lish themselves over time, by building a track record of trustworthiness.
On the other hand, the problem of
conventional signals is they are not
always reliable, and hence may cause
problems of biases and stereotypes.
These t wo types of signals together on
different platforms may have a mixed
effect of making people comfortable
engaging in the platform in the first
place, while at the same time, causing
some biases and discrimination, albeit sometimes unintentionally.
From a HCI point of view, platforms
that facilitate the gig economy should
distinguish assessment signals from
conventional signals, in order to highlight the right kind of information at
critical points when people are making
decisions. For example, when requesting
to pool in an Uber, or deciding whether
to pick up a passenger, the neutral information to rely on to make the decision
should be the efficiency of the route,
rather than some factors that may cause
systematic bias and results in redlining,
underserving a specific population (such
as those with a lower economic status).
The gig jobs of today are different from
the gig jobs of the 1950s and a key role
of policy makers is to extend the pro-
tections that regular jobs afford to gig
jobs. There is also a parallelism be-
tween the digital revolution we are cur-
rently experiencing and the Industrial
Revolution. The points of similarity are
at the moment limited to the social and
economic tensions generated by ac-
celerating technological development.
The history of the Industrial Revolution
is one of pain, misery, and poverty at
the beginning and of rising living stan-
dards at the end. Is the gig economy a
first step in a similar progression? Or is
the technology of today fundamentally
different from the technology of the
past, in that machines are becoming
more intelligent and substituting hu-
mans? Nobody has a definitive answer
to questions of this type.
We suspect, however, the machines
will not be able to replace humans anytime soon. Indeed, very hard problems,
such as whom to trust and how to measure the trustworthiness of others, do
not seem to go away. Actually, because
of the nature of the gig economy, problems of this kind appear more prevalent than ever before. We highlighted
some designing principles for creating
platforms that facilitate trust. A suggestion for students interested in creating
socio-technological systems is thus to
focus not just on the technical aspects
that make a website work, but also on
the social elements that facilitate interactions in a platform. We think combining sociological imagination with
technological skills is a fundamental
step for achieving a successful design.
[ 1] Sundararajan, A. The Sharing Economy: The End
of Employment and the Rise of Crowd-Based
Capitalism. MIT Press, Cambridge, 2016.
[ 2] Nunberg, G. Goodbye jobs, hello ‘gigs’: How one word
sums up a ne w economic reality. NPR. January 11,
[ 3] Bessen, J. Learning by doing: The real connection
between innovation, wages, and wealth. Yale
University Press, New Haven, 2015.
[ 4] Riegelsberger, J., Sasse, M. A., and McCarthy, J.
D. Shiny happy people building trust?: Photos on
e-commerce websites and consumer trust. In
Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems (CHI ‘03)(Florida,
April 5-10). ACM, New York, 121-128,
[ 5] Donath, J. Signals in social supernets. Journal of
Computer-Mediated Communication 13, 1 (2007),
Paolo Parigi is the associate director of computational
social science at IRiSS, Stanford University. Parigi is
interested in studying trust in online communities. He is
currently trying to find ways to quantify the impact of trust
among Uber users. He has a Ph.D. in sociology and an M. A.
in quantitative methods, both from Columbia University.
Xiao Ma is Ph.D. student in information science at Cornell
Tech, and a past research intern at Airbnb. Her research
focuses on self-disclosure, identity, and trust in online
communities. Ma has a B. S. in microelectronics from
Peking University in Beijing, China.
© 2016 Copyright held by Owner(s)/Author(s).
Publication rights licensed to ACM.
creates challenges for people on both
ends of the job market when assessing
the trustworthiness of each other. After
all, with all these choices of potential
people who can do the job, why should
one choose one over the other?
The design of platforms becomes
of key importance for how people interact in the gig economy. Specifically,
we highlight how people use different
signals to assess the trustworthiness
of other participants on the platform,
and how these socio-technological
platforms should take into consideration how people communicate and
form impressions online in order to
design a neutral platform where the
“good” workers can be rewarded while
the “bad” actors be punished.
Take the design decisions around
photos for example. We know from different eye tracking studies that online
photos with faces attract people’s attention. Riegelsberger and colleagues
found “the presence of photos reduced
participants’ ability to identify vendors
with good and bad reputations—the
perceived trustworthiness of poorly
performing vendors was increased,
whereas that of vendors with good
reputation was decreased” [ 4]. With
the “vendors” in the gig economy becoming individuals, the platform as a
marketplace should design carefully,
by providing the right piece of information to users when they are making different kinds of decisions.
One useful framework to address
this design challenge gig economy platforms face is signaling theory. According
to signaling theory [ 5], there are different types of signals: some are assessment signals, where the signal actually
reflects the quality of the characteristics
it represents; others are conventional
signals, which is when the connection
of signal to the quality of characteristics
it represents is merely through convention, such as making promises. In the
context of a sharing economy, an ideal
reputation system with reviews provides
signals that are in principle assessment
signals. While profile elements, such as
names, photos, as well as self-descrip-tion constitute conventional signals.
The problem with assessment signals is the “cold start” problem that
all reputation systems face. Participants of the platform need to estab-
Policy makers have
a fundamental role
to play that is
from what most
the gig economy