02/02 - Ton of Twos Thursday!

Sometimes I wake up in the middle
Of the night and no less than 3/5 animals are snoring at once while Matt is babbling to himself in his sleep. It's the best kind of quasi-quiet.

 

Title: Transcribe Product Details from Product Image - First HIT $2.00 bonus | PANDA Requester: Matt Garner [A3E68Y3V2QD0IE] (Contact) TO: [Pay: 3.78] [Fair: 4.40] [Comm: 4.67] [Fast: 4.71] [Reviews: 21] [ToS: 0] Description: Update product brand, name, and package size from product image Time: 20 minutes HITs Available: 4 Reward: $0.13 Qualifications: ShopHero Grocery Product Image Transcription Qualification v1 is not less than 1; HIT approval rate (%) is not less than 95;

HIT exported from Mturk Suite v1.3.7​

 

Okay.. Percy's suck donkey eggs now.

 

I get more of a ball vibe from them but they pay well enough.

 

I've got an olde english bulldogge so you can imagine how noisy she is.

 

Title: Rate videos on whether they are racy. | PANDA Requester: Sergey Schmidt [A4IJUXTMQNS4J] (Contact) TO: [Pay: 4.00] [Fair: 5.00] [Comm: 1.00] [Fast: 5.00] [Reviews: 12] [ToS: 0] Description: Given a video, decide if it is racy or not. Time: 15 minutes HITs Available: 6 Reward: $0.10 Qualifications: Location is US;

HIT exported from Mturk Suite v1.3.7​

come get yo dime

 

Thank you for your interest in our Product Label Transcription qualification test.
Unfortunately, at this time, we are not accepting additional qualifications for this series of HITs.
Please back check again at a later time.

 

Edits were made to calculations and conclusions based on incorrect reading of data. I apologize for any inconvenience this may have caused.

System Information

Browser: Chrome 56.x
Operating system: Ubuntu 14.04
Scripts: HitDB, Overwatch, Panda Crazy (separate window), Hit Scraper

Data

I. All HITs, 10 per page, newest first, cache disabled

1. 65.2 KB, 1.62 s
2. 65.0 KB, 1.51 s
3. 65.1 KB, 1.31 s

Average: 65.1 KB, 1.48 s​

II. All HITs, 10 per page, newest first, cache enabled

1. 15.9 KB, 1.01 s
2. 16.3 KB, 1.33 s
3. 16.0 KB, 950 ms

Average: 16.07 KB, 1.10 s​

III. All HITs, 100 per page, newest first, cache disabled

1. 111 KB, 5.85 s
2. 111 KB, 5.94 s
3. 112 KB, 6.15 s

Average: 111 KB, 5.98 s​

IV. All HITs, 100 per page, newest first, cache enabled

1. 63.5 KB, 5.72 s
2. 63.7 KB, 6.82 s
3. 58.4 KB, 6.06 s

Average: 61.87 KB, 6.2 s​

V. All HITs, 10 per page, newest first, cache enabled, incognito (no scripts)

1. 16.2 KB, 879 ms
2. 15.8 KB, 921 ms
3. 15.8 KB, 921 ms

Average: 15.93 KB, 907 ms​

VI. All HITs, 100 per page, newest first, cache enabled, incognito (no scripts)

1. 63.7 KB, 5.82 s
2. 61.1 KB, 5.90 s
3. 61.5 KB, 5.74 s

Average: 62.1 KB, 5.82 s
​

I got lazy and didn't feel like doing cache-disabled incognito for 10 and 100.

Calculations

Cache-enabled
II loads in ~17.74% of the time that IV does. Alternatively, IV takes ~563% as long as II to load.

Using scripts can slow down the page a decent amount.
II incurs a penalty of ~21.28% over V.
IV incurs a penalty of ~6.53% over VI.

Cache-disabled
I incurs a penalty of ~34.55% over II.
IV incurs a penalty of ~3.68% over III, despite utilizing the cache.

Conclusions

Loading less HITs per page results in less resources being required (obvious). However, that is quite a large performance hit (no pun intended) to load more resources -- though, not an equal increase in time to load to the increase in number of HITs (namely, 10 -> 100) due to the fact that many of the resources remain the same.

Disabling the cache hurts 10 HITs per page a lot while strangely increasing the speed at with 100 HITs per page loads. Either way, odds are you aren't clearing your cache every five seconds, so this won't apply.

Scripts have less of an effect on the loading of the page, it seems, than the increase in resources required for the page aside from turking scripts.

---

Hope you enjoyed this informative post.

@geniuscrackhead sparked the testing with the question regarding sufficient download speed.

 

I think most people will only really be limited by PRE but thanks for the detailed info.

 

Title: The Wisdom of Crowds Challenge | PANDA Requester: Stanford Social Systems Lab [A26Z3IU9PUIHK] (Contact) TO: [Pay: 3.25] [Fair: 5.00] [Comm: 3.00] [Fast: 5.00] [Reviews: 6] [ToS: 0] Description: Can you beat the crowd? Jump into this challenge to find out! Time: 20 minutes HITs Available: 58 Reward: $0.40 Qualifications: Location is US;

HIT exported from Mturk Suite v1.3.7​

fun

 

These are fun I did some last week I think I read somewhere you can do up to 50 hits

 

And got 0 of these lol

 

Title: Transcribe 1 hour 11 min of audio in English (#35) | PANDA Requester: Josh [A29TV859IOXFX9] (Contact) TO: [Pay: N/A] [Fair: N/A] [Comm: N/A] [Fast: N/A] [Reviews: 0] [ToS: 0] Description: Transcribing an interview with one person, with occasional time codes Time: 36 hours HITs Available: 1 Reward: $31.00 Qualifications: None;

HIT posted from Mturk Suite v1.3.7​

 

Title: WhatsApp survey | PANDA Requester: Comms-Quant-UX [A388Z19SP010HT] (Contact) TO: [Pay: 4.13] [Fair: 5.00] [Comm: N/A] [Fast: 5.00] [Reviews: 9] [ToS: 0] Description: - Time: 1 hour HITs Available: 293 Reward: $0.15 Qualifications: Total approved HITs GreaterThanOrEqualTo 1; HIT approval rate (%) GreaterThanOrEqualTo 90; Location EqualTo IN;

HIT posted from Mturk Suite v1.3.7​

For those from India

 

Is that an india only survey?

 

yep subcontinents only

 

I only got 1 this time. I managed 25 last time and 10 the first time I saw them. I was hoping they would go unnoticed and one day I could cap. I have lost hope now.

 

Y'all was motherfuckin' active today.

 

@Outer1 - Heard from Jake Fingert in the form of a big old rejection. I don't get why he's communicating with you and not with me. Also don't get why he'll bonus you $60 for finishing the work, but won't pay me the same fair wage for my partial work (which he has in his possession). Time to 1 his TO