Feedback A weekly column filled with strange stories, unbelievable marketing claims, and perplexing instructions.

Locked Out

Feedback believes that the expression “couldn’t make up for it” is frequently misinterpreted. This doesn’t imply a limit to creativity; rather, it signifies that some developments can’t fit into a fictional narrative because people insist, “That will never happen.” The issue is, those individuals are mistaken, as reality can be quite absurd.

One of the key organizations in cryptography is the International Association for Cryptographic Research. This non-profit organization is committed to advancing cryptographic science. Recently, the IACR held elections for new officers and board directors while also adjusting its bylaws. They executed this ingeniously, leveraging their expertise as cryptographers. They utilized an online voting platform called Helios, promising “verifiable online elections.”

Helios is remarkably efficient. Every vote is recorded, ensuring that your vote is counted and remains unchanged. Tampering is thus rendered impossible. Simultaneously, every vote is kept confidential. The system employs advanced encryption technology, where all encrypted votes are compiled into an encrypted tally, and only the tally is decrypted.

But how is this tally decrypted? An organization has to appoint multiple directors. The IACR chose three individuals, each granted one-third of the cryptographic keys. All three needed to enter their portions of the keys to decipher the tally and view the results. This was a complete process—partial decryption wasn’t an option.

Then the expected occurred. “Unfortunately, one of the three trustees lost their private key irretrievably, an honest but unfortunate mistake, and as a result, we are unable to compute the decryption share.” I wrote to the IACR on November 21st. “Consequently, Helios was unable to finalize the decryption process, making it technically impossible to obtain or verify the election’s final outcomes.”

The IACR was compelled to: void the elections and restart the entire process. They now plan to “implement a two-out-of-three threshold mechanism for managing private keys and distribute clear written procedures for all trustees to follow before and during elections.” Feedback is eager to scrutinize that “documented procedure” even if its sole instruction is “Remember” in large bold letters on the cover page.

We are also intrigued by how what the IACR calls “human error” can undermine even the most meticulously designed systems. Each time a Silicon Valley entrepreneur speaks of the imminent emergence of human-level artificial intelligence, we internally cringe. The first iteration of such AI will probably align with the average human’s capabilities.

Floating Like a Raisin

The creativity of science journalists in coming up with new and engaging units of measure is always impressive. On November 17th, New York Times featured a story about “a small radio-frequency tag powered by solar energy that weighs just 60 milligrams and retails for $200.” Entomologists utilize this tag to monitor monarch butterflies during their migration across North America.

Anthony Weaver highlighted a sentence aimed at illustrating how heavy the tag is compared to the butterfly. “Most monarchs weigh between 500 and 600 milligrams, meaning each tagged transcontinental traveler is like half a raisin carrying three grains of uncooked rice.”

In the feedback section, we can all agree that this explanation is far clearer than stating “about 1/10 of body weight” could manage. Or as Anthony phrased it, “When I envisioned myself as a half-raisin on a cross-continental trek with rice to Mexico, I genuinely grasped how butterflies relate to science.”

No, this is not a call for you to submit similar examples from your page. New Scientist. Don’t even think about it.

Boys Club

The reason you may not receive feedback on social media is simply because you lack the mental bandwidth to figure out how to stand out on six different platforms with entirely different algorithms. Nonetheless, we half-watch things, which is why I found the spontaneous experiment on LinkedIn intriguing. Women in the field altered their names and pronouns to appear male, resulting in a spike in engagement.

For instance, social media consultant Simone Bonnet observed a 1600 percent rise in profile views after changing her pronouns to “he/him” and her name to “Simon E.” According to The Guardian, others experienced similar increases. Meanwhile, Daniel Hyers, who possesses an ideal LinkedIn name, attempted the reverse by “changing his name to Daniela for 4 days.” Day 1: a -26% drop.”

Sakshi Jain from LinkedIn stated that feedback indicates the platform’s “algorithms and AI systems are not utilized.” Demographic data (age, race, gender, etc.) are factored in as signals to assess the visibility of content, profiles, and posts in users’ feeds. We don’t dispute that, but we also theorize that unintended emergent outcomes play a significant role. Algorithm bias.

Meanwhile, Feedback is in the midst of creating a new LinkedIn page, which we will dub Mansplaining.

Got a story for Feedback?

You can email your article to Feedback at feedback@newscientist.com. Please include your home address. This week’s and past feedback are accessible on our website.