ArticlePDF Available

Abstract and Figures

Will the digital revolution actually transform the process of innovation? A professor from NYU spent three years with NASA's engineers and scientists to uncover the significant opportunities and challenges involved with new models for R&D work.
Content may be subject to copyright.
XRDS • WINTER 2 016 • V OL.23 • NO .2
Where Are
the Flying
Will the digital revolution actually transform the process of innovation?
A professor from NYU spent three years with NASA’s engineers and scientists
to uncover the significant opportunities and challenges involved with new
models for R&D work.
By Hila Lifshitz-Assaf
DOI: 10.1145/3014050
It’s the 21st century, and despite an incessant buzz around innovation (not to burst your
bubble), we see no flying cars, and have not returned to the moon since the ‘70s. Can the
new web-enabled models of innovation accelerate the pace of research and development
(R&D) on these future-enabling technologies? What does it mean to be an engineer or a
scientist in such a future?
Growing up, I hoped to fly on a daily basis with an android companion, and yet today I
still drive to work and write on a computer that is only a thinner and faster version of the
one I had as a kid. The only robot I have is a floor vacuum cleaner, which often gets stuck.
What really happened? Is this the 21st century we dreamt of? This frustration of
being nowhere close to our dreams led
me to investigate how the digital revo-
lution might change the process of sci-
entific and technological innovation,
and bring the future into the present.
In order to understand the future of
R&D, I first want to put things in a his-
torical perspective. Innovation was ini-
tially led by the lone-inventors, such as
Leonardo De Vinci and other famous
early thinkers, who worked in their lo-
cal communities. Then, the Industrial
Revolution hit in the 18th century, and
the first labs were born. Ever since, in-
novation has been initiated mainly by
experts organized as groups, and by
labs within large private and public
organizations. Innovation has mostly
been a product of such organizations
and their collaborations. Could this be
the primar y reason we are still stuck—
the tunnel vision of experts, or the dis-
abling nature of bureaucracy in such
organizations? Some assert now is the
time for change, to democratize inno-
vation, and to use tools that digital rev-
olution gave us to open the boundaries
Image b y Esteb an De Arma s
XRDS • WINTER 2 016 • V OL.23 • NO .4 59
Phot o Credit T K
XRDS • WINTER 2 016 • V OL.23 • NO .2
the Amazon, I set out to the Houston
Space center, equipped with my field
research methodology tools, and
ready to delve into the world of space
and open innovation.
I designed my study as an in-depth
longitudinal field study to be con-
ducted at NASA’s Space Life Science
Directorate (SLSD) and in its related
units between 2009—2012 (with pe-
riodic follow ups from 2012 to pres-
ent). The SLSD focuses on research
and development in life sciences. Its
mission is to optimize human health
and performance through all phases
of spaceflight especially in response
to risks, such as radiation, bone loss,
and vision impairment (challenges
similar to those described in the
movie “The Martian”). In the three-
year time period, I collected data and
conducted iterative cycles of analysis
and collection; both pre- and post-
open innovation experiments. As the
study was designed to be longitudi-
nal, I collected all data before, dur-
ing, and after the open innovation ex-
periment. My aim was to understand
the day-to-day experience of organi-
zational members working in open
innovation context. I, therefore, col-
lected data from abundant sources,
both qualitative and quantitative. My
primar y data sources came from field
research methods like obser vation,
interviews, and project work docu-
ments. I sought to immerse myself
fully in the work-world of the people I
was studying. For instance, I partici-
pated in a summer training session
for interns, sat in weekly meetings,
spent entire days in each of the nu-
merous R&D units and labs, joined in
conferences, as well as participated
in the debriefing to astronauts on
their last shuttle flight, which was
also attended by friends and family. I
of innovation process to everyone. Will
the digital revolution be as transfor-
mational to the innovation process as
the industrial one?
The new web-based models for in-
novation, usually referred as open,
peer-production, or distributed in-
novation [1,2], is inspired by the
open-source software movement.
This movement has demonstrated
the possibility of successful innova-
tion outside the traditional econom-
ic, and organizational boundaries.
Worldwide, thousands of individu-
als freely develop highly sophisticat-
ed products, successfully competing
with dominant designs of an indus-
try laden with proprietar y software
products. This phenomenon is now
spilling over from software design
to a wide array of product and ser-
vice classes, and makes me wonder
if the same philosophy, organiza-
tional structure, and openness can
also produce scientific and techno-
logical innovation? As the digital
footprint of many products increas-
es and computational sciences be-
come prominent, many argue that
this will be the future of R&D. In or-
der to better understand this poten-
tial, and the future of R&D, I went to
NASA and researched its R&D pro-
cesses for almost three years. This
was before, during, and after they
experimented with open web-based
models for innovation as an approach
for solving their most strategic R&D
challenges. While there I conducted
an in-depth longitudinal field study,
which uncovered major opportuni-
ties and challenges involved in the
open innovation model.
I began my study by first searching
for leading organizations that were
experimenting with new models of
innovation. After inter viewing many
such companies in various sectors
(such as IBM, Starbucks, Pfizer, Lego,
and Procter & Gamble), I realized the
open innovation model is often used
for marketing reasons, with no in-
teraction or impact on the core R&D
process of the organization. How-
ever, upon meeting the head of Space
Life Science at NASA (the U.S. Na-
tional Aeronautics and Space Admin-
istration), I realized their approach
was very different. NASA tackled
their most strategic R&D challenges
by experimenting through new on-
line, open innovation models. They
designed the open innovation ex-
periment and held it simultaneously
and in parallel with the traditional
models to attempt to solve their tech-
nological and scientific R&D chal-
lenges. This meant their R&D profes-
sionals were deeply engaged in the
experiment and formulated chal-
lenges for open innovation online
platforms. This willingness to test
new models of R&D fascinated me.
Akin to an anthropologist going to
Resource allocation,
and attribution
and award systems
all focus exclusively
on problem
solving and not
on solution seeking.
Table 1. The open innovation platforms and communities used by NASA Space Life Science Directorate
Platform Established
Solvers Type of Problems Posted Range of Participants
Average Range
of Awards
Innocentive 2001 355,000 + Scientific and technological
problems, modular and complex
Wide range- scientists, technologists and
business (Most with advanced degrees)
TopCoder 2000 400,000 + Computer science and web design
problems, mostly modular
Software engineers, computer scientists,
Yet2 1999 130,000 + Technological problems,
mostly modular
Engineers, technologists, entrepreneurs
and small businesses
5k$- 100K$
XRDS • WINTER 2 016 • V OL.23 • NO .4 61
would start my mornings by reading
space news and went to bed reading
about life on Mars and why haven’t
we made it there yet. I interviewed
all the scientists and engineers in
relevant labs; they shared their proj-
ect documents, which enabled me to
study further. I proudly wore a NASA
name tag, albeit temporary, and
people generously welcomed me into
their professional world, for which I
am forever grateful.
Once the open innovation experi-
ment began, I collected data not
only from NASA, but also from open
innovation platforms. For instance,
when looking at the external solv-
ers who participated in the experi-
ment, I gathered data on type and
number, professional background,
geographical location, proposed so-
lutions, award amounts, and more.
This comprehensive data collection
enabled me to study the interaction
between the traditional R&D mod-
el—that is, worldwide experts work-
ing in a leading organization—and
the new world of online communi-
ties, hosting a wide range of pro-
fessionals and amateurs looking to
solve challenging problems.
NASA SLSD’s R&D professionals de-
cided to experiment with the usage
of open innovation platforms and
communities to seek solutions for
some major R&D challenges in strat-
egy planning for that year. The ex-
periment, which was expected to last
one year, was conducted in parallel
with traditional R&D work on these
problems. For one year, NASA’s R&D
professionals worked on these prob-
lems both internally and contract-
ing and collaborating with domain
experts from other organizations,
following the best practices of tradi-
tional R&D model. Simultaneously,
they posted these challenges online
on three of the most leading open
innovation platforms and commu-
nities: Innocentive (https://ww, Topcoder (https://
ww, and
( (see Table 1).
On these platforms, anyone can try
and solve these problems, usually
members of such online communi-
ties come from a range of industries
and have diverse professional back-
grounds (from students, to working
professionals and retired profession-
als, spread across the globe). The R&D
professionals met with the platform
providers a few times, who began the
process by providing a one-day intro-
ductory workshop, and then created
short problem descriptions that were
posted online for two to three months
on open platforms.
In total, 14 R&D problems from 11
R&D units related to a variety of sci-
entific and technological fields like
microbiology, heliophysics, mechani-
cal engineering, radiation, material
science, medical devices, and more,
were posted on the three open plat-
forms. These were the most impor-
tant and strategic R&D challenges for
that year. One hundred NASA profes-
sionals along with their collabora-
tors and contractors worked on these
problems; while 3,000 individuals
from 80 countries belonging to vari-
ous industries also worked on the
same problem set on the open inno-
vation platforms (see Figure 1). Three
months after the experiment began,
approximately 300 solutions were
submitted. The quality of proposed
solutions exceeded expectations and
surprised NASA’s R&D. Upon evalu-
ation, it was concluded that three
problems were solved completely and
four to six, partially. All the solutions
were turned in within astonishingly
short timeframes with a few even sur-
passing their solution criteria. InThe
speed and success of the open proc-
ess was mesmerizing.
One solution in particula r, b ecame
known as the “home run” of the open
innovation experiment. Prediction
of solar particle events, popularly
known as solar storms (see Figure 2),
is a well-known and well-researched
problem both at NASA and within the
global heliophysics community. So-
lar storms are extremely dangerous
to current and future space missions
and are considered a known threat
to satellite-based communication.
A severe solar storm could disable
satellites and the Internet, ground
all aviation, silence telecommunica-
NASA tackled
their most strategic
R&D challenges
by experimenting
through new
online, open
innovation models.
Figure 1. The geographic spread of external solvers who tried solving NASA’S R&D
problems at the open innovation experiment: 3,000 solvers from 80 countries.
XRDS • WINTER 2 016 • V OL.23 • NO .2
Figur e 2 image by N ASA/J PL.
the discipline and traditions of helio-
physics. Using ground radio-based
equipment instead of the traditional
satellite-based data, he created an
algorithm that could forecast solar
flares eight hours in advance, with 75
percent accuracy and a three-sigma
confidence level, well beyond the ex-
pected result. When the heliophysi-
cists and radiation professionals at
NASA tested his solution on their op-
erational systems, they achieved even
higher accuracy between 80 to 85 per-
cent. The head of the R&D unit that
worked on this problem was stunned.
However, the story does not end
with a rosy picture of scientific break-
through, ribbons, and awards. Adopt-
ing the open innovation model raised
several serious challenges that nei-
ther NASA’s management nor its sci-
entists and engineers had foreseen.
In particular, it led scientists and en-
gineers to re-examine their roles and
professional identities.
tions, and damage the electric grid.
Significant financial and intellectual
resources are invested by the helio-
physics community and radiation
experts, within NASA and worldwide,
towards the development of better
solar events forecasts. At the time,
the best algorithms could predict
a flare just one to two hours in ad-
vance, with 50 percent accuracy and
a two-sigma confidence interval. The
open innovation challenge sought
an algorithm that could predict an
event anywhere between four and 24
hours in advance. This problem was
posted in December 2009, with an
award amount of $30,000. Within a
three-month period, more than 500
individuals expressed interest in
trying to solve this problem, and 11
submitted solutions. The winning
submission came from a semi-re-
tired radio engineer from rural New
Hampshire. His solution brought an
approach that was entirely outside of
Figure 2. An example of a solar storm.
XRDS • WINTER 2 016 • V OL.23 • NO .4 63
edge products. This issue led me to
join a cross-disciplinary study that
investigated how individuals take
roles upon themselves in Wikipedia
and whether over time, these roles
create a cumulative stable production
of knowledge. We found at an indi-
vidual level, there was a high degree
of mobility across roles and articles.
Whereby individuals transitioned be-
tween roles and ar ticles, and many got
into the platform quickly, took a role
in one article, and then left. Despite
this freedom and mobility on the in-
dividual level, we found at the orga-
nizational level, over time, there was
stability for these emergent roles. We
called this dualistic nature of knowl-
edge co-production “turbulent sta-
bility” [5]. These results suggest the
plausibility of new ways of organizing
for knowledge production, without
traditional management mecha-
nisms. Perhaps organizations need
to see themselves as platforms with
a small team of full-time employees
working to orchestrate knowledge
production and a large base of in-
dividuals who choose the right job
for themselves. Think about it, if we
each choose the right project and the
right role for ourselves, will our cre-
ative prowess be unleashed and R&D
productivity skyrocket?
[1] Benk ler, Y. The Wealth o f Networks : How Social
Production Transforms Markets and Freedom. Yale
Univ ersit y Press . New Haven , Connec ticut , 2006.
[2] Von Hipp el, E. Sour ces of Innov ation. Oxford
University Press, 1988.
[3] Lif shitz-A ssaf, H. Di smantli ng Knowl edge
Bound aries at N ASA: F rom Probl em Solve rs
to Sol ution See kers. Av ailable at S SRN:
http :// stract= 2431717, 2016
[4] The Ne xt Rock et Scient ist: YO U. http s://open.n asa.
[5] Ara zy O., Dax enberge r J., Lif shitz-A ssaf H., N ov O.,
and Gur evyc h I. Forth coming. Tur bulent St abilit y
of Eme rgent Ro les: The Du alistic N ature of S elf-
Organizing Knowledge Co-P roduction. Information
Syste ms Researc h (ISR).
Hila Li fshit z-Assaf is a n assist ant prof essor of
information, operations, and management sciences
at New Yo rk Unive rsity S tern Scho ol of Busine ss.
She is al so a facult y associ ate at the B erkman
Cent er for Int ernet and S ociety a t Harvar d Univers ity.
Her re search r eceived m ultiple a wards and
commendations, including the cross disciplinary
INSP IRE grant f rom the Nat ional Scie nce Founda tion.
© 2016 ACM. 1528-4972/16/12 $15.00
Moving forward to integrate open in-
novation at NASA, turning the experi-
ment into a day-to-day reality proved
a far greater challenge than expected.
The open web-based models raised
multiple questions for the scientists
and engineers. What does it mean to
use such a model to solve all of their
work projects? How will their roles be
defined then? Should they continue in
the same work model (in a lab, in an
organization), or instead work like the
scientists and engineers in open, on-
line communities? These questions
had no clear answers. The open inno-
vation communities were preoccupied
with their online participants. This left
a lot of room for interpretation. Some
scientists and engineers adopted this
model and re-invented their role while
others saw it as a threat to their iden-
tity and how they were trained and re-
jected the model completely. This abil-
ity to re-think one’s role turned out to
be crucial in adopting new web-based
innovation models [3].
Those who adopted this model
and became orchestrators of external
innovators invited everyone to solve
NASA’s problems, as they themselves
developed new capabilities and skills
to manage this new way of conducting
R&D. They shifted their attention to
the individuals outside the boundaries
of their organization. One of the scien-
tists addressed this approach in a blog
post, calling on “you” (i.e., anyone) to
become the next rocket scientist:
For over half a centur y, NASA has in-
spired people across the world to look to
the heavens and wonder what secrets are
hidden within the cosmos. Solving those
mysteries has long been the domain of
lab-coat wearing scientists in govern-
ment agencies and universities. However,
with the advent of the Internet, social
web, and open source data, it has become
possible for anyone to make scientific
discoveries about our universe. Find out
how you can actively contribute to space
exploration and how the collective pow-
er of the Internet is enabling the future of
scientific research [4].
Indeed, if you, the reader, want to be
a part of solving NASA’s R&D challeng-
es, there are two important initiatives
aimed at opening NASA’s innovation
process and adopting web-based mod-
els for R&D currently visit open.NASA
( and the NASA
Tournament Lab (http://www.nasa.
This transformation raises ques-
tions about the type of training aspir-
ing engineers and scientists should
receive. The existing educational and
professional training programs teach
and reward individual problem solv-
ing, rather than producing solutions
by orchestrating crowds of solvers or
the integratng knowledge from differ-
ent disciplines. As suggested in this
study, a potential shift in the role of
an R&D professional offers important
implications for innovation-related
policies, given that currently, resource
allocation, incentives, and attribution
and award systems all focus exclusive-
ly on problem solving and not on solu-
tion seeking.
An alternative approach to open
innovation is not to define the roles of
scientists and engineers in organiza-
tions, but rather let them select their
role and their projects, similar to the
model of the open innovation on-
line platforms. These platforms and
communities produce knowledge
and innovation in a way that funda-
mentally stands in contrast to the
control-based approach of regular or-
ganizations. Many scholars and prac-
titioners question the efficacy of this
open approach and its ability to pro-
duce stable and high-quality knowl-
Some assert now is
the time for change,
to democratize
innovation, and to
use tools that digital
revolution gave us to
open the boundaries
of innovation
process to everyone.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Increasingly, new forms of organizing for knowledge production are built around self-organizing coproduction community models with ambiguous role definitions. Current theories struggle to explain how high-quality knowledge is developed in these settings and how participants self-organize in the absence of role definitions, traditional organizational controls, or formal coordination mechanisms. In this article, we engage the puzzle by investigating the temporal dynamics underlying emergent roles on individual and organizational levels. Comprised of a multilevel large-scale empirical study of Wikipedia stretching over a decade, our study investigates emergent roles in terms of prototypical activity patterns that organically emerge from individuals' knowledge production actions. Employing a stratified sample of 1,000 Wikipedia articles, we tracked 200,000 distinct participants and 700,000 coproduction activities, and recorded each activity's type. We found that participants' role-taking behavior is turbulent across roles, with substantial flow in and out of coproduction work. Our findings at the organizational level, however, show that work is organized around a highly stable set of emergent roles, despite the absence of traditional stabilizing mechanisms such as predefined work procedures or role expectations. This dualism in emergent work is conceptualized as "turbulent stability." We attribute the stabilizing factor to the artifact-centric production process and present evidence to illustrate the mutual adjustment of role taking according to the artifact's needs and stage. We discuss the importance of the affordances of Wikipedia in enabling such tacit coordination. This study advances our theoretical understanding of the nature of emergent roles and self-organizing knowledge coproduction. We discuss the implications for custodians of online communities as well as for managers of firms engaging in self-organized knowledge collaboration.
Full-text available
R&D professionals are known for keeping their knowledge work within clearly defined boundaries and protecting it from external non-professionals. The history of scientific and technological innovation is rich with cases of rejection of meritorious innovation when created outside disciplinary boundaries. Recently, a new model—typically called “open” or “peer-production” innovation—has been challenging these boundaries. According to the model, knowledge work should be conducted in the open by anyone who chooses to contribute. This study investigates how the open innovation model impacts R&D professionals and their work through a longitudinal in-depth field study at NASA. The open innovation model led to a scientific breakthrough at unprecedented speed; yet it challenged not only the knowledge work boundaries but also the professional identity of the R&D professionals. Only R&D professionals who changed their professional identity and refocused it from “problem solvers” to “solution seekers”, dismantled their knowledge work boundaries and shifted the locus of innovation outside the traditional boundaries. Adopting open innovation without a change in professional identity resulted in no real change in the R&D process. This paper reveals how such processes unfold and illustrates the critical role of professional identity work in changing knowledge work boundaries and shifting the locus of innovation.
Presents a series of studies showing that the sources of innovation vary greatly; possible sources include innovation users, suppliers of innovation-related components, and product manufacturers. These types of roles are known as functional areas. Specific areas of innovation are marked by having innovators predominantly in one specific functional area. Using empirical data from industrial histories, the authors show that this innovation-function relationship has held in scientific instrument, semiconductor and printed circuit board assembly process innovations. Users are predominantly the innovators in these fields. Also identifies a few industries where manufacturers are typically the innovators and a few others where suppliers tend to be. Analysis of the economic rents of innovation expected by potential innovators can often, if not always, by itself predict the functional source of innovation. Innovating firms will do so only when these rents prove attractive. Two factors suggest that this will tend to limit exploitation of the innovation to a functional area. First, it is difficult for innovators to adopt new functional relationships to their innovations. Second, innovators face a poor ability to capture innovation rents by licensing their innovation-related knowledge to others. This hypothesis and its implications are tested against the empirical datasets used initially. The role of informal R&D know-how trading is also discussed and analyzed using the Prisoner's Dilemma. Guidance is given to innovation managers and policymakers. (CAR)
With the radical changes in information production that the Internet has introduced, we stand at an important moment of transition, says Yochai Benkler in this thought-provoking book. The phenomenon he describes as social production is reshaping markets, while at the same time offering new opportunities to enhance individual freedom, cultural diversity, political discourse, and justice. But these results are by no means inevitable: a systematic campaign to protect the entrenched industrial information economy of the last century threatens the promise of today's emerging networked information environment. In this comprehensive social theory of the Internet and the networked information economy, Benkler describes how patterns of information, knowledge, and cultural production are changing-and shows that the way information and knowledge are made available can either limit or enlarge the ways people can create and express themselves. He describes the range of legal and policy choices that confront us and maintains that there is much to be gained-or lost-by the decisions we make today.