12/10 - Satirical Saturday!

Does it offer anything additional over HF, or is it just a preference thing like how some people use HM over HS?

 

Hold on. I guess I've skipped the overwatch talk. So worker.mturk has a PRE limit separate from main mturk? If so I definitely need to get on that

 

my favorite feature in it is to "search all" - so it doesn't scrape just the new HITs, but runs each search individually. Not to remain 'on', obviously, but to catch up after being away or whatever.

edit:

another advantage is listening to ethraiel's dreamy voice.

 

It is fairly simple with it searching for requester name or other searches to look for hits. The one thing I really like is it will speak with TTS when something pops up (if you have that turned on) so I can be watching stuff and hear it say "[Requester name here]"

Yup worker.mturk isn't the same PRE pool as far as I am aware. This is why Hit finder can run in worker.mturk as well.

 

I tried it but didn't really get it i guess. I mean knowing it has a different PRE pool could be useful but in making new searches it seems to basically do the same thing PC does... or am I totally off? lol. I'm guessing the latter.

 

I have HF running all day every day so I don't miss anything that's posted. It does sound a lot better than HM for sure.

The TTS sounds cool. Thats the main reason I didn't use HM, wasting page request on just searching for known matches vs HS that could search for known matches and show you new stuff. I know it didn't before, can't speak to now but I know a few people said that they switched it to sharing.

 

oh, it also gives the full details on the page like gid and links which is super helpful instead of only a text log like HM. but, again, HF has that too.

 

Title: HELP EXPAND KNOWLEDGE OF HOW INVESTORS MAKE DECISIONS | PANDA Requester: Deni [A2X97TX7U7OJ62] (TO) TO Ratings: No TO (Submit a new TO rating for this requester) Description: The task will take approximately 15-20 minutes, and will be completed using web-based software. There’s no need for any advance preparation.  Benefits to you: - $2.00 for participation - knowing you're helping to impact future accounting pract Time: 2 hours Hits Available: 1 Reward: $2.00 Qualifications: Location is US

 

Title: Survey. Usually takes less than two minutes! | PANDA Requester: Gabriel S Lenz [A687MKEY2JOMB] (Contact) TO: [Comm: 5.00] [Pay: 4.05] [Fair: 5.00] [Fast: 4.72] [Reviews: 21] [ToS: 0] Description: Contribute to research! Time: Unable to pull duration HITs Available: 1 Reward: $0.25 Qualifications: HIT approval rate (%) is greater than 90; Location is US

This HIT was posted through TurkerHub xPort Script v2.1 • Learn More • Install Here​

 

Wow... those surgery C-Sats are gross... and fascinating lol. I hope they are lenient because i had no idea what I was looking at. The way they used the robot arms was cool though. Do you guys do these a lot? I see them talked about often but at 1.15 for nearly 20 mins that seems like a bit much to me.

 

C-sats are c-money and I love them! They don't usually take as long but I'm guessing you had to watch that training video they make you watch the first time. They are interesting and most of the time is just watching the video so a lot of people (myself included) are big fans.

 

Well I had to do the training video and then watch the real video and then do the ratings. I did write one small blurb. Good to know the training video doesn't come back each time. I felt a bit unqualified to be rating a surgeon's performance lol. I'm guessing they are pretty good about not rejecting them? I mean I was objective but not a lot of reference points and I got the training one correct. I guess I would try another one.

 

I kind of think that is the point. They want to see what an average Joe that is paying attention will think. I haven't ever seen them reject as well so just do your best work and you will be fine.

 

Good to know... Thanks.

 

They dont reject, but they will give you a qual that will block you from doing their hits. I've only seen this rarely done in the past.

 

http://www.cureus.com/posters/188-c...amazoncom-mechanical-turk-project-vs-facebook

They actually did a study to compare mturk folks to surgeons and found that turkers were just as good at judging on their scale while being cheaper and faster. Yay crowdsourcing!

 

This is the hit I got the rejection on the other day...but the requester emailed me today and sent me the link to redo and said when they got my new code they would reverse it. Very cool- fingers crossed I did not mess it up this time. I honestly don't see how I did in the first place though. But nice requester.

 

Interesting - I got rejected too. I was quite sure I answered everything correctly, but didn't pursue because apathy. Maybe I should follow up.