Skip to content
 
If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(BBC-US)   Algorithms: Good for business, bad for workers   (bbc.com) divider line
    More: Obvious, Algorithm, examples of algorithms, US developer HireVue, classic example, Fair conduct standards, algorithmic management, absence of concrete evidence, Amazon's fulfilment centre  
•       •       •

706 clicks; posted to Fandom » on 28 Aug 2020 at 9:32 PM (13 weeks ago)   |   Favorite    |   share:  Share on Twitter share via Email Share on Facebook



6 Comments     (+0 »)
 
View Voting Results: Smartest and Funniest
 
2020-08-28 9:35:56 PM  
Fark user imageView Full Size
 
2020-08-28 10:14:58 PM  
Letting self-learning algorithms make important decisions is a really, really bad idea.

Does nobody remember the nice and accurate wolf detecting algorithm that on further examination detected whether snow was present? Do you really want to let your business run on stuff like that?
 
2020-08-29 1:10:23 AM  
Blaming algorithms for unfair employee treatment is like blaming math for your business losses. They are a tool. If used correctly they're a boon. If used incorrectly they are a hammer.

FTFA: At Amazon's fulfilment centre in south-east Melbourne, they set the pace for "pickers", who have timers on their scanners showing how long they have to find the next item. As soon as they scan that item, the timer resets for the next. All at a "not quite walking, not quite running" speed.

It isn't the algorithm that's the problem. It's the dick who plugged in the assumption that people can run all day every day. Double the amount of time and the problem goes away. (Note that doubling the amount of time is an algorithm itself.)
 
2020-08-29 3:24:13 AM  
I've seen too many people act like AI and alogyrthms are impartial and omniscient dispenser of truth, like when the Google AI aonwas even more discriminatory in its hiring pratice, completely forgetting all the biases assumptions that were fed into it
 
2020-08-29 9:57:43 AM  

NotARocketScientist: Blaming algorithms for unfair employee treatment is like blaming math for your business losses. They are a tool. If used correctly they're a boon. If used incorrectly they are a hammer.

FTFA: At Amazon's fulfilment centre in south-east Melbourne, they set the pace for "pickers", who have timers on their scanners showing how long they have to find the next item. As soon as they scan that item, the timer resets for the next. All at a "not quite walking, not quite running" speed.

It isn't the algorithm that's the problem. It's the dick who plugged in the assumption that people can run all day every day. Double the amount of time and the problem goes away. (Note that doubling the amount of time is an algorithm itself.)


It's more than just that. Many warehouses have a rule that their employees are expected to pick a certain number of items per hour. Some are absolutely asinine about this, but some make multiple efforts to do it fairly. It's usually checked as an average every 1-2 weeks, so having one bad day won't skew the numbers much. The number comes in the first place by the most direct method possible - they determine the number by having people actually do the job at a normal walking pace, with reasonable time spend to complete tasks, then drop about 1/3 off of that and say "this is the reasonable minimum speed for this job". They even give supervisors the ability to cancel quota tracking for a day, so if a major equipment malfunction or something else that's in no way the employee's fault happens, they just ignore that day. Many even try to favor positive reinforcement over negative - the quota is intentionally set low, but there's a significant cash bonus for going 15-20% over it, sometimes even a scaling bonus so the people who really go above and beyond have that reflected on their paycheck. It's not perfect, but it sounds at least mostly fair, right?

All of this ignores one thing that causes this system to actually be pretty terrible, and neither the human testers nor the algorithm are seeing it. If everyone simply picked things in the order they came up, the fast and slow items would indeed cancel out (and they could even further ensure this by making sure the realllllly slow items count as multiple times against the item quota), but this assumes that everyone's actually doing that. Workers quickly learn that certain items are almost certain to get them fired if they pick too many, while others will let them slack off and easily pass the standard. The people who get farked over are the honest ones - not only are they not trying to game the system, but because all the good picks are pounced on by the people who do, they're suddenly forced to work 30-50% faster to even make quota... while the dishonest employees are doing half of the work and getting paid more for it. A system meant to ensure fair job retention and bonuses ends up accomplishing exactly the opposite.
 
2020-08-29 2:38:24 PM  

NotARocketScientist: Blaming algorithms for unfair employee treatment is like blaming math for your business losses. They are a tool. If used correctly they're a boon. If used incorrectly they are a hammer.

FTFA: At Amazon's fulfilment centre in south-east Melbourne, they set the pace for "pickers", who have timers on their scanners showing how long they have to find the next item. As soon as they scan that item, the timer resets for the next. All at a "not quite walking, not quite running" speed.

It isn't the algorithm that's the problem. It's the dick who plugged in the assumption that people can run all day every day. Double the amount of time and the problem goes away. (Note that doubling the amount of time is an algorithm itself.)


I'm willing to bet the constant speed assumption was intentional.  It's everywhere.  A perversion of the "stretch goals" concept that was meant to provide workers with positive incentives for achieving more, but these days is usually used to single people out and gives a reason to fire someone when they burn out.
 
Displayed 6 of 6 comments

View Voting Results: Smartest and Funniest

This thread is closed to new comments.

Continue Farking





On Twitter



  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.