Author: bipashab

03/25/2020 – Bipasha Banerjee – All Work and No Play? Conversations with a Question-and-Answer Chatbot in the Wild

bipashab Leave a comment

Summary

The paper by Liao et al. talks about conversational agents (CAs) that are used to answer two research questions. The first was to see how CAs interact with users, and the second was to see what kind of conversational interactions could be used for the CAs to gauge user satisfaction. For this task, the authors developed a conversational agent called Cognitive Human Interface Personality (CHIP). The primary function of the agent is to provide HR assistance to new hires to a company. For this research, 377 new employees were the users, and the agent provided support to them for six weeks. CHIP would answer questions related to the company, which is quite natural for newly employed individuals to have. IBM Watson Dialog package has been utilized to incorporate the conversations collected from past CA usages. They made the process iterative, where 20-30 user interaction was taken into account in the development process. The CA was aimed to be conversational and social. In order to do so, users were assisted with regular reminders. Participants in the study were asked to use a specific hashtag, namely, #fail, to provide feedback and consent to the study. The analysis was done using classifiers to provide a characterization of user input. It was concluded that signals in conversational interactions could be used to infer user satisfaction and further develop chat platforms that would utilize such information.

Reflection 

The paper does a decent job of investigating conversational agents and finding out the different forms of interactions users have with the system. This work gave an insight into how these conversations could be used to identify user satisfaction. I particularly was interested to see the kind of interaction the users had with the system. It was also noted in the paper that the frequency of usage of the CA declined within two weeks. This was natural when it comes to using the HR system. However, industries like banking, where 24-hour assistance is needed and desired, would have consistent traffic of users. Additionally, it is essential to note how they maintain the security of users while such systems use human data. For example, HR data is sensitive. The paper did not mention anything about how do we actually make sure that personal data is not transferred or used by any unauthorized application or humans.

One more important factor, in my opinion, is the domain. I do understand why the HR domain was selected. New hires are bound to have questions, and a CA is a perfect solution to answer all such frequently asked questions. However, how would the feasibility of using such agent change with other potential uses of the CA? I believe that the performance of the model would decrease if the system was to be more complex. Here the CA mostly had to anticipate or answer questions from a finite range of available questions. However, a more open-ended application could have endless questioning opportunities. To be able to handle such applications would be challenging. 

The paper also only uses basic machine learning classifiers to answer their first research question. However, I think some deep learning techniques like those mentioned in [1] would help classify the questions better.

Questions

  1. How would the model perform in domains where continuous usage is necessary? Examples are the banking sectors. 
  2. How was the security taken care of in their CA setup? 
  3. Would the performance and feasibility change according to the domain? 
  4. Could deep learning techniques improve the performance of the model?

References 

[1]https://medium.com/@BhashkarKunal/conversational-ai-chatbot-using-deep-learning-how-bi-directional-lstm-machine-reading-38dc5cf5a5a3

Read More

03/04/2020- Bipasha Banerjee – Combining Crowdsourcing and Google Street View to Identify Street-level Accessibility Problems

bipashab 2 Comments

Summary

The paper by Hara et al. attempts to address the problem of sidewalk accessibility by using crowd workers to label the data. The authors had different contributions in addition to just making crowd workers label images. They conduct two studies, a feasibility study and an online crowdsourcing study using AMT. The first study aims to find out how practical it is to label sidewalks using reliable crowd workers (experts). This study also gives an idea of the baseline performance and acts as a validated ground truth data. The second study aims to find out the feasibility of using Amazon Mechanical Turks for this task. They have evaluated the accuracy of image-level as well as pixel-level. The authors have conducted a thorough background study on the current sidewalk accessibility issues, the current audit methods, and that of crowdsourcing and image labeling. They were successful in showing that untrained crowd workers could identify and label sidewalk accessibility issues correctly in the google street view imagery. 

Reflection

Combining crowdsourcing and google street view to identify street- level accessibility is essential and useful for people. The paper was an interesting read and the authors described the system well. In the video[1], the authors show the instructions for the workers. The video gave a fascinating insight into how the task was designed for the workers, explaining every labeling task in detail. 

The paper mentions accessibility, but they have restricted their research for wheelchair users. This works for the first study as they are able to label the obstacles correctly, and this gives us the ground truth data for the next study as well as establishes the feasibility of using crowd workers to identify and label accessibility effectively. However, accessibility problems on sidewalks are also faced by other groups like people with reduced vision, etc. I am curious to see how the experiments would differ if the user-group and the need changes?

The experiments are based on google street view, which is not known to be the best at certain times. There are certain apps that help people get real-time updates on traffic while driving like the app Waze [2]. I was wondering if google maps or any other app insert dynamic updates for street walks, it would be beneficial. It would not only help people but also help the authority in determining which sidewalks are frequently used and the most common issues people face. The paper is a bit old. But, newer technology would surely help users. The paper [3] by the same author is a massive advancement in collecting sidewalk accessibility data. This paper is a good read based on the latest technology.

The paper mentions that active feedback to crowd workers would help improve labeling tasks. I think that dynamic, real-time feedback would be immensely helpful. However, I do understand that it is challenging to implement when using crowd workers, but an internal study could be conducted. For this, a pair or more people need to work simultaneously, where one label and the rest give feedback or some other combinations. 

Questions

  1. Sidewalk accessibility has been discussed for people with accessibility problems. They have considered people in wheelchairs for their studies. I do understand that such people would be needed for study 1, where labeling is a factor. However, how does the idea extend to people with other accessibility issues like reduced vision?
  2. This paper was published in 2013. The authors do mention in the conclusion section that with improvement in GSV and computer vision will overall help. Has any further study been conducted? How much modification of the current system is needed to accommodate the advancement in GSV and computer vision in general? 
  3. Can dynamic feedback to workers be implemented? 

References 

[1] https://www.youtube.com/watch?v=aD1bx_SikGo

[2] https://www.waze.com/waze

[3] http://kotarohara.com/assets/Papers/Saha_ProjectSidewalkAWebBasedCrowdsourcingToolForCollectingSidewalkAccessibilityDataAtScale_CHI2019.pdf

Read More

03/04/2020- Bipasha Banerjee – Pull the Plug? Predicting If Computers or Humans Should Segment Images

bipashab 4 Comments

Summary

The paper by Gurari et al. discusses the segmentation of images and when segmentation should be done by humans and when is a machine only approach applicable. The work described in this paper is interdisciplinary, involving computer vision and human computation. They have considered both fine-grained as well as coarse-grained segmentation approaches to determine where the human or the machine perform better. The PTP framework describes whether to pull the plug on humans or machines. The framework aims to predict if the labeling of images should come from humans or machines and the quality of the labeled image. Their prediction framework is a regression model that captures the segmentation quality. The training data was populated with masks to reflect the quality of the segmentation. The three algorithms used are Hough transform with circles, Otsu thresholding, and adaptive thresholding. For labels, the Jacquard index was considered to indicate the quality of each instance. Nine features were proposed derived from the binary segmentation mask to catch the failure. It was finally derived that a mixed approach performed better than completely relying on humans or computers. 

Reflection 

The use of machines vs. humans is a complex debate. Leveraging both machine and human capabilities is necessary for efficiency and dealing with “big data.” The paper aims to find when to use computers to create coarse-grained segments and when to replace with humans for fine-grained data. I liked that the authors published the code. This helps in the advancement of research and reproducibility.

The authors have used three datasets but all based on images. In my opinion, detecting images is a relatively simple task to identify bounding boxes. I work with texts, and I have observed that segmentation results of large amounts of text are not simple. Most of the available tools fail to segment long documents like ETDs effectively. Nonetheless, segmentation is an important task, and I am intrigued to see how this work can be extended to text. 

Using crowd workers can be tricky. Although Amazon Mechanical Turk allows requesters to specify the experience, quality, etc. of workers, however, the time taken by a worker can vary depending on various factors. Would humans familiar with the dataset or the domain annotate faster? This needs to be thought of well, in my opinion, especially when we are trying to compete against machines. Machines are faster and good at handling vast amounts of data whereas; humans are good at accuracy. This paper highlights the old problem of accuracy vs. speed.

Questions

  1. The segmentation has been done on datasets with images. How does this extend to text? 
  2. Would experts on the topic or people familiar with databases require less time to annotate?
  3. Although three datasets have been used, I wonder if the domain matters? Would complex images affect the accuracy of machines?

Read More

02/26/2020 – Bipasha Banerjee – Explaining Models: An Empirical Study of How Explanations Impact Fairness Judgment

bipashab 1 Comment

Summary 

The paper highlights one of the major problems that the current digital world faces, algorithmic bias, and fairness in AI. They point out that often ML models are trained on data that in itself is bias and, therefore, may result in amplification of the existing bias. This often results in people not trusting AI models. This is a good step towards explainable AI and making models more transparent to the user. The authors used a dataset that is used for predicting the risk of re-offending, and the dataset is known to have a racial bias. Global and local explanations were taken into account across four types of explanations styles, namely, demographic-based, sensitivity based, influence based, and case-based. For measuring fairness, they considered racial discrimination and tried to measure case-specific impact. Cognition and an individuals’ prior perception of fairness of algorithms were considered as measures of individual difference factors. Both qualitative and quantitative methods were taken into account during the evaluation. They concluded that a general solution is not possible but depends on the user profile and fairness issues. 

Reflection 

The paper by Dodge et al. is a commendable effort towards making algorithms and their processing more clear to humans. They take into account not only the algorithmic fairness but also the humans’ perception of the algorithm, various fairness problems, and individual differences in their experiments. The paper was an interesting read, but a better display of results would make it easier for the readers to comprehend. 

In the model fairness section, they are considering fairness in terms of racial discrimination. Later in the paper, they do mention that the re-offending prediction classifier has features such as age included. Additionally, features like gender might play an important role too. It would be interesting to see how age and other features as a fairness issue perform on maybe other datasets where such biases are dominant. 

The authors mentioned that a general solution is not possible to be developed. However, is it possible for the solution to be domain-specific? For example, if we change the dataset to include other features for fairness, we should be able to plug in the new data without having to change the model.

The study was done using crowd workers and not domain experts who are well knowledgeable with the jargon and are used to being unbiased. Humans are prone to be biased with/without intentions. However, people who are in the legal paradigm like judges, attorneys, paralegals, court reporters, law enforcement officers are more likely to be impartial because either they are under oath or years’ of practice and training in an unbiased setup. So, including them in the evaluation and utilizing them as expert crowd workers might yield better results.

Questions

  1. A general solution for a domain rather than one size fits all?
  2. Only racial discrimination is considered as a fairness issue. Are other factors only used as a feature to the classifier? How would the model perform on a varied dataset with other features like gender as a fairness issue?
  3. The authors have used the dataset for the judicial system, and they mentioned their goal was not to study the users. I am curious to know how they anonymize the data, and how was the privacy and security of individuals handled here?

Read More

02/26/2020 – Bipasha Banerjee – Will You Accept an Imperfect AI? Exploring Designs for Adjusting End-user Expectations of AI Systems

bipashab 1 Comment

Summary

The paper talks about user-expectation when it comes to end-user applications. It is essential to make sure that the user-expectations are set to an optimal level so that the user does not find the end product underwhelming. Most of the related work done in this area highlights the fact that user disappointment occurs if the initial expectation is set to “too high”. Initial expectations can originate from advertisements, product reviews, brands, word of mouth, etc. They tested their hypothesis on an AI-powered scheduling assistant. They created an interface similar to the Microsoft Outlook email system. The main purpose of the interface was to detect if an email was sent with the intention of scheduling a meeting. If so, the AI would automatically highlight the meeting request sentence and then allow the user to schedule the meeting. The authors designed three techniques, namely, accuracy indicator, example-based explanations, and control slider, to design for adjusting end-user expectations. Most of their hypotheses were proved to be true. Yet, it was found that an AI system based on high recall had better user acceptance than high precision. 

Reflection

The paper was an interesting read on adjusting the end-user expectation. The AI scheduling assistant was used as a UI-tool to evaluate the users’ reactions and expectations of the system. The authors conducted various experiments based on three design techniques. I was intrigued to find out that the high precision version did not result in a high perception of accuracy. An ML background practitioner always looks at precision (false positive). From this, we can infer that the task at hand should be the judge of what metric we should focus on. It is certainly true that here, displaying a wrongly highlighted sentence would annoy the user less than completely missing out on the meeting details in an email. Hence, I would say this kind of high recall priority should be kept in mind and adjusted according to the end goal of the system.

 It would also be interesting to see how such expectation oriented experiments performed in the case of other complex tasks. This AI scheduling task was straight-forward, where there can be only one correct answer. It is necessary to see how the expectation based approach fairs when the task is subjective. By subjective, I mean, the success of the task would vary from user to user. For example, in the case of text summarization, the judgment of the quality of the end product would be highly dependent on the user reading it. 

Another critical thing to note is the expectation can also stem from a user’s personal skill level and subsequent expectation from a system. As a crowd-worker, having a wrongly highlighted line might not affect as much when the number of emails and tasks are less. How likely is this to annoy busy professionals who might have to deal with a lot of emails and messages with meeting requests. Having multiple incorrect highlights a day is undoubtedly bound to disappoint the user.

Questions 

  1. How does this extend to other complex user-interactive systems?
  2. Certain tasks are relative, like text summarization. How would the system evaluate success and gauge expectations in such cases where the task at hand is extremely subjective?
  3. How would the expectation vary with the skill level of the user? 

Read More

02/19/2020-Bipasha Banerjee -In Search of the Dream Team: Temporally Constrained Multi-Armed Bandits for Identifying Effective Team Structures

bipashab 1 Comment

Summary:

The paper aims to find a Dream Team by adopting teams to different structures and subsequent evaluation. The authors try to identify the ideal team structure using “the multi-armed bandit” approach over time. The dream team structure selects the next exploration task based on the reward from the previous job. They explored a lot of background research on HCI groups, the structural contingency theory from organizational behavior, multi-armed bandit. A network of five bandits was created with different dimensions, namely, hierarchy, interaction patterns, norms of engagement, decision-making norms, and feedback norms. Each of the dimensions has different possible values. For example, for hierarchy, there can be three possible values – none, centralized (where a leader was elected), decentralized (majority vote). Global temporal constraint and dimensional temporal constraint are taken into consideration to determine at what stage the teams are prepared to embrace changes and also take into account if too many dimensions change at one. The authors used the popular game Codenames for the Slack interface. They used Amazon Mechanical Turk to employ 135 workers and assigned them based on five conditions, namely, control, collectively chosen, manager chosen, bandit chosen, and Dream Team is chosen. There were 35 teams with seven teams per condition. It was found that Dream Team based teams outperformed other teams 

Reflection 

The paper was a nice read on selecting the ideal team structure to maximize productivity. The paper did extensive background research on team structures and included theories from HCI and organizational behavior. Being from a CS background, I have no idea about what team structure is and the theory involved behind selecting the ideal structure. It was a very new concept for me, and the difference between the approaches taken by the HCI domain and Organizational behavior was intriguing. The authors described their approach in detail and mathematically, which makes it easy to visualize the problem as well as the method.

The most interesting section was the integration with Slack, where the Slack bot was utilized to guide the Team with broadcast messages. It was interesting to see how different teams reacted to the messages the Slack bot posted. Dream Teams mostly adhered to the suggestions of the Slack bot whereas, some of the other team structures chose to ignore them. It would be good if the evaluation is also done on a different task. The game is relatively simple, and we don’t know how the Dream Team structure would perform for complicated tasks. It would be intriguing to see how this work could be potentially extended.

The paper highlights a probabilistic approach to proposing the ideal team structure. One thing that was not very clear to me is how the slack bots suggest other than taking into consideration the current score and the best approach. Is it using NLP techniques to deduce the sentiment of the comment and then posting a cross-comment? 

Question

  1. The authors used slack to test their hypothesis. How would dream-team perform for real-life software development teams?
  2. The test subjects were Amazon Mechanical Turks, and the project was reasonably simple (codenames game). Would Dream Team performs better than the other structures when it is domain-specific, and experts are involved? Would it lead to more conflicts?
  3. Could we use better NLP techniques and sentiment analysis to guide the DreamTeams better?

Read More

02/19/2020-Bipasha Banerjee – The Work of Sustaining Order in Wikipedia: The Banning of a Vandal

bipashab 1 Comment

Summary: 

The paper discusses software development tools that help in moderating content posted or altered in the online encyclopedia, popularly known as Wikipedia. Wikipedia was built on the concept that anyone with an internet connection and a device could edit pages on the platform. However, such platforms with “anyone can edit” mantra are prone to malicious users, aka Vandals. Vandals are people who post inappropriate content in the form of text alteration, the introduction of offensive content, etc. Humans can be moderators who can scan for offensive content and remove them. However, this is a tedious task for humans to do. It is impossible for them to monitor huge amounts of content and look for small changes hidden in a large body of the text. To aid humans, there are fully automated softwares that are responsible for monitoring, editing, and overall maintenance of the platform. Examples of such software are Huggle and Twinkle. These tools work with humans and help in keeping the platform free from vandals by flagging users and taking appropriate actions as deemed necessary.

Reflection:

This paper was an interesting read on how offensive content is dealt with in platforms like Wikipedia. It was interesting to learn about different tools and how they interact with humans and help them in making the platform clean of bad content. These tools are extremely useful, and it makes use of machine affordance of dealing with large amounts of data. However, I feel we should also discuss the fact that machines need human interference to evaluate its performance. The paper mentions “leveraging the skills of volunteers who may not be qualified to review an article formally”, this statement is bold and leads to a lot of open questions. Yes, this makes it easy to hire people with lesser expertise, but at the same time, it makes us aware of the fact that machines are taking up some jobs and undermining humans’ expertise.  

Most of the tools mentioned are flagging content based on words, phrases, or deletion of enormous content. These can be defined to be rule-based rather than machine learning. Can we implement machine learning and deep learning algorithms where the tool learns from user behavior as Wikipedia is data-rich and could provide a lot of data to the model to train on? The paper mentioned that “significant removal of content” is placed higher on the filter queue. My only concern is sometimes a user might press enter by mistake. For example, take the case of git. Users write codes, and the difference is generally recorded and shown in the diff from the previous commit. If a coder writes new lines of code may be a line or two and press enter erroneously before or after the entire piece, the whole block shows as “newly added” in the diff. This is easy for a human to understand, but a machine flags such content, nonetheless. This may lead to extra work which normally would have been not in the queue or even lower.

The paper talks about the “talk page” where the warnings are posted by tools. This is a very good step as public shaming is needed to stop such baneful behavior. However, we can incorporate a harsher way to “shame” such users. This may be in the form of poster usernames on the main homepage for every category. This won’t work for anonymous, but maybe blocking their IP address would be a temporary fix, I feel like human and computer interaction is well defined in the paper and the concept of content controlling bots make our life easier.

Questions:

  1. Are machines undermining human capabilities? Do we not need expertise any more?
  2. How can such tools utilize the vast amount of data better? E.g., for training deep learning models.
  3. How could such works be extended to other platforms like Twitter?

Read More

02/05/2020 – Bipasha Banerjee – Guidelines for Human-AI Interaction

bipashab Leave a comment

Summary

The paper was published at the ACM CHI Conference on Human Factors in Computing Systems in 2019. The main objective of the paper was to propose 18 general design guidelines for human-AI interaction. The authors consolidated more than 150 design recommendations from multiple sources into a set of 20 guidelines and then revised them to 18. They also performed a user study of 49 participants to evaluate the clarity and relevance of the guidelines. This entire process was done in four phases, namely, consolidating the guidelines, modified evaluation, user study, and expert evaluation of revision. For the user study portion, they recruited people from the HCI background with at least a year of experience in the HCI domain. They evaluated all the guidelines based on their relevance, clarity, and clarifications. Then they had experts review the revisions, and it helped in the detection of problems related to wording and clarity. Experts are people with work experience to UX or HCI, which are familiar with heuristic evaluations. Eleven experts were recruited, and they preferred the revised versions for most. The paper highlights that there is a tradeoff between specialization and generalization.

Reflection

The paper did an extensive survey on existing AI-related designs and proposed 18 applicable guidelines. This is an exciting way to reduce 150 current research ideas to 18 general principles. I liked the way they approached the guidelines based on clarity and relevance. It was interesting to see how this paper referenced the “Principles of Mixed-Initiative User Interfaces”, published in 1999 by Eric Horowitz. The only thing I was not too fond of is the paper was a bit of a monotonous read about all the guidelines. Nonetheless, the guidelines are extremely useful in developing a system that aims to use the human-AI interaction effectively. I liked how they used users and experts to evaluate the guidelines, which suggest the evaluation process is dependable. I do agree with the tradeoff aspect. To make a guideline more usable, the specialization aspect is bound to suffer. It was interesting to learn that the latest AI research is more dominantly found in the industry as they have up-to-date guidelines about the AI design. However, there was no concrete evidence produced in the paper to support this theory.

Questions

  1. They mentioned that “errors are common in AI systems”. What kind of errors are they referring to? What is the percentage of error these systems encounter on an average?
  2. Was there a way to incorporate ranking of guidelines? (During both the user evaluation as well as the expert evaluation phase)
  3. The paper indicates that “the most up-to-date guidance about AI design were found in industry sources”. Is it because the authors are biased in their opinion or do, they have a concrete evidence to state this?

Read More

02/05/2020 – Bipasha Banerjee – Power to the People: The Role of Humans in Interactive Machine Learning

bipashab Leave a comment

Summary 

The article was an interesting read on interactive machine learning published by Amershi et al. in the AI magazine in 2014. The authors pointed out the problems with traditional machine learning (ML). In particular, the time and efforts that are wasted to get a single job done. The process involves time-consuming interactions between machine learning practitioners and domain experts. In order to make this process efficient, continuous interactive approaches are needed to make the model interactive. The authors mentioned that the updates in the interactive strategies are quicker, get updated quickly based on user feedback. Another benefit of this approach that they pointed out where users with little or no ML experience could interact as the idea is input-output driven. They gave several case studies of such applications as the Crayons system. They mentioned some observations which demonstrate how the end users’ involvement affected the learning process. Some novel interfaces were also proposed in this article for interactive machine learning like the assessment of model qualities, timing queries to users, among others.

Reflection

I feel that interactive machine learning is a very useful and novel approach to machine learning. Having users get involved in the learning process indeed saves time and effort as opposed to the traditional approach. In traditional ML approaches, the collaboration between the practitioner and the domain experts are not seamless. I enjoyed reading about interactive based learning and how users are directly involved in the learning process. Case studies like learning for gesture-based music or image segmentation demonstrate how users provide feedback to the learner immediately after looking at the output. In traditional ML algorithms, we do involve human components during training. It is mainly in the form of annotated training labels. However, whenever domain-specific work is involved (e.g., the clustering of low-level protein problems), the task of labeling by crowd workers becomes tricky. Hence, this method of experts and end-users being involved in the learning process is productive. This is essentially a human-in-the-loop approach, as mentioned in the “Ghost Work” text. However, human interaction said it is different from the human interaction that occurs when humans are actively trying to interact with the system. The article mentioned various observations when dealing with humans, and it was interesting to see how humans behave, have biases. This was brought forward by last week’s reading about affordances. We found that humans do tend to have a bias, in this case, positive bias whereas machines tend to have an unbiased opinion (debatable as machines are trained by humans, and that data is prone to bias).

Questions

  1. How to deal with human bias effectively?
  2. How can we evaluate how well the system performs when the input data is not free from human errors? (E.g., humans tend to demonstrate how the learner should behave, which may/may not be the correct approach. They tend to have biases too)
  3. Most of the case studies mentioned are interactive in nature (teaching concepts to robots, interactive Crayons System, etc.). How does this extend to domains that are non-interactive like text analysis?

Read More

01/22/2020 – Bipasha Banerjee – Ghost Work

bipashab 1 Comment

Paper: Ghost Work, How to Stop Silicon Valley from Building a New Global Underclass

Summary: 

The book highlights one of the major players of the artificial intelligence realm that is the “people.” By people, we often think about the customers. However, ghost workers are people in the system who work behind the scenes. These are on-demand workers who are working to keep the system running without any hiccups. The work is unique for these people. Any company who needs a certain task that requires humans to complete, they can request for such services. The task can be anything from flagging adult content, verifying human credentials, labeling images to create training data. On-demand workers are individuals who are not considered full-time or hourly wage workers. They are paid according to the task at hand. Hence, they need to be vigilant to take up tasks as soon as they appear and not lose the job to others who are looking. Tasks can be classified as a micro and macro task. Micro-tasks are small tasks that take less amount of time, but lots of humans to complete the task. Macro-task, on the other hand, are larger projects like developing webpages or building apps. Some common platforms of on-demand works are Amazon Mechanical Turks, Universal Human Relevance System, Lead genius, and Amara.

Reflection 

The topic was fascinating and gave a unique perspective on “gig-workers.” I was aware of Amazon Mechanical Turks and that humans were an integral part of the automation process. However, this gave a more in-depth idea of what role humans play in this. It was intriguing to know how such individuals support themselves and their families. However, the lack of regulated law makes it difficult to accurately estimate how much such people earn and how exactly do they support their lives. 

One important thing to note here is for all of the companies about most of the workers were in the age-group 18-37, and most had a college degree. This means that younger people find it comfortable to use modern technology to depend on such a unique method of earning money. The workers are generally not paid well enough, and they often need to be extremely skilled to make this their sole livelihood. 

The book-chapters made it very clear that in various ways, such employees are an integral part of the technical system, and they are, at times, better performers than regular employees. Does this mean that companies can shift their work model from hiring more and more full-time employees to move to the on-demand model completely? That is still debatable. I do not think that such a model would be liable when it comes to long term commitments to deliver projects. Although, for full-time employees, job security might be the reason they are comfortable. However, having no financial guarantee, in the long run, is bound to be detrimental for a person. In my opinion, the need to perform “flawlessly” would cause people to reach their breaking point eventually. I firmly believe that such roles make people live an unhealthy work-life balance as they would continuously search for tasks to get more gigs.

Questions

  1. Can use more “gig workers” prove to be efficient/profitable for companies?
  2. Other than personal reasons, what makes people take up such jobs? Can someone aspire for such roles, or necessity drives them?
  3. Would older people be able to do such tasks? If yes, how can we measure the efficiency and compare the productivity against the average workforce age of 18-37?

Read More