Talk:Main Page

From crowdresearch
Jump to: navigation, search

WinterMilestone 1 ANK

Milestone 1 Mechanical Turk account in review

Alternate crowd sourcing websites ( Micro workers )

I took up simple jobs of voting or filling up a form, I felt that navigation through the site was a little difficult and in general the appearance of the website seemed like from the 90’s. Also many jobs ask for proof of having done the job. I wanted to simply drag and drop a screenshot I had saved but was unable to do so. In turn, I typed out the information in the text box provided. An attachment option seems like an obvious feature that should be there, while completing your job.


Mobile Works It’s interesting to see that India seems to be big source of human resource for crowd sourced work and how the app comes into execution.

- small jobs divided into many sub jobs that can be done on phones that are capable of simple OCR tasks.
- telecommunication rates are cheap in india, especially if you use the internet packs for basic search and surfing task but rates become expensive with an increase in usage.
- There is potential for many more people in Asia or India specifically working on these platforms in their spare time. It gives them a good opportunity to earn some extra cash.
- The app has a mechanical approach that seems to work efficiently wherein workers are able to earn $20-25 for an hour worth of work.

What could improve - The number of people in India using smart phones is constantly increasing. ( and this could help the number of people working on simple OCR tasks upgrade to more complex and more rewarding tasks.
- And the number of people investing in smart phones is more than people owning desktop computers which gives crowd sourcing website a serious incentive to make simple, intuitive apps for smart phones.


Strengths - better interface, navigation seems to be easier as compared to Mechanical Turk
- The 3 option evaluation system that gives a filtering for accessibility of jobs.
- What Daemo has to its advantage is that it seems to be working of the weaknesses and areas where Mechanical Turk falls short.
- The Boomerang method of evaluation where in an early feedback mechanism proved to improve work results.

Flash Teams


- A flash team is a linked set of modular tasks that draw upon paid experts from the crowd, often three to six at a time, on demand.
- The basic unit of a flash team is a block, which gathers one or more paid experts from the crowd to perform a focused task. Blocks can be connected via shared inputs and outputs.
- Foundry allows end users to author and manage flash teams. In authoring mode, the user can specify the expertise for each block and enter details about each block’s requirements.

In earlier platforms, there was mention of inadequate information that led to below bar quality of work. Foundry seems like an efficient solution to that, wherein each block has a well thought of problem that needs to be solved and defined properly so that the the worker can effectively solve the problem without confusion. The result seems to be encouraging as well. Not only is the work done at a high quality but also Foundry seems to save time and work efficiently.