Newsletter 20 – July 2022

Welcome to your Turkopticon Newsletter 

We have had a busy start to the year and with your help, we can carry on trying to make turking better for us all.

To leave reviews please visit our homepage turkopticon.net.

Have you been affected by a mass rejection? We’d love to hear your story. Please contact us.

See the bottom of the newsletter for how you can submit short articles.

 

Our Next Open Forum

If you have thoughts on Turkopticon’s mass rejection campaign or any other issues facing turkers, we would like to invite you to our open community forum on Thursday, August 4th. Turkers can stop by between the hours of 1-3 pm EST. As always, we will try to answer questions about the platform itself, turker tools and forums, how to get started on AMT, and more. Please fill out the registration form at the link below and we will provide the Zoom link. Camera use is optional. Feel free to come for 2 minutes or the full 2 hours. We look forward to chatting with you all next week!

Signup at https://turkopticon.net/open-forum

If you want to learn more or have questions, please email us at volunteercoms@turkopticon.net.

 

When You’re Managed by an Algorithm

By Lilly Irani

Which accounts does Amazon suspend and which does it elevate as trusted Masters accounts? They keep these secrets well hidden so I got to reading patents to better understand. The process, it turns out, may involve machine learning algorithms that determine whether to essentially fire you or give you privileged Masters status.

 

One worker lost her account for two weeks before Amazon admitted it made a mistake.  After investigation, Amazon found that the worker’s account had been suspended because she and her son, also a Turker, had logged in from the same Wi-Fi router at home and, thus, had the same Internet Protocol (IP) address. Amazon flagged the second login on the same IP address, thinking it was a worker trying to get paid to do the study twice with different accounts. Nothing in Amazon’s terms of service prohibits multiple workers from working in the same place. Though Amazon admitted the error, it offered no compensation for the two weeks of income that the Turker lost.

 

A 2014 Amazon patent also holds some clues to Masters status. The patent suggests that Amazon may calculate a “judge error rate” or confidence rating it assigns to workers behind the scenes. It might judge workers with lower “speed of task performance” as worse. Could Turkers with kids running around in the background or repetitive stress injuries be less likely to get Masters?

 

Workers’ organization Uni Global wrote in its Top 10 Principles for Workers’ Data Privacy and Protection that “workers must have access to, and influence over, data collected on them.” They also state that “workers must have a full right of explanation when data is used.”

 

What data rights and processes should Turkers have to be treated fairly at work?

 

This article is an excerpt from a blog post “When Managers Rely on Algorithms of Suspicion: Fraud Logics and Their Fallouts.” Proceeds from writing this piece will be donated to Turkopticon.

 

https://www.cigionline.org/articles/when-managers-rely-on-algorithms-of-suspicion-fraud-logics-and-their-fallouts/

Turkopticon is looking for suggestions!

Do you have suggestions for how to make Turkopticon better and would like to share them with us?

 

Please email volunteercoms@turkopticon.net and someone will be in contact to further hear about your suggestions and see how we all can work together to make Turk better for all of us.

 

Even if you think your ideas are ones we should know please don’t be afraid to send them our way because we listen to all suggestions and want everyone to be heard!

Published by admin

Turkopticon helps the people in the 'crowd' of crowdsourcing watch out for each other—because nobody else seems to be! Almost half of the Mechanical Turk workers who wrote their Bill of Rights demanded protection from employers who take their work without paying. Turkopticon lets you REPORT and AVOID shady employers.