IEEE Technology and Society Magazine - June 2018 - 9

best practices of survey design and
academic division of labor. Eric Gilbert and Karrie Karahalios' essay
on developing a Twitter interface
based on theories of "social ties"
explores the difficult task of balancing the expectations of social science
research against the feedback from
user testing. Despite succeeding from
a technological perspective, a question of representativeness lingers
over the project, as start-up style beta
testing is not necessarily compatible
with norms of sample selection in the
social sciences. Following their essay,
Aaron Shaw's narrative of incorporating crowdsourcing into a large-scale
content analysis project discusses
the problem of user error in research
software and expands into a broader
reflection on how academic labor
is valued. But this decision comes
with its own time-consuming problems. Whereas professionally trained
research assistants often disagree
over interpretation, Shaw notes that
disagreements in crowd-sourcing are
often also the result of usability issues
in the project's software interfaces
and documentation. Shaw's account
of addressing miscommunications
between himself and software engineers on the importance of collecting
data in specific ways will ring true to
anyone who has participated in complex collaborations and is a must read
for anyone considering one, regardless of their role.
The collection also includes a trio
of essays on doing social research
using "big data" retrieved from online
sources. In their conversationally
styled essay, Michelle Shumate and
Matthew S. Weber offer two complementary, detailed defenses of software carpentry through a discussion
of programming web-crawling tools.
Their essay offers a look at the same
issues raised by those before them

june 2018

in the collection but from the other
gaining trust during interviews as conside of the interface. While the labor
siderations of how databases prejuof programming may not yet be
dice our sense of historical memory.
institutionally rewarded in
the humanities and social
sciences, Shumate and
Web er offer a strong argument for the ways that it
Each project involves a moment
can enrich those products
when a researcher could not
of research that are. Brent
Hecht and Darren Gergle
help but notice the ways their
extend discussion of this
topic by examining how disengagement with their project
parate data sources often
was thoroughly organized
make very different assumptions about metadata and
through software.
its presentation within a
database. Their essay raises
important questions about
the bias of certainty present in re If Hargittai and Sandvig have aimed
search involving big data. Similarly,
the book at fostering conversations
Brooke Foucault Welles' discusses
among seasoned practitioners in the
how larger data sets can actually
fields represented about the longcreate a greater degree of uncerterm effects of recent developments
tainty by examining how seemingly
in Internet-based technologies on
familiar concepts like "friendship"
their home disciplines, then in that
are understood via comparatively
regard it succeeds wonderfully even
narrow observational contexts. In
if every reader will take away someaddition to offering detailed looks at
thing different from it. Nonetheless,
the labor behind large-scale data coleach chapter is generally written in
lection and preparation, these three
a manner beneficial to specialists
essays show that close involvement in
and non-specialists alike, making the
the algorithmic processes performing
collection also an excellent choice
it will lead to increased confidence in
for an upper-level classroom, scholthe theories derived from it.
ars already wrestling with similar
Digital Research Confidential's
problems themselves, or technologibreadth, in short, is ambitious and
cal professionals looking to better
representative of the way that digital
understand the computationally-drivmedia research in the humanities
en research of their humanist and/or
and social sciences is often pepsocial scientist collaborators.
pered across academic departments
Reviewer Information
on most campuses rather than cenMichael L. Black is an Assistant Protralized under a single umbrella.
fessor of English as UMass Lowell,
Readers will likely find that not all
Lowell, MA. His current book project
subjects or methods offer insights
examines the influence of market
applicable to their own research.
pressures and technical discourses
For instance, someone interested in
on popular conceptions of software
combing through digital archives may
usability and personal computing.
not find the lengthy discussions on

∕

IEEE Technology and Society Magazine

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - June 2018