Happy Monday. My name is David Meyer and I’m going to be regularly writing the Data Sheet lead essay, so, pleased to meet you. A bit about me: I’m a South African who’s lived in Berlin for 11 years and was in the U.K. for 13 years before that; I’ve been a tech journalist for most of my career; and I also front an alternative-rock band called The Board, so expect the odd musical reference here and there. Right, then—on with the show!
Twitter bots get a bad rap sometimes, which is understandable, given that many of these automated accounts were created to spew spam, manipulate opinion, and/or harass people. Their prevalence on the platform also makes it difficult to establish just how many of Twitter’s accounts represent real people, as owner and CEO Elon Musk repeatedly mentioned during last year’s tempestuous takeover.
They’re not all bad, though. Some bots publish urgently useful information about things like earthquakes. Others extend the functionality of Twitter itself, like the self-explanatory Thread Reader App. There are bots for cheering people up and opening their eyes to the universe around them, bots that let people anonymously confess their dark deeds, and bots that monitor lawmakers’ attempts to massage their images. In short, bots and their creators are a big part of what makes Twitter interesting, meaningful, and weird.
Musk being Musk, this situation needed shaking up. Last Wednesday, Twitter announced it would—with just a week’s notice—end free access to its application programming interface (API). That meant external developers and researchers would have to pay between $149 and $2,499 a month to plug their services or research programs into Twitter’s systems, either to make automated tweets or to read and analyze Twitter data. The makers of prominent Twitter bots began saying their goodbyes.
But wait! Over the weekend, Musk tweeted that “responding to feedback, Twitter will enable a light, write-only API for bots providing good content that is free.”
On the plus side, being responsive to criticism is a good thing. However, Musk and his team really ought to have known better before making last week’s announcement, the downsides of which were obvious to anyone familiar with the different types of actors on Twitter and the many ways in which people use the service. Musk may have claimed he was trying to introduce more friction for “bot scammers and opinion manipulators,” but the undifferentiated demand for payment was a slap in the face for enthusiasts who have invested considerable time and effort in making Twitter a worthwhile experience.
Its partial reversal was also characteristically chaotic. “What is a ‘light, write-only API’?” tweeted Luca Hammer, a developer whose free FediFinder service—which helps people move to rival network Mastodon by identifying which of the people they follow on Twitter are also active there—will be scuppered by the shift. “What is ‘good content,’ and will there be humans to check each piece of content from bots if it’s good enough?”
These are good questions, particularly the latter. I’m old enough to remember when Musk was all about free speech and leveling the playing field. Now, by design or otherwise, he has dragged his company into the business of deciding which content is good or bad. For a lean organization with an overworked CEO, that seems like a heck of a task to be taking on.
Want to send thoughts or suggestions to Data Sheet? Drop me a line here.
David Meyer
Data Sheet’s daily news section was written and curated by Andrea Guzman.
