Taking power back from the algorithm

easyDNS is pleased to sponsor Jesse Hirsh‘s “Future Fibre / Future Tools” segments of his new email list, Metaviews

Algorithmic folklore and the ability to game the system

 

I was driving home from an errand run early this (Monday) morning when I saw two of my goats in the middle of the side road near our house.

We had lent our two male goats to a neighbour, as we didn’t want them knocking up our young female goats (as yet). This neighbour had underestimated both the goats ability to escape containment, as well as their desire to reproduce.

The two bucks had made it nearly three quarters of a kilometer, and more than half way home. However they had not yet figured out how to cross the busier road by the time I pulled up and lured them into my truck.

It was another adventure in being a goatherd, but also a reminder that the users rarely act as anticipated or desired. We like to joke that goats are stubborn and difficult to contain, but they’re nothing compared to humans.

This is a big reason why we remain AI skeptics, and in particular, skeptical of the ability of AI or black box systems to control or manipulate humans.

Often the easiest example to prove this can be found among young people and their intrinsic or intuitive appropriation of technology.

Strategic knowledge is an imaginative praxis that uses deliberate engagement tactics to feed algorithms, in order to specify which content should grace a users’ feed. As a foil to strategic ignorance, strategic knowledge is an affirmation that human recipients of algorithmic decisions can wrest power from humans behind the code.

By flipping the script, the subjects being experimented on become researchers engaging in experiments.

In a black box system or society the blind leading the blind is by necessity and as long as the knowledge is shared openly, it’s not that hard for others to access or participate in the knowledge generation process.

By sharing experiences, asking questions, and crowdsourcing answers, teens are developing an algorithmic folklore while discerning the potential motivations behind TikTok’s software engineering.

Other TikTok users create multiple accounts to segment their different interests, or to exert control over what content the algorithm recommends to them and how their content should be recommended to others. Through selective interaction with particular types of content, users can reach different niches ranging from “garden-tok” to “academic-tok.”

These methods are relevant and powerful for any user, and help provide both agency and intelligence when it comes to using these platforms strategically.

Although they do raise the issue of the digital divide, and whether we’re seeing clear evidence of the power differential that exists between sides.


On the one hand are the people who have access to the dashboards and metaviews that depict what happens on these platforms. On the other are power users who have to figure it out for themselves, but in so doing find a different kind of power and success.

That’s particularly potent when played by an artist who is not just figuring this stuff out for fun, but also for their art and profit.

He started attaching his music to his viral tweets, suspecting that was the way to make it pop off. One day, his mind scanning the internet like a Google algorithm, he noticed an emerging theme: Country trap videos — collisions of hip-hop beats and country tropes — were gaining popularity. What if he wrote a country-themed banger that was also funny and told a story? In 2018, he bought a $30 beat on YouTube, wrote some lyrics — “Cowboy hat from Gucci, Wrangler on my booty” — and posted it, like his other songs, to SoundCloud that December. He named it “Old Town Road” because it sounded like a “real country place” and deluged the internet with memes attached to the song, hoping one would go viral. He even, famously, posted “What’s the name of the song that goes ‘take my horse to the old town road’” on a part of Reddit dedicated to helping people track down earworms. The song spilled over to TikTok, a new barometer for whether a song is a hit, and caught fire. “A lot of people like to say it’s like a kid accidentally got it,” he told Joe Coscarelli, a culture reporter for The Times. “No, this is no accident. I’ve been pushing this hard.”

Artists who are hustling to make money in our contemporary economy understand all too well the role that algorithms play in helping to influence what is popular and what is ignored. Yet in this case the application is not literal, but metaphoric.

Mixing trending topics and fusing memes into each other in the hopes that it catches fire or goes viral. Just like algorithmic folklore, it reflects an attempt to reverse engineer the system, in this case reverse engineer fame.

Bonus points if it also brings controversy or contention.

In March, the song charted on Billboard’s Hot 100, Hot Country and Hot R.&B./Hip-Hop charts at the same time. When Billboard removed the song from its country list, citing an edict that this song about horses did “not embrace enough elements of today’s country music,” fans protested at the perceived racial slight — was the message that Black people didn’t belong in country music? — which only brought more attention.

It’s difficult to quantify the kind of attention economy we find ourselves in, as the value of that attention is both fleeting and contextual.

Yet the real difficulty stems from the algorithm as intermediary, and the opacity it provides. It’s a false sense of secrecy or security as people on both sides find ways to see past and game the system.

Rather than offer fairness or merit-based rewards, the systems just encourage gaming, attempts to manipulate, and overall foster distrust.

Like goats, humans are difficult to contain, and it makes a whole lot more sense to work with them rather than against them. Ample treats help certainly help.

Leave a Reply

Your email address will not be published. Required fields are marked *