Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Life of a Data Labeler

scale.ai

Post reply on HN
Print view View original

21–30 of 31 posts

Re: Life of a Data Labeler

#21

Post by cardigan »

irjustin wrote: post #19
cardigan wrote: post #17

Earlier quoted context omitted.

I'm a software engineer at Scale and the author of this blog post. Let's talk about Venezuela for a second. ~75% of the population lost >19lb in body weight in a year according to this survey: https://www.upi.com/Top_News/World-News/2017/02/19/Venezuela... It's unbelievable that we haven't figured out how to prevent people from living like that in the 21st century. I think it's totally deplorable and honestly an affr…

I believe the OP assumed, as a company, Scale means well and that the team is potentially bringing better wages to people who don't traditionally have it. The outright dismissal of OP's criticisms/comments doesn't allow for a discussion in that data entry is a very tedious job along with maybe there are people who are unhappy with the work but very happy with the pay (which I think is perfectly okay). The comment rea…

Sorry I should've explicitly said this, but I was responding to just this part, not the part about the job being tedious:

  But I do see a lot of data around company engagement. 
  There’s basically no chance this level of positivity is ubiquitous for them
When a company hires many regular people in a place like Venezuela, why wouldn't there be ubiquitous positivity? I'm not from the US; I've lived in a few places with extreme poverty; maybe this is based too much off of personal experience. i.e, I don't think the data the OP is referring to applies here. Getting money for basic needs (if you wouldn't have it otherwise) likely dominates most other concerns, including the work being tedious. From another comment, it looks like the OP's other experience with labelers is in the US - where the next best alternatives they can imagine are a lot better - so it makes sense that those labelers aren't as happy.

I guess what you're asking is - how many of these people find their job tedious? I will say it's a lot less tedious than you might imagine as a first impression - e.g, if you see something weird in some data, you talk to other people about it; if you get good, you train people; when doing a new project, you're learning from your coworkers; if you get really good, you might be asked to help develop training materials, etc. So there's a lot of interacting with other people, and a sense of community. For me personally, I honestly find it meditative to label a lot of data - it feels kind of like tending to a large garden, maybe even fulfilling some deep OCD/obsessiveness desire. Some of our labelers find it meaningful that they're contributing to robotics / self driving cars - e.g, the last interviewee in the blog post.

Back to your question though - how many of these people find their job tedious? I'm not sure how to ask the question to them in a way which gives a satisfactory answer. e.g, I'd expect if we just asked "do you feel like your work is tedious?" the answer would be dominated by people's realistic alternatives, and wouldn't have much to do with the job itself. (so we'd get similarly positive responses) If you can think of a better way to frame that question, I'm happy to ask it and post the responses here :-)

Re: Life of a Data Labeler

#22

Post by infinite8s »

gexla wrote: post #15
king_magic wrote: post #5

Earlier quoted context omitted.

RE: your first paragraph: Yeah, I mean I sort of get it. But if I really had to choose between the mental tedium of labeling data (which I end up doing quite a bit of myself) vs. backbreaking outdoor work, or something like repetitive food service prep... I’d pick the latter every single time. At least for me, data labeling is just brutally mind numbing and draining.

> But if I really had to choose between the mental tedium of labeling data (which I end up doing quite a bit of myself) vs. backbreaking outdoor work, or something like repetitive food service prep... I’d pick the latter every single time. In developing countries, status is an important element which people in developed nations may not think about. For example, how many derogative terms can you rattle off for people…

Interestingly, in the US the skin color status has switched when most jobs went indoors (both white and blue color work). Tanning became a status for people with lots of leisure time.

Re: Life of a Data Labeler

#26

Post by cardigan »

sars1996 wrote: post #25

And how am I supposed to know if the attractive, well-educated, contented folks chosen by this company's PR department are actually representative of its workforce?

Check out my other comment on this thread. I interviewed these folks and wrote the blog post (got lots of feedback from friends and design help from our awesome designers) - I'm a software engineer, not a PR department, lol :P

I tried to pick the answers which were most well written, not the ones which were most positive/negative. Personally I do think this is representative of the labelers who've stuck with us, but at this point don't have the energy to argue this; hopefully my other comment is convincing (i.e, there isn't a high bar to making people happy when their next best alternative is a lot worse)

Re: Life of a Data Labeler

#27

Post by UweSchmidt »

A similar opportunity (and maybe a step up) would be 'crowd tester'; if a Dollar or Euro goes a little farther in your country than in the west, then 1€ for a bug reproduction up to 10€ for a major bug could be an attractive additional income ...

... while of course not providing with a stable regular job, and not necessarily doing much to strengthen the local economy in a fundamental sense.

Re: Life of a Data Labeler

#28

Post by YeGoblynQueenne »

joshvm wrote: post #11
b_tterc_p wrote: post #3

I know nothing about this particular company. I also see no reason to assume it’s particularly bad (it could very well be above average). But I do see a lot of data around company engagement. There’s basically no chance this level of positivity is ubiquitous for them, especially when the target workforce appears to be people otherwise below the poverty line. Although pointless tedium is not the most important factor…

There's a great paper about image labelling, written by an expert labeller: https://arxiv.org/abs/1210.3448 One interesting side to this is citizen science; websites like Zooniverse where you can get random members of the public to do your labelling. Technically there's little difference to what Scale does, just that it might be pictures of Pelicans or galaxies rather than cars on the road. In my experience what you'…

Thanks for pointing out that paper.

Re: Life of a Data Labeler

#29

Post by Causality1 »

"Life of a data labeler" is a wildly inaccurate description for "we asked some of our labelers for times when working for Scale made a positive difference in their lives."

What's the median hourly pay for a Scale data labeler? What fraction of their employees enjoy working for Scale? What's their six month retention rate? Amazon's Mechanical Turk has a reputation for grinding people up and burning them out; what makes Scale better?

Re: Life of a Data Labeler

#30

Post by biofunsf »

Causality1 wrote: post #29

"Life of a data labeler" is a wildly inaccurate description for "we asked some of our labelers for times when working for Scale made a positive difference in their lives." What's the median hourly pay for a Scale data labeler? What fraction of their employees enjoy working for Scale? What's their six month retention rate? Amazon's Mechanical Turk has a reputation for grinding people up and burning them out; what make…

Specifics are intentionally hidden, but it seems like the median hourly wage is somewhere around $1-$2.50 an hour.

The labeler side of scale is hosted at "remotasks.com" and they run IP checks to try and ensure that only people from certain countries can sign up to label data. Also a Facebook account is required for signup...

Post reply on HN