So, first of all, I'm not sure why, but it seems that the formatting of the application is off... I have Windows 11. Has anyone else had this problem? And more importantly, could someone tell me how to use it in the most ELI5 way possible. Kindergarten words please. I'm a 24-year-old boomer. I've used PandaCrazy and SearchCrazy til this point, but it's been unreliable. The transition has been tough and the guide that the creator made was too complicated for my technologically-illiterate brain to understand. I've been relying on a mobile app called "turkdroid", which is *okay*, but it doesn't have an auto feature, so I have to go in, click "once", and hope that someone will return the HIT. Some questions I have about the program: 1. How do you catch HITs by search terms? 2. If so, can you auto-accept HITs that are caught by search terms? (say, for example, that I wanted to catch HITs with the word "psychology" in the title or description - is there a way to make them auto-accept?) 3. How do you make it so only qualified HITs show up? 4. Literally, any general information explained very simply would help me out a lot. If there is anything important I should know about it's functionality, please do let me know. Thank you in advance. Also, so sorry if this is posted in the wrong section :') Sent from my SM-G781U1 using Tapatalk
Your screen resolution is probably throwing it off. I had this issue with my laptop. Two solutions (pick one or the other): You can play with your screen resolution, but you will have to do this before opening QBC. If you change the resolution while QBC is open, it wont fix anything until you close and reopen QBC. You will just have to play around to find the right one. You can disable fullscreen optimizations for this program. Find the QBC icon and right click it. Go to properties. Click on the compatibility tab. Click on the empty box next to "Disable fullscreen optimizations" to disable. You won't need to use these anymore. QBC is going to run all of your pandas. If you run external pandas, you are going send to many requests to Amazon. Amazon only lets you send so many before they throttle you. A panda is a request to accept a HIT. Basically, all of these programs just send requests to accept a HIT until it finally gets lucky. If you want to catch HITs by search terms, you just need to go the Watchers tab on the left side of the screen. Then go to the keywords tab at the top. Then you will want to click the green button. Add whatever keywords you want. Just keep in mind if you have just one word, you will be casting a wide net, so pick a very specific word to search for, or use two words. E.g. Apple vs Baked Apples. Play around with it till you get what you want. In order for QBC to catch them, you will want to click on the panda icon. In my example it is green, which means I turned this feature on. Note: Keyword Watchers are typically for batches, not surveys. You will need to keep an eye on your panda manager to turn these off, or you will need to add some more criteria, which quite frankly, will cause you problems down the line. If you mean, how do you filter out unqualified HITs from the Scrapers tab, you just go to the filters option. Note: I put arrows around it since I am using darkmode and it might be hard to see.
Instead of Keyword Watchers, you should focus on YV and TV Watchers (unless your goal is to only work on very specific surveys). To break it down: YourView Watchers - looks at requesters you have worked for in the past. TurkerView Watchers - looks at requesters in general. You may or may not have worked for the requester before. The way both of these work is by referencing data at TurkerView. For our purposes, we are only concerned with the fact that QBC looks at the hourly rate on TurkerView. If it sees something in the Scraper that fits, its going to automatically create a panda to try to catch it. To set one of these up, you will need to go to the Watcher tab. Then click on the little green buttons with plus signs in the respective areas. Then just play around with adding and defining your criteria. Lots of schools of though on this, but less is more. Note: For YV Watchers to work, you need to upload and review nearly every HIT. Think of the law of averages. You do not necessarily need to write a review, but it will be helpful for others if you do. As mentioned above, YV Watchers look at requesters you have worked for in the past, and if you have no history, YV Watchers are not going to work. Why these are important watchers? QBC does a great job of managing pandas. It will turn them on and off at the appropriate times so you do not have to. If you are focused mostly on surveys, these two watchers are essentially set it and forget it. There is a lot more, but this is good enough for now. As you keep using QBC, you will start to learn about its other features. Tons and tons of information on the forums, just dig around.
Definitely some helpful ideas posted here. One thing I was curious about was if setting your Maximum Reward to $0.00 lessens the chance of it actually catching something. Does this make the area "to wide" and end up missing on other HITS if you were to decrease the scope of what it's looking for? I assume that if you set a decent hourly rate of $8-9/hour and then use no Maximum that it "should" be ok?
I personally feel less is more. The less restrictions I give QBC, the more it will find. Using a Max Reward, as I understand it, allows you to hyper focus on a set of HITs for a specific length of time. For example, you set both your Hourly to $12 and your Max Reward to $12, you are telling QBC, don't catch HITs that could take me more than 1 hour to complete (plus or minus). Hourly to $12 and your Max Reward to $6, no HITs that could take more than 30 minutes. Etc... One could argue, for example, if 2 HITs were posted at the same time of equal hourly rate, but not in value, what would happen if a Max Reward were set and not set: Max Reward set, potential of 0 HITs meeting your criteria HIT 1 = 0 reward HIT 2 = 0 reward Max Reward set, potential of 1 of 2 HITs meeting your criteria HIT 1 = potential reward with 30 pandas firing consecutively HIT 2 = 0 reward Max Reward set, potential of 1 of 2 HITs meeting your criteria HIT 1 = 0 reward HIT 2 = potential reward with 30 pandas firing consecutively Max Reward set, potential of 2 of 2 HITs meeting your criteria HIT 1 = potential reward with 30 pandas firing every other second HIT 2 = potential reward with 30 pandas firing every other second Max Reward blank, potential of 2 HITs HIT 1 = potential reward with 30 pandas firing every other second HIT 2 = potential reward with 30 pandas firing every other second I am not a statistician, but there probably is some probability you could calculate from this thought experiment. I personally feel that your chances are greater uncapped though.
I appreciate the super in-depth reply. I'm still always trying different approaches to see what works best. The last thing I was curious about in relation to this was enabling to many watchers, but doesn't seem like that has much effect either. For example, one YV watcher, 2 TV watchers, and one Unrated watcher. Wasn't sure if 4 watches decreases chances of it catching other HITS because it's multitasking a bit to much.
I do not think you need to worry about multitasking boggling down QBC. The only thing I would be concerned with is having two TV or two YV watchers that overlap in restrictions running at the same time. You do not want to accidently cut off a HIT that might satisfy 1 or the other. @CT would have to weigh in on how the logic works.
Not a problem, once one triggers it's going to kick out of the checks and move along with other stuff. In fact YV/TV watchers are sort of meant to overlap, and only cover each other on the off chance there is a discrepancy so yeah totally fine to let it ride.
