Needfinding in AMT: from the clicker games to World of Warcraft and the alternative Magic World - amdp

From crowdresearch
Jump to: navigation, search

After reviewing data and articles about AMT, and going through the AMT workers websites and reddit areas, the idea that comes out is about an underlying context-related background-motivation that binds workers to the AMT market. This background context starts off with the clicker-game addiction[1], then apparently supporting websites let workers find a new social world and the mirage they could influence the AMT context talking one another and having unfrequent conversations with their job workers.

The theory is that AMT workers hidden need is then a mixture of the behavioural automatism we can identify in completing tasks, so the mere completion of tasks and derivate satisfaction and realistic -not real- rewarding, and the feeling to participate to a social world which triggers the mirage of a magic similar and alternative social “nation” (not a case “The Turker Nation” is one of the most important website for workers) in which also the fight for some rights could have a more direct and effective agreement and compassion betwween people than in the real world. This remembers the addiction generated by “surrounding” world like World of Warcraft and seems to adhere to science-fiction apparent worlds like Idiocracy, The Truman Show, The Matrix. In the AMT world an easy world is proposed, made of simple tasks that satisfy and give real-istic benefits, but that has risks and difficulties in the long run as well as the external world, even though those risks start when the addiction has been already enacted.

We developed this idea also following the cooperation theory model by Merletti De Palo, A. et al.. [2]

We also want to remind that naming “Turks” or “Turkers” the AMT workers or any other worker that may do a simple task online could be offensive for the Turkish people.

1. Entering AMT: the clicker game starts

“On the other hand, Cookie Clicker gives you one rule—get more cookies—and waits in hiding to see whether you’re gullible enough to keep doing so. And naturally you do, because it’s a game and it’s there to be played. Cookie Clicker is the bleakest, and funniest, manifestation of that age-old warning, from parents and non-gamers, that playing games is a waste of time. It is a game with no end, no reward, no reason, and yet you play it anyway. Not because it’s fun or rewarding but because you want more cookies; because you’re invited to get more cookies. It is the worst game of all time, but also one of the funniest.

When I asked my friend at work whether I should give Cookie Clicker a go, he said that I shouldn’t. He had an unpronounceable amount of cookies, a Lovecraftian amount of cookies. He told me not to play Cookie Clicker, yet he had been playing it for weeks.

He also said, “If you play Cookie Clicker you’re trying to escape something in your life.” I was two weeks from a deadline, and horribly uninspired. I visited the Cookie Clicker website and started clicking. I got a grandma. Later, an antimatter condenser, then a time machine so I could get future cookies. I played Cookie Clicker for a very long time. I played it into the billions.

And I was not having fun.”

From “Cookie Clicker” and the Bleak Futility of Games by Shaun Prescott

Many surveys and articles seem to indicate that AMT workers (since now on: workers) do the tasks mainly for money. We personally want to go against this idea, or at least to integrate it, because we want to suggest that money is part of the motivation and not the only one. Moreover, money plays a relative role as motivation, as there are other meta-motivations that lay underneath the money motif. One of these is the behavioural addiction that simple tasks may trigger. Why would in fact the distribution of US AMT workers resemble that of the US nation? What could be so attractive as a one dollar per hour wage in people earning thousands of dollars per month?

As in the premise, clicker-games offer a simple start, but then they become addictive due to the fact the “missions” (using mission instead of tasks to be in line with World of Warcraft terminology) give them some enjoyment. In this aspect, we refer to Enjoyment as E = b+r as in [2]. Enjoyment is then the result of a real benefit and a relational/emotive benefit. In this case we have both: the (little) real benefit is the monetary reward, while the relational/emotive benefit is the completion of the task, pretty much like in clicker games. When we add some atmosphere and meaning to the task-mission, we get better results because of social, intrinsic and context-related motivations as described in “An Assessment of Intrinsic and ExtrinsicMotivation on Task Performance in Crowdsourcing Markets” by Rogstadius et al.[3] and in “Breaking monotony with meaning: Motivation in crowdsourcing markets” by Chandler and Kapelner [4].

2. The context gets its space

When the clicker-game triggers its dynamic with the help of the “realistic” idea that you get paid, the clicker game gets to a second level of interaction, because with the little payment its effect, wether little and apparently insignificant for most of the workers at first, it becomes parts of their lives with the monetary reward, becoming a clicker game which has “real” effects, even if just in the monetary order of few dollar cents. Compared to a Social Network, it enacts all the dynamics that people may feel in a real world social context thanks to external websites like The Turker Nation, but overcomes the basic SN aspects because of the “doing” aspect and because what you do there becomes real in your pocket, or, if we want to consider the low wages aspects, it becomes real-istic.

Following the Enjoyment composition model in [2], we could describe the seven conditions of cooperation in the AMT context:

Equivalence is maintained amongst workers according to the average payment, but is matter of strong discussion when it is not addressed like in the case of low paid tasks-missions compared to other task-missions. Equivalence is instead matter of discussion outside of AMT by comparing the standard outer world wages to those in AMT. Moreover, the tools in the hand of the requesters (the “bosses” or companies who create the task-missions) are way more impactful on the workers than vice-versa - “Whilst AMT provides means to rate Turkers’ reputations, there is no equivalent means for rating Requesters [..] Hasty judgments by Requesters result in unfair treatment of Turkers, and can rob them of pay for completed work and access to future work through potentially no fault of their own” (See “Being a Turker”, Martin et al., [5]).

Trust is a very important issue both for requesters and workers in AMT, overall when the trust legacy is broken by requesters denying the quality of the job and therefore avoiding the satisfaction-payment, and making the worker to risk to lose her reputation in the AMT system (after 5 rejections your possibilities of addressing HITs is so reduced you may choose to get out of the system). This generates both a relational and real benefit crisis and unleashes rage. If we observe the very low amount of economical advantage of these missions-tasks and the great anger generated we can start doubting money is the only real motivator. But trust (or fighting mistrust) is one small piece of the cake.

Care is a double sided aspect of the AMT context. If we see the idea that people can be treated with unfair manners or unfair evaluation, other than unfair wages, we could think of a negative context regarding care aspects. But there are hidden aspects which bring positive and motivational hints in play: examples could be the “sense of purpose”[6], or the “anonymity, flexibility to work when you want, for whom you want, on what HITs you want are all major incentives for working on AMT. Thus Turkers also orient to “positive invisibility” – the freedom from surveillance, control, and intervention in their personal affairs.”[5]

Transparency is well maintained in the context until it gets to work approval. The amount of money you get is there, the dynamic is always the same and hoaxes and transparency leaks happen only when some requesters do not approve the task-mission (called HIT in AMT).

Freedom compared to the outer world is a high valuable for AMT workers as they can choose when to work and in what conditions like sofa, bed, while eating, smoking, wearing any kind of clothes and remaining in their complete comfort zone even regarding races, religious beliefs and sex. The real problem for many of them is they do not see that they are inside a machine-generated momentum, a cycle that is ignited by the society in which they live and that make them to crave for something isn’t not going to give them a change but just a static dynamic from which they can’t apparently upgrade without leaving.

Common codes: the established language, english, and the similarities in the tasks contribute to a high level of common codes amongst the workers. We’d rather say this has anything to see with motivations, but it can contribute to make the AMT context simple and comparable.

Diversity has a low apparent space online, but a huge space in the workers life. The workers should homologate to the AMT procedures but they are quite free with no diversity limitations when they proceed to complete the task-missions

The alternative magic world

“Based on their responses to various media reports on AMT and Turkers, it is clear that while it is considered important for Requesters and Turkers to adhere to Amazon’s terms of service, they do not seem to want the US government to legislate and regulate AMT. This is not because they are happy with everything on AMT, but because the Turkers believe that their power to influence and manage the way the market works comes most fruitfully from their collective individual actions.” [5]

This concept of not needing any Government ‘invasion’ leads to the idea that the “Turkers” nation is a real social context that tends to be detached from the institutions and works as a “nation-like” set within the real world national set (or setting). We could see the AMT labour market as a CPR, a common pool resource, that if it has too many free-riders on both the requesters and the worker sides could become useless. According to Elinor Ostrom ideas of decentralization[7] it is quite normal that the need the workers have is not an external, “unaware” ruler who institutes new norms, but instead a facilitator-mediator which collects a set of norms agreed by the two parties (the requesters and the workers), and instead of monitoring them and make those norms being applied with force or punishment/reward systems, simply states them publicly and in transparence, allowing public monitoring to verify the relative respect of the rules.

In the case of Ostrom’s example, the utilization of CPRs was a process involving communities learning that lasted even centuries. The fact that AMT has become a “nation” in a very short time is quite peculiar. It means, like in the case of 2nd life or of the World of Warcraft, that the labour market online triggers social network aspects, and from those a whole sense of community and belonging. What are the motivations behind the “need for money” that lead workers, to invest their time and skills in the AMT labour market if the external labour market, especially in the US and in online jobs, could offer better possibilities?

“Absolutely unpredictable, great and terrible!”

The main idea is the need for an easy world: there is some difficulty in leaving an addictive alternative world made of easy satisfying tasks and a social environment, with those missions-tasks being easier to accomplish than the real world ones. It’s pretty much like clicker games or mission games like WoW. Entering the external labour market means aligning with a dress code, learning new things and facing your own lack of skill, avoid managing kids at home, marketing yourself instead of behaving the way you normally are used to do at home, commuting or simply passing a lot of time in an environment who you do not like.. in one simple expression, challenging ourselves at many levels. When some narrative and ethic emotion-triggering context is added to the mission-tasks, the action is comparable to the addictiveness we can find in the WoW context, where the game designers had the ability to create, explain and diffuse tasks the best way and generate the best context in which people could get addicted to the clicking-game dynamic.

Even in World of Warcraft, anyways, the idea is that you are in a better and easier world. Then you find the same dynamics and difficulties, the same lack of stability, but thanks to the micro-tasking both in AMT and WoW you get an easy satisfaction and that makes you happy anyways and willing to continue.

The competitiveness of the market makes such a pressure that even in the case of poor people, a very simple life made of a small room in co-housing with some restricted amount of food and water and the possibility to earn basic subsistence means with low effort, together with the social aspects of belonging to the category of the “turkers” in the “turkers nation”, can represent a way to satisfy the basic Maslow’s needs (Phisiological = food and water, Safety = not being fired and having to find another job, Belonging = social aspects + Turkers Nation, Esteem = requester’s approval, Self-actualization = missions accomplished). The same dynamics may work for rich people who have to cope with failures and personal regrets in a higher-class competitive context, or for carers, like mothers who do not want to leave their kids, to have granular tasks with filler-style possibilities to occupy the time without delegating the love-giving to external people. This goes quite strongly in the same direction as Riane Eisler’s requests to include caregiving activity into the GDP of a nation.

What is that AMT workers do not see?

At first, they don’t see that their efforts could become unpredictable. But even when they meet that eventuality, they go on, as the magic trigger already put them in the "cycle". And e-workers start to prefer e-jobs.

“My SO makes $20-$40 a day, but treats it like a "real" job. I'm a casual turker (my "real" job pays way more) so I'll make $100 every few months.” “I used to be able to make upwards of 50/week with a couple hours a day. Now I have an actual job, so my effort towards mturk has really dipped. I now make about 10-20/week.”

They just do not see the endless static dynamic of the AMT world, the WoW world, the Adventure Capitalist or Cookie Clicker clicker-game world, and of any other model of the real world that becomes addictive and does not allow our identities to grow and recognize themselves in their development.

Conclusion: the workers' needs

People need the crowdsourced work environment to have a narrative, a social world, they need it to be easy and to be an adventure. (in verbs: they need to enjoy, to find an easy context, to live stories, to find a social environment) They need those elements to be bound to the environment. Do we want to bind people to the current online labour market environment as it is?

The answer could be: only if we want people to fulfil their own missions, and not only the labour market ones. So we get to a deeper level of needs:

Workers need a world that makes working (both online and real) fair and not leading to endless and static cycling. (They need to evolve, to feel trust, to feel free)

Workers need a labour market that constantly improves their lives and their development desires. (They need to desire and accomplish; to recognize and identify themselves in what they accomplish)






[5] Being a Turker, Martin D. et al,

[6] Social desirability bias and self-reports of motivation: a study of amazon mechanical turk in the US and India by Judd Antin et al.

[7] Elinor Ostrom, Governing the commons