Crowdsourcing in Interface Design
Various Aspects of Crowdsourcing
Crowdsourcing software development is a developing area of software engineering. As a Human-computer interaction (HCI) software development methodology, crowdsourcing calls for collaborative participation in software development tasks, including the creation of user specification, design, coding and software testing. Typically, members of an enterprise (enterprise crowdsourcing) or people contracted by an entity (mobile crowdsourcing) conduct these tasks. There is a growing interest in the application of crowdsourcing in the field of Information technology, as well as in the professional business landscape. This paper explores various aspects of crowdsourcing, including its invention, development, benefits and challenges in the context of HCI and design projects. Similarly to other software development approaches, crowdsourcing has an impact on budgets and timelines of design projects.
Additionally, the paper explores the legal, ethical and societal issues linked to crowdsourcing in interface design and possible solutions to the concerns related to these issues.
Digital economy is increasingly becoming an integral part of the IT environment using technology to improve communication, trade, research, education, and living standards. Crowdsourcing harnesses the collective intelligence of a group of individuals with diverse backgrounds in order to yield greater intelligence or increase productivity (Hughes, 2011). Crowdsourcing has been successfully applied to computation and problems that are easily solved by humans. In the past years, crowdsourcing has emerged as a research theme, as well as an innovative service platform for harnessing skills of a large Internet-connected population. For this reason, crowdsourcing research intersects with various domains and presents new challenges, such as crowdsourcing as a novel methodology for research; computation and problem-solving; development of new application and services based on HCI sensing technologies; design of improved crowdsourcing platforms with quality control mechanisms; gamification of work and incentive-centered design; application of crowdsourcing in business process outsourcing realm. Evidently, crowdsourcing promises numerous advantages despite the social, ethical and legal challenges that accompany the methodology in HCI. To have a better understanding of crowdsourcing as an effective methodology in HCI, this paper explores various aspects of HCI including its invention, development, benefits and challenges.
The Invention and Growth of Crowdsourcing in the Field of Interface Design
Grier (2005) notes that the first computers were humans. This argument is consistent with the observation that human computation entails completing tasks that cannot be computed by distributed computers on demand and scale. In other words, the use of individuals to perform computations predate the Automatic Computers (ACs) built to perform numerous routine computations in the present world (Lease & Alonso, 2013). The term “crowdsourcing” was coined by Jeff Howe and was defined as the process of taking a task that is conventionally performed by known agents and outsource it to an undefined, usually larger group by way of an open call (Hecht, Teevan, Morris, & Liebling, 2012). The utilization of open source principles is another perspective of viewing crowdsourcing. Wikipedia is an apt example of crowdsourcing. Therefore, crowdsourcing can be viewed as a methodology or tool by which a human computation system can distribute tasks. According to Lease and Alonso (2013), human computation entails a computation performed by individuals, whereas human computation systems organize participants’ efforts to complete a computation.
As noted by Lease and Omar (2013), a variety of business and technological terminologies are presently being used in regards to crowdsourcing, such as human computation, collective intelligence, peer-production, citizen science, and serious games. This is an illustration of the development in crowdsourcing. This paper focuses on compensating online software crowdsourcing, which entails performing software development tasks, particularly interface design by distributed developers who are compensated financially by requesters (Lease & Alonso, 2013; Nguyen, 2012). In this context, the term requesters refer to organizations, groups or individuals that are in need of crowdsourcing services. Thus, crowdsourcing is a socio-technical work system composed through a set of web-based relationships that connect individuals, organizations, groups and work activities. Some of the prominent crowdsourcing platforms include oDesk, ManPower, MobileWorks, Crowdflower and Freelancer (Lease & Alonso, 2013). 99Designs and TopCoder are examples of crowdsourcing platforms designed for specific expertise. While the mentioned platforms are designed for professional work, there are other platforms meant for nefarious intentions, including crowdturfing, gold farming, and CAPTCHA solving. Regarding growth, many distinct forms of crowdsourcing co-exist in the present-day world under the same umbrella term. Other models include a game with a purpose and peer-production. Wikipedia is one of the apt examples of a complex, large and quality-peer production. Observably, most of the applications are built based on collective intelligence. While volunteer-based crowdsourcing platforms may be effective, user participation may be influenced by other factors including compensation and motivation.
The Impact of Crowdsourcing on the Field of Interface Design
Crowdsourcing has the potential to strengthen a flexible or mobile workforce and mitigate challenges such as shortages of software engineering experts in specific geographical locations or complicated software development tasks. For individuals, crowdsourcing creates opportunities for social mobility and income generation in areas where economies are stagnant, and governmental structures discourage local investment. For example, through crowdsourcing, a software developer in India can work on a software design project hosted in the United States. Consequently, the developer in India earns income without the need to apply for an American visa and travel to the United States. In the same context, the final product will have taken into account feedback from various locations, especially if it is an application, for example, a website that would be accessed by people in the global community.
Benefits of Incorporating Crowdsourcing in Interface Design
To non-programmers, the HCI simply provides recommendations for interface design such as icons, forms, login screens and data display. However, as noted by Nguyen (2012), interface design requires sufficient computer science and software engineering knowledge. As cited by Nguyen (2012), a recent research showed that HCI is a multidisciplinary field. For these reasons it is evident that crowdsourcing impacts interface design. Through crowdsourcing a requester can access knowledge and skill from different disciplines including emotions, crowdsourcing, ergonomics, artificial intelligence and business intelligence. Thus, the requester can develop services and applications that have diverse features. Nguyen (2012) also asserts that an effective interface designer should have expertise in an array of topics (fields), because it is virtually impossible to design an effective interactive interface using only one computer discipline in isolation. Figure 1 is a depiction of HCI as a multidisciplinary field.
HCI as a Multidisciplinary Field
Crowdsourcing enriches interface design by availing various experts to requesters. For example, sociology enables developers to have a better understanding of targeted audiences, minimizing any social or ethical risk associated with diversity at the same time.
Challenges of Incorporating Crowdsourcing in Design Project
Just like any other outsourcing and collaborating system, crowdsourcing faces numerous challenges including economic, social and political factors. Opponents of crowdsourcing argue that it ships job opportunities overseas leaving the locals unemployed. For instance, Donald Trump views crowdsourcing as a threat to the United States economy, because it creates an unemployment gap in the United States. In this regard, the opposition is both economic and political. Similarly to the other systems of outsourcing business, software crowdsourcing transfers responsibilities of the development team to a third party, especially if participants do the development from the external environment of a company. To that end, legal, technical, and ethical challenges emerge. Arguably, the requester loses control over the development process to third parties. Managerial resistance accompanies this concern, especially if critical information assets are involved. For example, it is understandable if the management of a government security agency like the FBI or CIA rules out the possibility of crowdsourcing the development of electronic document management systems on security grounds. In contrast, a private company resisting the development of an interactive system through crowdsourcing is a depiction of resistance to change, especially if there is no security threat.
The existing Internet-based crowdsourcing tools offer an opportunity to improve the ability to get distributed software development done within budget and in time. One of the noted points in crowdsourced software development process is that successful deployment of crowdsourcing systems depends not only on information technology but also largely on leadership and strategic management of the participants. Given that crowdsourcing relies heavily on ICTs for connectivity, it is undeniable that crowdsourcing performance is influenced by the availability of connecting technology. In this context, availability refers to timely access to the development platform or shared resources. Despite being dispersed, participants’ success is also attributed to the fact that the team has goals aligned with their vision. Additionally, the success of participants also relies on the attention to the team’s adoption of elements such as identifying team mission and defining team success. During the distribution of roles, development roles should be crafted based on the scope of the project.
As crowdsourcing is naturally a human-centric methodology, attention to human factors is likely to affect the quality of data gathered, information processed and resultant products. Hence, poorly collected data can result in dysfunctional products or inaccurate content. Typically, poor data, code and resultant products are associated with ignorant or lazy participants and poorly designed interfaces. Furthermore, if task instructions are poorly communicated, the product will be erroneous. In other words, the implementation of crowdsourcing in interface design relies on the effective communication channels, design methodology, and data collection methods. Approaches to quality assurance grounded on human factors rather than statistical methods are complementary in crowdsourcing, allowing all the factors affecting success to be investigated independently (Lease & Alonso, 2013). However, requesters should be reminded that the quality of the gathered data varies in different task designs. This point adds to the argument that the implantation of crowdsourcing in interface design is also affected by the task design. As a consequence, investigation of quality assurance algorithms should consider the variability in crowdsourcing data.
Generating Interest in Design Project
Different people are motivated or inspired by various incentives, which can occur independently or jointly (DeNisi & Pritchard, 2006). For this reason, it is important to design appropriate incentives for the crowdsourced task through incentive engineering. Some of the incentive mechanisms that can be explored to generate interest in a design project include fun, pay, barter, altruism, prestige and learning (Miner, 2015; O’Neil & Drillings, 2012). For instance, in the case of citizen science, nonprofits and scientists distribute various tasks to volunteers motivated by learning, altruism and socialization opportunities. In the context of Games with a Purpose, work is motivated by prestige, socialization and opportunities for fun. One of the strategies that can be used to help participants move towards a corporate approach to crowdsourcing is the development of an agreement on a shared set of terms. Some shared terms and conditions can serve as guidelines for best practices and direction towards success. Technologically, and in agreement with Chhay and Kleiner (2013), crowdsourcing performance can be improved by refining the communication strategy in place. That is to say, there should be multiple and safe communication channels to ensure that there is seamless communication between the distributed virtual team members. For instance, a hybrid of satellite and fiber internet connection provides a failsafe for internet outage if one of the modes fails.
Retaining skilled participants and top-management employees is a taxing endeavor but can be addressed from various perspectives. Industry research makes it evident that bureaucratic structures and authoritarian leadership contribute to participant turnover because it lowers commitment and devotion. Accordingly, participants can be retained by motivating them to increase their performance regardless of the hurdles facing them within crowdsourcing work environments. Likewise, the operations managers should create structures and processes that encourage joint decision-making. In other words, if a requester wants to ensure that its crowdsourced participants remain in the organization, it has to identify the vital aspects of the company or its HR department that makes a developer want to stay and emphasize them. Apart from pay, these include job security, pleasant working environment, desirable benefits and an opportunity for learning and career advancement. Furthermore, the requester should try to minimize the influence of external factors that may entice existing developers and executives to leave by making the roles as desirable as possible.
Evaluating the Skill Set and Quality of Code Submitted By Potentially Unknown Users
Software engineering and interface design play a vital role in industry, government, commerce, entertainment, research, education, communication systems, medicine and many other areas of contemporary society (Grier, 2005). Ineffective interfaces undermine financial, energy, defense, health care, and other critical infrastructures. For this reason, it is important to evaluate skill sets and quality of code submitted by anonymous developers. Additionally, it is essential to perform a background check of the developers to eliminate the threat of inside jobs, terrorism, and theft of intellectual property. In regards to software quality, all codes submitted must be checked for quality by independent verifiers. Most importantly, all developers should follow all the predefined software quality standards as directed in the working contract.
Whether crowdsourced developers are new to interface security, or they are familiar with the existing risks, they must note that designing and coding secure applications or patching the existing ones is demanding. To manage the design risks, the developers must define the security measures for the interface design. In other words, security interface requirements must be identified either internally or externally, but under strict guidelines. It follows that guidelines on interface design must be predefined before the actual design. Additionally, the developers must be familiarized with the company’s quality assurance policies. Furthermore, the HCI architecture must be defined at the beginning of any software development. Moreover, the designed interface must be verified through application code review and real-time testing of the interface. Quality standards are used to establish whether the predefined quality objectives were achieved. For this reason, the established quality team should identify the quality standards related to the interface design project. After the project schedule and tasks have been highlighted, the interface design project manager will identify quality requirements and standards for interface project (Schwalbe, 2009).
Additionally, the interface project manager will document how the interface design project will demonstrate compliance with these identified software standards and requirements. This will be delivered through a crowdsourcing quality team. The quality team will be responsible for performing quality checkpoints, as well as verification of achievements of the interface project design milestones. The quality team may consist of the HR manager, performance manager, QA administrator, and the software development manager. The diversification is a strategic initiative meant to provide expertise as well as a sense of balance when providing checkpoint reviews. Nevertheless, the Project Manager should ensure that all the crowdsourced developers comply with company’s foundational policies and guidelines. Irrespective of the project scope and the size of the crowdsourced team, the PMO Quality Policies should be used as the foundational quality standards for a project. The idea of PMO is service focused on quality and customer satisfaction. To guarantee process quality, attention should be given to comprehensive documentation of the interface project design. Software test and review meeting need to be regularly held to ensure that processes flow smoothly and misunderstandings are eliminated. Additionally, internal audits regarding finances and human capital should be performed regularly.
According to Jamwal (2010), quality entails all aspects and essential features of the software related to the fulfillment of the presented user requirements. Software Quality Factors (SQFs) are features that software users consider substantial (Jamwal, 2010). When code quality factors are connected with crowdsourcing system quality metrics, they present a crowdsourcing software quality model. Normally, code quality factors can be expanded into functionality (generality, security, capabilities, and feature set), reliability (predictability, recoverability, severity/frequency of failure, mean time to failure and accuracy), usability (documentation, aesthetic, human factors, consistency), supportability (adaptability, testability, compatibility, reconfigurability, testability, extensibility, maintainability, and serviceability), and performance which entails efficiency, speed, response time, throughput and resource consumption (Ghayathri & Priya 2013; Jamwal 2010). The quality factors mentioned above can be used to evaluate the quality of the software code submitted by all developers to ensure uniformity. Speaking of quality factors, conciseness is the ability of code to satisfy its functional requirements utilizing a number of resources. A software code is said to be concise if the developer presents a code segment that does not have irrelevant information. Code consistency can be either internal or external.
According to McCall’s software quality model, internal consistency is the ability of code or program to meet user specifications. In contrast, external consistency is the degree to which a software code contains constant symbols, terminology, and notation. It is evident from the definition that internal consistency is important for the coding phase. Accuracy is another code quality, which refers to the extent to which the code’s output is precise in order to satisfy its expected use. Expandability is the efforts required to extend the software’s performance or capabilities. In McCall’s Quality Model, the quality of crowdsourced software quality will be measured subjectively based on developers or answering questions. As highlighted by Jamwal (2010), not all interphase design traits are covered in this quality model. For example, the architectural integrity of the interface design is included as an aspect of quality in the model. Nevertheless, some of the aspects measured include modularity and simplicity (Jamwal, 2010). Consequently, this quality model is recommended for crowdsourced interface designs.
The Impact of Crowdsourcing on Budget and Timeline of a Design Project
The impact of crowdsourcing on timeline and budget relates to various factors, including the level of expertise, the underlying technology, risk factors, geolocation of the participants, and projects scope. Similarly to any other project, planning is essential to eliminate project creep in terms of time and budget. Companies that crowdsource globally are much more likely to swell their budgets and elongate their schedule compared to companies embracing enterprise crowdsourcing. The rationale is that more time, human and financial resources must be allocated for external mobile crowdsourcing compared to internal crowdsourcing whereby the existing infrastructure and human capital come in handy. As expected, the level of technology and security concerns also contribute to the operating costs and investment expenditures. For financially stable companies such as Apple and Samsung, enterprise crowdsourcing would have an insignificant impact on their budgets and as well as timelines because project templates will be used. New entrants, however, have to consult experts not only in the field of crowdsourcing but also in project management as a measure to ensure that their project schedules and budgets are not overstretched.
One of the most important aspects of controlling and monitoring a crowdsourced design project is managing the project costs against the approved budget. In this regard, the budget is defined as the amount of money that has been authorized for use on a crowdsourced design project. Irrespective of scope creep or potential inflation in project costs, the budget amount always remains the same. As an IT project, procurement in the design project entails purchasing products, work, and services (Adesanya, 2008).
In crowdsourcing, the procurement route should define the aspects of a requester’s relationship with the contracted interface design team and coding participants. Project delivery entails getting quality projects completed on budget and in time as well as ensuring that informational and tangible assets are maintained long-term (“Project delivery strategy,” 2010). By nature, crowdsourcing procurement is a delivery route that creates, manages, and fulfills contracts relating to the provision of supplies or engineering and construction works, hiring, acquisition, and disposal. This definition compliments the argument that crowdsourcing can impact the project budge or schedule based on the procurement path. As a process, crowdsourcing procurement can be documented as understandably related actions performed or occurring in a definite manner that results in the attainment of a milestone or completion of a deliverable (Adesanya, 2008). In this context, the milestone is the development of an interactive interface for a crowdsourcing platform. Project cost management entails the processes involved in planning, estimating, budgeting, and funding so that the project can be completed within the scheduled time and approved budget. As a strategic effort, design project costs should be broken down using the work breakdown structure (WBS) which serves as the basis for the cost account. The WBS break down design project costs to a manageable and reportable level. As part of the project planning, the cost is allocated at the task level for all materials, equipment, and labor. Additionally, provisions should be made for all the identified risks.
In project management, budget analysis is an umbrella term that refers to a set of tools used to manage a budgeting process. The budgeting process entails allocation of resources to various task and risk assessment. In essence, the project manager must regulate the project budget well to fulfill its objective. The tools and techniques used in cost estimation and management vary depending on the complexity and phase of the project. This section describes some of the methods that can be used by project managers to estimate the cost of an interface design project and create a budget. Cost estimation is essential in comparing and selecting projects. During the initiation phase, economic factors can be used to choose between competing projects. An estimate of the cost of each project is made to compare the internal rate of return or paybacks between projects. The estimates should be accurate for the comparisons to be meaningful. Furthermore, some resources and times used in estimating costs should be appropriate for the complexity and size of the project. The methods used in estimating project cost during the conception phase are often faster and utilize fewer resources than the methods used to create detailed estimates in the other phases of the project.
Legal, Societal, and Ethical Issues Raised by Crowdsourcing
Similarly to any technology-based products, software crowdsourcing presents some potential dangers. Arguably, software crowdsourcing is an innately risky method of interface design. To underscore this observation and assist in illustrating the ethical implications of software crowdsourcing, software development can be viewed as an empirical process (Wohlin et al., 2012). In reality, experimental processes go beyond controlled development environments to cover humans on a social platform. Numerous risks exist in almost every phase of software development because of limited resources (Vernon, 2014). For this reason, crowdsourced software products are tested for a predetermined number of cases. It follows that the crowdsourced interface design is used in real life conditions that were never tested during the software development lifecycle (SDLC). Logically, the crowdsourced interface designers expect their design to function safely even in uncontrolled scenarios. These observations induce issues of professionalism, ethics, and legality. Essentially, ethics entails a set of beliefs, values, habits and attitudes revered by an entity or group (British Computer Society, 2014). In software engineering and particularly crowdsourced software development, legal and ethical disputes surface more often when there are differences of expectations regarding what constitutes the esteemed state of software engineering affairs or right course of action.
As crowdsourcing gains traction, there are also ethical, societal, and legal risks. Firstly, risk emerges from entities that exploit the public for company benefit and legal questions emerge over the rightful owner of the resultant intellectual property. Given that crowdsourcing is a process harnessing collaboration from the internal and external environment of a company to attain innovations, solve problems and improve efficiency at different levels across various industries, there is a chance that ethical, societal and legal elements will intersect. In other words, the external environment induces new ethical, legal and societal risk to the crowdsourcing company. For example, getting results from software crowdsourcing entails tapping human capital with diverse backgrounds including nationality, religion, race, sexual orientation, gender, and age. For this reason, numerous demographic factors come to play in regards to laws, ethics, and cultural values. In the same perspective, crowdsourcing may be exploitative in nature due to variance in economic development and technological capability.
This is evident especially in the difference of remuneration based on nationality. Unsurprisingly, a developer in Latin America, India or Africa is more likely to be paid less that a developer at the same level in the EU, Canada or the United States. People are ardent about being part of a software development community. They will contribute and may never realize that the crowdsourcing company is exploiting them. Legal questions over intellectual property (IP) rights and the party that should claim ownership of the resultant product or content are increasing. As of this writing, the majority of the companies engaging in crowdsourcing had binding contracts with the participants, which stated that the companies own the IP rights. Participants can retain the IP rights in only a few cases. Observably, the challenge is in the allocation of shares equivalent to the value of the input of each participant. The other challenge relates to the lack of laws distributing and protecting ownership of ideas of individuals participating in human computation.
Alleviating Concerns Raised by Legal, Ethical, and Societal Issues Linked to Crowdsourcing
To address concerns discussed in the preceding section, companies should design crowdsourcing methodology with the ability to not only create value but also protect the ideas of the participants. Arguably, most brands that crowdsource are inexperienced in this specific aspect. Therefore, moving forward, these companies should consider a consultative approach to crowdsourcing. Failure to do so exposes crowdsourcing companies to legal and ethical risks with their participants. Before any laws are passed to protect participants in crowdsourcing, companies should develop structures for allocating ownership. Additionally, gamifying crowdsourcing can create a transparent way of allocation points to the contributors. Gamification entails rewarding participants (Burke, 2011).
Due to diversity regarding culture, age, nationality, religion and race, it is vital for executives in crowdsourcing companies to develop and enhance their leadership competencies and cross-cultural communication skills to perform effectively on their crowdsourced interface design projects, which can be delivered through training. The goal of crowdsourcing based training programs is to enhance managers’ skills and knowledge, as well as their cultural awareness for the benefit of the organization by facilitating effective teamwork and establishing a warm and healthy work environment encouraged by recognizing and tolerating differences in culture among people. According to Osland and Bird (2006), such a training program would provide leaders and managers with an opportunity to learn more about diversity beyond their company’s internal environment. Even though diversity trainings may fail to change cultural values of an individual, the training can catalyze the improvement of diversity awareness, transfer knowledge, and educate leaders in ways of accepting diversity (Osland & Bird, 2006).
There is a growing interest in the application of crowdsourcing in the field of Information technology, as well as in the professional business landscape. After its invention, the development of crowdsourcing in the realm of software engineering in general and HCI design, in particular, has grown exponentially. The rationale of software crowdsourcing is to harness the collective intelligence and skill sets available on cyberspace through the Internet. Normally, an entity would perform enterprise-related tasks internally or outsource them to a third party. Remunerated crowd work provides remarkable opportunities for improving social mobility, productivity, and the global economy through collaborative engagement of a diverse and geographically distributed workforce to perform complex tasks at scale and on demand. Given that crowdsourcing is dependent on Internet-enabled technologies, it follows that the methodology is vulnerable to cybercrime threats and risks. Additionally, scalability and availability of the Internet enabled platforms is also an issue. Besides the technical issues, there is a number of legal, societal, and ethical issues including privacy concerns, the anonymity of workers coupled with poor quality of work due to the pressures of quantity. Furthermore, distributed locations of team members, working on a single design project, call for the purposive approach in the interpretation of various statutes, including labor, privacy, and intellectual property rights.