06/10 - Munday Monday

i have a new worst fear

 

Anne Cormia [A2UFI24BK2XR9U] Memory Test - $2 - 15 min!(~ 15 minutes) - $2.00 | PANDA

Low

Unrated

Unrated

$10.29 / hour
00:11:40 / completion time

​

Pros:
none

Cons:
Disturbing timed pictures

Qualifications: Total approved HITs >= 100; HIT approval rate (%) >= 95; Adult Content Qualification EqualTo 1; Location In US; Exc: [-331843427-151821] DoesNotExist ;

To read Anne Cormia's full profile check out TurkerView!​

 

411Richmond [A38RXMI7DFPL8G] Verify a single value from a receipt - $0.01 | PANDA

Underpaid

Unrated

Unrated

$2.57 / hour
00:00:14 / completion time

​

Pros:
They're almost always available.

Cons:
They're woefully underpaid

Qualifications:

To read 411Richmond's full profile check out TurkerView!​

 

Title: Entity-centric dialogue creation (expected time 1-2 min) | PANDA Requester: Deep Dialogue [A1DGWH87NMS7M5] (Req TV): $19.03/hr/hr (TO): [Pay: N/A] [Fair: N/A] [Comm: N/A] [Fast: N/A] Description: - Time: 10 minutes(s) HITs Available: 496 Reward: $0.55 Qualifications: Total approved HITs GreaterThanOrEqualTo 1; HIT approval rate (%) GreaterThanOrEqualTo 90;

do you guys know if we can do more than 1 deep dialog? or how many we can do? I don't want to read and learn everything then get capped at 1.

 

most hits on mturk right now...I feel like a lot of them are a trap

 

Title: Copy Editing (WARNING: This HIT may contain adult content. Worker discretion is advised.) BATCH: 367002 | PANDA
Requester: FA-Turkmaster [A1RQC8OYQOA480]
TurkerView: [ $6.28 / hour ]
Description: Edit several answers to questions correcting spelling, grammar, casing, punctuation, etc.
Duration: 60 Min
Available: 507
Reward: $0.20
Qualifications:

• Copy Editor GreaterThan 0,
• Adult Content Qualification EqualTo 1,

[31R68PGVQCY8GD547KDWTSYTYJLB0S]​

 

going to eat chips and watch tv instead. see you later!

 

I've done quite a few of them with no issue. I don't think there's a cap.

 

seems to be endor #1 and #2 if other person leaves, you are SOL and have to return it and get locked out of endors for an hour

 

Isn't there a script to force submit in those instances?

 

it seems to have an internal submit, and i tried the submit script i have and it did nothing

 

Florian Cova [A2F6IFQVPPPJV0] Surveys about philosophical issues - $0.50 | PANDA

Generous

Unrated

Unrated

$14.40 / hour
00:02:05 / completion time

​

Pros:

Cons:

Qualifications: HIT approval rate (%) > 95; Total approved HITs > 50; Location EqualTo US; Adult Content Qualification EqualTo 1;

To read Florian Cova's full profile check out TurkerView!​

 

With the "traffic light" ML hits, how far away must a traffic light be to become "irrelevant" due to distance? — Or if a traffic light is so distant that it's like 2 pixels wide, barely even discernible as a traffic light, can I mark it "no visible lights" (because whoever spot-checks these probably won't notice the tiny speck that took me 70 seconds to find), or again should I mark that barely-visible distant light as present but "irrelevant"?

 

One limitation of those traffic-light ML hits is that we can't mark on the image.

We can only say if there's a relevant traffic light or not, and what color(s) it is,
but we can't mark on the image.

Sometimes when there's some ambiguity, I've occasionally wanted to
mark on the image to clearly indicate which lights I consider relevant.

 

They're underpaid to begin with, so I hadn't worried too much about the what ifs. Their algorithms likely look for what the majority chose as an answer anyway, so your first instinct without overthinking it is probably fine. And they don't do rejections, so personally, I don't worry about it. If they wanted higher quality answers they would use quals and increase the pay, but with these they haven't, so they can eat whatever simple answer I give them and like it.

 

Are they still using blocks instead of exclude quals or am I thinking of a different requester