Milestone 1 pentagram

From crowdresearch
Revision as of 23:39, 4 March 2015 by Karthiksenthil (Talk | contribs) (TaskRabbit VS. microWorkers)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

This is the submission page for Milestone 1 by Team pentagram.

Experience the life of a Worker on microWorkers

We tried out the role of a worker on the crowd-sourcing platform microWorkers. Our observations and comments on the experience are as follows :


  1. Setup an account for the team and looked up for available jobs.
  2. Took up some simple jobs like signup tasks, search and verify tasks, commenting, etc.
  3. At the end of experiment, cumulatively we were able to make $1.05


  1. Simple, easy-to-use UI.
  2. Flexibility and options like ordering records based on different parameters like payment, success rate, etc.
  3. Job listings are well categorised.
Good features
  1. The format for description of every task is very simple and easy to understand for any worker.
  2. Gives highlights of the task (like Time taken to finish, review period, employer profile)
  3. Easy to use mechanism to submit the proof of completion of task. Simple answering type to the questions asked by employer.
  4. Reporting misleading/non-working jobs.
  5. Well detailed user profile. This adds transparency to know more about the employer/requester.
  6. Incentives on the profile page to get reputations on particular categories. For example,one of our team member was personally motivated to get a five star reputation in the Twitter category as he was very active on that social media and hence the tasks seemed easy.
  7. Unique feature: Hire group jobs. Ability to join a team of experienced workers and get handpicked for a particular task by employer.
Disliked features
  1. There is no user personalised task suggestion system.For example, if a worker is constantly good at doing Twitter based tasks, there is no exclusive suggestions for him with only Twitter tasks.
  2. Work involved in the tasks and the payment corresponding to them are not justified in many cases.
  3. Reviews of submissions is sometimes very slow.
  4. The "not satisfactory review" for a submission are not very informative at times.

Experience the life of a Requester/Employer on microWorkers

We tried out the role of an employer on the crowd-sourcing platform microWorkers. Our observations and comments on the experience are as follows :


  1. Employers can post their work to be done, as campaigns.
  2. A minimum of $0.75 is required to get a campaign approved.
  3. Work submitted by workers is reviewed by the employer.
Good features
  1. Very detailed and exhaustive UI to describe the details of a campaign, and at the same time very easy to use and user-friendly.
  2. A detailed list of acceptable and non-acceptable types of campaigns. Can be found here
  3. All jobs posted are reviewed and approved by the microWorkers' admins before opening them to the workers. This avoids illegal,violent,spam and scam activities.
  4. Flexibility in choosing workers from specific countries.
  5. Unique feature: Hire groups. Assigning tasks only to group of select expert workers in a domain.
  6. Feature to award bonus to workers.
  7. Transactions page : Dedicated page to show all the monthly transactions in detail.
Disliked features
  1. Personally need to review and rate submissions to a task. It would be better if this is either automated or in-turn crowd-sourced.
  2. Hard to ensure that submissions are not duplicate or invalid.
  3. Approval process for a task can take a long time.
  4. The task creation process is itself cumbersome and time-consuming.

Explore alternative crowd-labor markets

TaskRabbit VS. microWorkers

microWorkers is a crowd-sourcing platform to connect people seeking to get human-only menial work done with people who are seeking such work. TaskRabbit is a crowd-sourcing platform to connect people who require household work done, with workers who are skilled at household work. That is, TaskRabbit is a specialised crowd-sourcing platform for household work. Here are some similarities and differences we noticed between the two platforms.

  1. TaskRabbit is more aesthetically pleasing to look at and use.
  2. TaskRabbit is more requester-oriented than worker oriented. That is, a lot of focus is given to the requirement of the person requesting the task and lesser to the people actually signing up to do the work.
  3. microWorkers has an easier worker policy. That is, the worker need not be vetted before the job, and need not necessarily have the required qualifications to do a certain job. Clickworker, another similar crowd-sourcing platform, on the other hand, require some amount of skills/proficiency before a job is assigned to an individual. TaskRabbit, owing to the nature of the work, is expected to perform careful background checks of all individuals applying as workers. This leads to much more deliberation and time required to register as a worker.
  4. TaskRabbit has regulated payment guidelines. The workers are paid by the hour, with a reasonable pay. Payment is done in a secure manner after the task is completed well. microWorkers on the other hand has an extremely uneven pay scale which is not regulated by any authority. So, it allows work requestors to ask for jobs to be performed for around $0.01 an hour, which is outrageously low, even if the work is menial.




  1. Reachability to a very large community.
  2. Very specific and remote tasks can be completed. For example,translation of a script in a regional variant of a language
  3. Interface is as simple as possible.
  4. Worker satisfaction(for payment) is high,as individually the tasks are very small when divided.
  5. Targetted at the lower echelons of society, it has the promise to bring about their upliftment.
  6. Quality rating system is an efficient method to decide the remuneration for workers.

Scope for improvements and weaknesses

  1. With growth of the worker community at an unusually higher pace(due to extreme simplicity of work) than the tasks at hand, a point of stagnation may be reached when workers do not get enough work for them to be profitable while spending for their mobile internet data.
  2. Pre-processing(includes cost determination) and distribution for iterations will be resource intensive if it has to assure equal division of work.
  3. Instead of an application,that may involve platform issues,a browser based application for the same maybe more feasible as most of the modern day phones support a web browser.This can also span the community that uses any terminal with a browser.
  4. Smaller the division of the document to be converted,more the network traffic at the server.Considering the iterations for accuracy,the traffic at the server will only double.A feasible solution might be, on the go partitioning of the document rather than a pre-processing stage.That is,the user determines how much work he is capable of doing at the current point of time(with a limit). Based on his/her request the partitioning of the document happens,thereby reducing the network traffic of transmitting small packets again and again.
  5. The time taken to complete a task (Efficacy) must also be taken into consideration while payment. A quick response to a task must fetch more money.
  6. A system of allotting different groups for answering, validation and correction may be implemented and repeated till a converging result is achieved.



  1. Wider work variety can be incorporated. For example, images/text can be sent through an SMS.
  2. Payments are made in terms of prepaid mobile recharge.
  3. Various transliterations of the same language can also be resolved using the technique described in the paper.

Scope for improvements and weaknesses

  1. Mobile airtime recharge methods are not currently standardised . An alternate solution for this can be freecharge or paytm .
  2. Payments in terms of mobile recharge is not always favored by some workers.
  3. Image quality over SMSs can be degraded. Instead mobile internet can be used to transfer the images.
  4. SMSs can be sometimes slow in delivery leading to latency.

Flash Teams


  1. Complex tasks(like software engineering) can be crowd-sourced - places more trust on workers
  2. A team can be created in short notice, has a lot of flexibility and results are obtained quickly
  3. It incorporates Organisational Behaviour studies to effectively manage the process
  4. Reusable blocks, creates expertise/specialisation
  5. Pipe-lining, improves throughput and communication
  6. The user who may not have a clear idea of the steps to be followed can plan out the task with existing blocks
  7. Skilled people with little entrepreneurial inclination will be benefited by joining such a platform

Scope for improvements and weaknesses

  1. Workers can be associated with quality and commitment ratings, based on the tasks they have completed.
  2. Workers may be allowed to enroll in groups for a particular task. There will be a better understanding among the workers and the desired outputs may be obtained more quickly.
  3. A list of services that can be provided by the platform can be put up for users who have little idea about the technical details of a task and the platform can create the entire block chain for the user.