only shortly after it was reported that temperatures in the Arctic and Antarctic were 30°C and 47°C, respectively, higher than normal.

Loose ends in a list of links

As the Verge reports, Stephen Wilhite, creator of the GIF, has died age 74 from COVID-19. Thank you for the dancing.

Der Spiegel published the English translation of research accusing Frontex director Fabrice Leggeri of covering up Frontex’s involvement in illegal pushbacks in the Aegean (German version here, but behind a paywall). That Frontex is complicit in pushbacks has been reported for years, consequences failed to materialise so far.

My body, my choice? Individualism made it harder to fully embrace this cornerstone of feminist rethoric. The slogan has been co-opted by anti mask mandate protesters. Why though? That’s a question answered in a piece in Geschichte der Gegenwart. I first read this angle in Bitch Magazine some months ago.

“My body, my choice” is highly individualistic and—in the end—fails to convey the ways we’re bound up with each other. Especially as Texas institutes a near-complete ban on abortions, it’s crucial that we embrace language and frameworks that emphasize our mutual responsibilities and interconnectedness.

That’s it for this week. Stay sane, hug your friends, and donate to Self-Defined.