Nick @internic@social.cleverlibre.org

This week, Super Bowl 2024 shattered records, with the #NFL championship broadcast on CBS becoming the most-watched televised event in #US history.

Also riding high from the big game? #Elon #Musk's #X.

A whopping 75.85 percent of traffic from X to its advertising clients' websites during the weekend of the Super Bowl was fake.

https://mashable.com/article/x-twitter-elon-musk-bots-fake-traffic

#Ukraine #Russia

Gift link. More than 100 Chinese propaganda accounts are still on X, months after their exposure by Merta and federal prosecutors. They are working to exacerbate tensions in the U.S. and denigrate both Biden and Trump, we found. https://wapo.st/3T4I0GI

Das #Fediverse ist sich oft selbst genug, viele wehren sich reflexhaft gegen alles Neue. Damit verspielt das dezentrale Netzwerk die einzigartige Chance, Motor für eine neue globale Öffentlichkeit zu sein. Doch genau an dieser sollten wir jetzt gemeinsam bauen. Ein Plädoyer.

https://netzpolitik.org/2024/threads-bluesky-und-mastodon-die-chance-ist-jetzt/

It should be a national scandal that you can be imprisoned based on results from software that can't be audited by anyone other than the company selling it.

Two members of Congress want to put a stop to this travesty: https://www.theverge.com/2024/2/15/24074214/justice-in-forensic-algorithms-act-democrats-mark-takano-dwight-evans

Trump just got fined $355 million plus interest for fraud. With interest that's about $450 million! He was also fined $83 million in another civil case - for libeling E. Jean Carroll even after he'd been fined $5 million for the same thing.

There are also 4 criminal cases underway against Trump. He calls this a 'witch hunt' - and I just learned that there's some truth to this. Stormy Daniels, the porn star he paid $130,000 to shut up, has unsuccessfully sued him three times. But she's now near the center of his Manhattan criminal case for violating federal election laws: he took the hush money from campaign funds. And she considers herself a witch.

So it's not about Trump being a hunted witch. It's that he has a witch hunting him.

Lovely exchange at work:

"Geoff, got a compass I can borrow?"

"Drawing circles or finding north?"

"Winter mountaineering in Scotland."

Much chat.

Old pic of mine for the flavour of the conversation.

Google, following the industry trend to AI-all-the-things, has released Magika - a machine learning model which can identify file types. It claims it can outperform traditional methods by 20 per cent.

I pitted it against BSD File on something I figured Google hadn't included in its million-file-strong corpus: CU Amiga's Mega CD-ROM coverdisc from November 1995.

Magika identified... one file correctly, a plain-text document. File? File got 'em all, and quicker too.

(An unfair test, I know!)

This paper titled "Methodological Irregularities in Programming-Language Research" does not mince its words.

> Substantial industry and government investments in software are at risk due to changes in the underlying programming languages, despite the fact that such changes have no empirically verified benefits. One way to address this problem is to establish rigorous evidence standards like those in medicine and other sciences.

ieeexplore.ieee.org/document/7999115

Golden Crown

The sunset is sinking slowly behind the mountain ridge behind us to the south, casting a shadow across most of the peak in front of us. Only the tip is left in sunlight.

3:30pm in November: Stöng (2917 ft/889 m) between Berufjörður and Breiðdalur, eastern #Iceland.

Please read the ALT tag for more...
@photography

#DailyPhoto #TravelThursday #WeatherPhotography #LandscapePhotography

https://bc.ctvnews.ca/air-canada-s-chatbot-gave-a-b-c-man-the-wrong-information-now-the-airline-has-to-pay-for-the-mistake-1.6769454

So, A, kind of an interesting precedent that a chatbot can incur liability for a company - but consider what this suggests for the use of LLMs that are well-known at this point to 'hallucinate' inaccurate information.

Meta has reduced payments to organizations that fact-check WhatsApp misinformation. The development comes just months before elections are set to take place in tens of countries around the globe. https://www.theinformation.com/articles/meta-cuts-fact-checking-efforts-on-whatsapp-as-elections-loom

It also comes as Meta also fired most of its Facebook election team, cutting staff from 300 to 60. The move is a trend across all major social networks, which appear to have embraced their kingmaker roles around the globe. https://www.washingtonpost.com/politics/2024/02/14/biggest-online-threat-2024-elections-isnt-ai/

Let me take a wild guess and say mentioning "#climate" will fall in the "political" category

Do you have Mastodon?

@gaqzi I think that @grimalkina's research on learning culture and learning debt is a must-read in this department. This paper impacted me SO MUCH when I read it ~ 2 years ago. In fact, I reference it so much that I have it bookmarked in my browser 😆

https://www.catharsisinsight.com/_files/ugd/fce7f8_2a41aa82670f4f08a3e403d196bcc341.pdf

The second interview in the Voices of Developer Thriving interview series is live, folks! 🎉

https://www.pluralsight.com/resources/blog/leadership/voices-of-developer-thriving-eli-mellen

This written interview series explores how the four components of the Developer Thriving framework manifest in the lives of software engineers.

Here, I chat with Software Jack-of-all-Trades Eli Mellen about how the following factors have impacted his thriving throughout his career as a software practitioner:
✅ agency
✅ learning culture
✅ belonging
✅ motivation & self-efficacy

I'm tired of sitting through #science talks where all plots are properly credited but artist's impressions and photographs aren't.

As a researcher and astrophotographer who works with super talented artists, this attitude really bothers me.

You study galaxies at the dawn of time. You look for biomarkers in the atmospheres of distant planets. You can most definitely find who made that cool image you just pasted in your slides.

To those who do credit #art : you're awesome 🤗

What's wild about the bluesky bridge thing is that it seems to be trying to follow all of the rules. It's using the same ActivityPub protocol as everyone else. It has a name and the author is trying to make sure it respects everybody else's moderation settings. You can block it or defederate from it. But very few people who I've seen talking about it seem to be placated by that. They're still mad for some reason.

Morning all. Quite a day yesterday, and today so far. I’m obviously taking a beating from everyone who thinks the Bluesky bridge should be opt in. OK.

I want to run one idea by you all. The way the bridge is currently designed, no fediverse profiles or other content are proactively bridged into Bluesky. If someone on Bluesky wants to see or follow someone on the fediverse, they have to manually request it on the bridge. That fediverse user’s posts are then only bridged going forward, and only if someone follows them.

What if, the first time someone on Bluesky requests to follow someone on the fediverse via the bridge, the fediverse user gets prompted, “X from Bluesky wants to follow you. Are you ok with connecting with Bluesky?”, maybe via DM. I assume that would still be considered opt in?

Realistically, most people in the fediverse will never hear about the bridge. Traditional opt in and opt out both generally expect people to proactively find a setting or take some action, often one that only a tiny fraction of people ever learn about. I don’t really care how many people discover or use the bridge, but this kind of just-in-time prompt, only shown when someone wants to follow or interact with them, feels like a useful improvement in that it puts the decision in front of them directly.

Thanks to @kio for the idea. It seems promising; I’m now planning to try it out well before launch. Let me know if you don’t like it.

I was teasing this scoop last week, and only held publication of it until the company confirmed it had addressed the problem:

The Minnesota-based Internet provider U.S. Internet Corp. has a business unit called Securence, which specializes in providing filtered, secure email services to businesses, educational institutions and government agencies worldwide. But until it was notified last week, U.S. Internet was publishing more than a decade's worth of its internal email -- and that of thousands of Securence clients -- in plain text out on the Internet and just a click away for anyone with a Web browser.

https://krebsonsecurity.com/2024/02/u-s-internet-leaked-years-of-internal-customer-emails/