Working Alone Together Coworking as Emergent Collaborative Activity

Abstract

Mobile professionals can choose to work in offices, executive suites, home offices, or other spaces. But some have instead chosen to work at coworking spaces: open-plan office environments in which they work alongside other unaffiliated professionals for a fee of approximately $250 a month. But what service are they actually purchasing with that monthly fee? How do they describe that service? From an activity theory perspective, what are its object, outcome, and actors? This article reports on a 20-month study that answers such questions.

Full text

http://jbt.sagepub.com/
Technical Communication
Journal of Business and
http://jbt.sagepub.com/content/26/4/399
The online version of this article can be found at:
DOI: 10.1177/1050651912444070
published online 30 May 2012
2012 26: 399 originallyJournal of Business and Technical Communication
Clay Spinuzzi
Activity
Working Alone Together: Coworking as Emergent Collaborative
Published by:
http://www.sagepublications.com
can be found at:
Journal of Business and Technical CommunicationAdditional services and information for
http://jbt.sagepub.com/cgi/alertsEmail Alerts:
http://jbt.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://jbt.sagepub.com/content/26/4/399.refs.htmlCitations:
What is This?
- May 30, 2012OnlineFirst Version of Record
- Sep 13, 2012Version of Record >>
by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from by guest on October 11, 2013jbt.sagepub.comDownloaded from

Articles
Working Alone
Together:
Coworking as
Emergent
Collaborative
Activity
Clay Spinuzzi
1
Abstract
Mobile professionals can choose to work in offices, executive suites, home
offices, or other spaces. But some have instead chosen to work at coworking
spaces: open-plan office environments in which they work alongside other
unaffiliated professionals for a fee of approximately $250 a month. But what
service are they actually purchasing with that monthly fee? How do they
describe that service? From an activity theory perspective, what are its
object, outcome, and actors? This article reports on a 20-month study that
answers such questions.
Keywords
coworking, activity theory, knowledge work, telecommuting
1
University of Texas at Austin, TX, USA
Corresponding Author:
Clay Spinuzzi, Department of Rhetoric and Writing, University of Texas at Austin, 208 W. 21st
St., Stop B5500, Austin, TX 78712, USA
E-mail: clay.spinuzzi@mail.utexas.edu
Journal of Business and Technical
Communication
26(4) 399-441
ª The Author(s) 2012
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1050651912444070
http://jbtc.sagepub.com

The term coworking has been used with increasing frequency over the last few
years, often in books that describe business trends (Botsman & Rogers,
2011; Ferriss, 2009; Hunt, 2009). The first book on coworking (Jones,
Sundsted, & Bacigalupo, 2009) does not provide a definition although it
quotes many coworkers and space proprietors who describe what cowork-
ing means to them.
Nevertheless, people do seem to agree that some sort of service
called cowor king exists. Mor eove r, t he y ar e will ing to pay for this ser-
vice.InAustin,Texas,forinstance, at least 13 coworking spaces have
opened since 2009, and most of them charge about $250 a month to
allow professionals to work there. Thes e spaces di ffer radically in ambi-
ence, amenities, l ocation, and clientele—and as important, their proprie-
tors and the coworkers who work there differ radically in how they
describe coworking in their talk and in a great variety of texts, including
business documents, collateral, advertisements, Web sites, and social
media.
Services such as coworking are edge cases of more general trends toward
distributed, interorganizational, collaborative knowledge work: work that
includes independent contracting, freelancing, virtual teams, and peer pro-
duction. Consequently, understanding coworking—in particular, how peo-
ple define it, who decides to engage in it, and why they do it—can help us to
develop theoretical and analytical tools for understanding other cases of dis-
tributed work.
In this article, I report on a 2-year study of coworking at nine coworking
spaces in Austin. I examine interviews with coworkers and
coworking-space proprietors and written materials (business plans, Web
sites, collateral, site reviews, social media) that describe those sites. Using
a fourth-generation activity theory (4GAT) framework and an approach
based on grounded theory, I examine how coworkers and coworking pro-
prietors define coworking, who coworks, and why they cowork. I specifi-
cally focus on how these professionals collaboratively construct
coworking through their talk and texts. Finally, I discuss the implications
for applying 4GAT to such emergent collaborative activities. This
approach, I argue, can help us to better account for other cases of distribu-
ted, interorganizational, collaborative knowledge work.
First, I review recent work changes that have made coworking a viable
option. Then I discuss using activity theory to analyze coworking, review
the methods I used for this study, describe my findings, analyze coworking
as a coherent 4GAT phenomenon, and discuss implications for workplace
writing and communication.
400 Journal of Business and Technical Communication 26(4)

Background
Toffler (1980) predicted that personal computing would lead to the
‘‘electronic cottage,’’ in which workers could do work at home. ‘‘Put
the computer in people’s homes, and they no longer need to huddle,’’ he
argued; ‘‘white-collar work . . . will not require 100 percent of the work
force to be concentrated in the workshop’’ (p. 199). Rather, they could cre-
ate, analyze, and transform texts in the comfort of their own homes, for that
is the sort of work Toffler envisioned happening in the electronic cottage—
knowledge work. Toffler’s prediction of the electronic cottage has been
repeatedly cited in the telework–telecommuting literature (e.g., Clark,
2000; De Jong & Mante-Meijer, 2008; Ellison, 2004; Ramsower, 1985),
particularly with regard to the obvious drawback: Working from home is
potentially quite isolating and erodes the boundaries between home and
work life (e.g., Gurstein, 2001; Kjaerulff, 2010; Kylin & Karlsson, 2008).
Yet, long-term employment trends (e.g., Burton-Jones, 2001; Castells,
2003; Malone, 2004; Zuboff & Maxmin, 2004) and developments in mobile
technology have tended to encourage more work from remote locations,
more cooperative work that is not collocated, and more federated work that
is contingent rather than permanent. Examples include independent con-
tracting and other forms of contingent labor (Burton-Jones, 2001), nomadic
work (Mark & Su, 2010; Su & Mark, 2008), distance work and telework
(Bradner & Mark, 2002; Paretti, McNair, & Holloway-Attaway, 2007), peer
production (Benkler, 2006; Mueller, 2010), and other forms of distributed
work (Spinuzzi, 2007). One recent industry report estimates that ‘‘the mod-
ern contingent labor umbrella encompasses over 22% of the average orga-
nization’s total workforce’’ (Dwyer, 2011, p. 2).
To get a sense of these employment trends, consider the growth of
nonemployer firms (firms that have no employees, earn receipts over
$1,000, and are subject to federal income taxes). The number of these firms
overall has increased 21% in the United States from 2002 to 2008—and
41% in the Austin–Round Rock Metropolitan Area during the same period.
And such firms have particularly increased in the information sector, which
includes ‘‘(a) producing and distributing information and cultural products,
(b) providing the means to transmit or distribute these products as well as
data or communications, and (c) processing data’’ (U.S. Census Bureau,
2011a). These firms have grown remarkably in both number (64% in the
Austin–Round Rock Metropolitan Area vs. 32% in the United States) and
receipts (105% in the Austin–Round Rock Metropolitan Area vs. 46% in the
United States; U.S. Census Bureau, 2011b). These changes all far outpace
Spinuzzi 401

the population growth in the Austin–Round Rock Metropolitan Area (22.2%)
and the United States (5.8%) during the same period (U.S. Census Bureau,
2011c). Apparently more people are working alone, especially in Austin.
But working alone can take a toll on people, who sometimes find them-
selves cut off from networking and trust-building opportunities, with lim-
ited access to infrastructure and without firm barriers between their
personal and work lives. For instance, Kjaerulff (2010) described how tele-
workers struggled with separating their work lives and home lives and
sought other teleworkers with whom to socialize during weekly lunches.
Similarly, Clark (2000) described how rural teleworkers struggled with pro-
fessional isolation and sought local networks of freelancers (p. 173).
Compounding the problem is cities’ increasing immobility, making com-
mute times longer and causing two industry analysts to predict that ‘‘the city
will become more permeable, punctuated by a series of places to work’’
(Dixon & Ross, 2011, p. 6). The Austin–Round Rock Metropolitan Area,
which was ranked fourth in the United States in terms of the lengthiest travel
time for commuters (INRIX, 2009, p. 12), seems to fit this profile well.
To sum up, on one hand, more people (nationally, but especially in Austin)
can work anywhere—telecommuting, collaborating electronically, running
their own businesses with mobile phones and laptops. On the other hand, their
freedom to work anywhere often means isolation, inability to build trust and
relationships with others, and sharply restricted opportunities for collaboration
and networking. One emerging solution to these drawbacks is coworking.
Coworking
In the United States, Brad Neuberg is generally credited as starting the co-
working movement in 2005 when he organized Spiral Muse in San Fran-
cisco (Botsman & Rogers, 2011; Hunt, 2009). By 2011, over 700 coworking
sites had opened globally (Deskmag, 2011), including the 13 in Austin. A
Google group and a wiki keep space proprietors in contact as does an annual
event held during the South by Southwest Interactive Conference.
But what is coworking? The coworking wiki defines it this way:
The idea is simple: that independent professionals and those with workplace
flexibility work better together than they do alone . . . [C]oworking spaces are
built around the idea of community-building and sustainability. Coworking
spaces agree to uphold the values set forth by those who developed the con-
cept in the first place: collaboration, community, sustainability, openness,
and accessibility. (Coworking, n.d.)
402 Journal of Business and Technical Communication 26(4)

This definition is useful but imprecise. What kind of service is coworking?
Who, beyond ‘‘independent professionals,’’ coworks? Why do they choose
to cowork?
To answer these questions, I turn to activity theory to provide a theoretical
and methodological framework that supports my case study of nine cowork-
ing spaces i n the Austin area.
Analyzing Coworking as Activity
To understand coworking, I draw on activity theory, a sociocultural
approach to understanding cyclical, collective human activity. Specifically,
I start with the third-generation activity theory (3GAT), what was formu-
lated largely by Yrjo¨ Engestro¨m (see Spinuzzi, 2011, for an extensive dis-
cussion). Rather than recapitulating earlier detailed discussions of 3GAT
(e.g., Geisler, 2001; Russell, 1997), I focus on the activity system, with its
actors, object, and outcome, and the contradictions that emerge in activity
systems.An activity system is a collective in which one or more human
actors labor to cyclically transform an object (a raw material or problem)
in order to repeatedly achieve an outcome (a desired result). For instance,
construction is an activity system in which actors (construction workers,
a foreman, an architect, etc.) labor to transform an object (raw material that
will become a building) in order to achieve their desired outcome (a build-
ing that provides a lucrative return). In 3GAT, activity systems form net-
works in which different activities interoperate. For instance, the actors’
tools come from different manufacturers; the actors themselves come from
different disciplines; the building is constructed according to specifications
that best accommodate the activity for which it is intended.
Contradictions are systemic disruptions that form within activities,
sometimes within parts of the activity, sometimes across parts of the activ-
ity, and sometimes across activities in the network. For instance, different
actors working on the same construction project might seek contradictory
outcomes: The foreman wants to finish the building as quickly as possible
whereas the investor wants it to return the best investment, and the architect
wants to be able to take pride in the building even if that means going over
time and budget constraints.
In construction, the object is quite concrete (literally). But in collabora-
tive knowledge work, the object is harder to define. Indeed, Engestro¨m
(2009) recently suggested that we need a fourth generation of activity the-
ory (4GAT) to address such work (p. 310):
Spinuzzi 403

Third-generation activity theory still treats activity systems as reasonably
well-bounded, although interlocking and networked, structured units. What
goes on betwe en activity systems is p rocesse s, such as the flow of rules
from management to workers. [But] in social production and peer produc-
tion, the boundaries and structures of a ctivity systems seem to fade a way.
Processes become simultaneous, multidirectional, and often reciprocal.
The density and c risscrossing of processes makes the distinction between
processes and structure somewhat obsolete. The movements of informa-
tion create textures that are constantly changing but not arbitrary or
momentary. (p. 309)
Like Engestro¨m (2009), others have seen challenges to 3GAT in how
knowledge work is organized (Bodker, 2009; Lompscher, 2006; Ruckriem,
2009). For instance, Yamazumi (2009, p. 212) argued that the knowledge
society has shifted from mass production to interorganizational collabora-
tion (cf. Castells, 1996, 2003; Toffler, 1970). This shift results in ‘‘new
types of agency [that] are collaborations and engagements with a shared
object in and for relationships of interaction between multiple activity sys-
tems’’ (Yamazumi, 2009, p. 213). As Engestro¨m put it, ‘‘social production
requires and generates bounded hubs of concentrated coordination efforts’’
(p. 310), hubs in which interorganizational collaboration is the object, or at
least a large aspect of it (cf. Adler & Heckscher, 2007; Gygi & Zachry,
2010). That is, 4GAT understands internetworked activities by examining
the interorganizational collaborations to which they contribute. These chal-
lenges correspond closely with the long-term employment trends and
changes in work organization that I have discussed.
In sum, 4GAT responds to the same trends that have led to coworking.
Thus, I follow the 4GAT line of analysis here, examining the phenomenon
of coworking as an interorganizational, collaborative object. What are the
aspects of that collaboration? What activities does it network? The 4GAT
analysis proffered here, I believe, fits the phenomenon of coworking well:
Although coworking initially seems to be an unproblematic service, it
means rather different things to different participants, and a 4GAT analysis
can tease out these differences and suggest further lines of inquiry.
Methods
This qualitative case study was approved by the institutional review board
at my university. Following Smagorinsky’s (2008) suggestions for
404 Journal of Business and Technical Communication 26(4)

developing the methods section, I describe my research questions, the sites
and participants, and my data collection, reduction, and analysis
procedures.
Research Questions
I sought to answer the following research questions based on three parts of
the activity system, the object, actors, and outcome:
Research Question 1: What is coworking? That is, how do these space
proprietors and the coworkers in their spaces define coworking in their
interviews and texts, and to what extent do they agree? What meta-
phors and analogies do they use to describe coworking? What is the
object of their activity, and what contradictions exist in their under-
standing of that object?
Research Question 2: Who coworks? That is, what potential coworkers
do proprietors target, and who actually decides to cowork in their
spaces? Who are the actors of the activity, and what contradictions
exist across them?
Research Question 3: Why do people cowork? What motivations do these
space proprietors and the coworkers in their spaces report in their
interviews and texts? What outcomes do the actors desire, and what
contradictions exist across those desired outcomes?
Sites and Participants
I interviewed proprietors at nine Austin-area coworking sites and toured
their facilities. I interviewed one group of proprietors the month before the
facility (Cospace) opened. I interviewed another proprietor twice: once
when she announced that she planned to open a coworking site (Link) and
again after the site opened. In the other cases, I interviewed as soon as I
found out about each site and could persuade the proprietors to be inter-
viewed. Site names and proprietor names are not pseudonyms. I also inter-
viewed 17 coworkers at the three most populated coworking sites (see
Table 1).
Data Collection
I collected data for the study from July 2008 to February 2011. Given the num-
ber of sites and the difficulty of setting up interviews with people who have
busy and fluid schedules, I collected data snapshots rather than longitudinal
Spinuzzi 405

data: The data represent points early in the life of the coworking spaces, not
necessarily the current state of these spaces. Given the research questions, I
focused not on how people acted out coworking daily but rather on how they
described the object of coworking, their characteristics as workers, and their
motivations for coworking. Thus, my data collection was built from these
self-descriptions rather than observations.It included both formal and informal
interviews with proprietors, photos from space tours, texts, coworker
interviews, and LinkedIn profiles.
Proprietor formal interviews. I conducted formal interviews with proprie-
tors of the coworking spaces (see Table 1). Interviews ranged from 29 to
77 minutes, averaging 48 minutes. When a space had multiple proprietors,
I interviewed them together.
Proprietor informal interviews and photos from space tours. In addition, I
conducted informal interviews with the proprietors, which I then posted
on my blog after soliciting their feedback. I also toured the spaces and took
photographs to record details such as layout, furniture, and amenities.
Finally, I informally observed the most populated spaces (Conjunctured,
Cospace, Link) by working at least 6 hours in each space.
Texts. In addition, I collected 84 texts, including collateral media, mem-
bership agreements, member lists, business plans, Craigslist ads, and Web-
site pages for each site (see Appendix A). I also collected any social media
Table 1. Austin Coworking Spaces, Proprietors Interviewed, and Coworkers
Interviewed
Site Proprietors interviewed Coworkers interviewed
Brainstorm Martin Barrera
Conjunctured Jon Erik Metcalfe, Dusty Reagan,
Cesar Torres, David Walker
CW1–CW5 (5)
Cospace Andrew Bushnell, Kirtus Dixon,
Pat Ramsey
CW6–CW10 (5)
Cowork Austin Blake Freeburg
GoLab Austin Steve Golab
Link Liz Elam CW11–CW17 (7)
Perch Lisa McTiernan
Soma Vida Sonya Davis, Laura Shook
Space12 Sam Lee, Paul Wang
406 Journal of Business and Technical Communication 26(4)

related to each site: Yelp and Google Places reviews, Foursquare and
Gowalla check-ins, and Facebook pages.
Coworker interviews. I interviewed 17 members of the three most popu-
lated coworking spaces: Conjunctured, Cospace, and Link (see Table 1).
I selected these members in a convenience sample—by approaching co-
workers who were working in the space during my visits. Later, I compared
the convenience sample to these sites’ membership directories, verifying
that it was roughly representative of each site’s coworkers in terms of indus-
try and gender. Interviews ranged from 9 to 45 minutes, averaging 21
minutes.
LinkedIn profiles. I collected all the available profiles on Lin kedI n.co m:
all 17 of the coworkers I interviewed and 13 of the 16 proprietors. The
profiles describe d these partici pa nts ’ education, job history, industry,
and job title.
Data Storage, Coding, and Reduction
After collecting the data, I transcribed all the interviews, resulting in 2,608
entries (paragraph-separated units) for proprietors and 1,893 entries
(paragraph-separated units) for coworkers. I coded entire entries rather than
text segments. I placed all data in a relational database, with tables for par-
ticipants, proprietor interviews, coworker interviews, texts, and photos, and
created summary characterizations for entries in each data type (Miles &
Huberman, 1994, pp. 54–55). My coding system was nonexclusive (i.e.,
each datum could be assigned multiple codes) and included starter coding,
open coding, and axial coding.
Starter coding. I began coding deductively, using descriptive starter codes
(Miles & Huberman, 1994, pp. 57–58) based on my semistructured inter-
view questions.
Open coding. Using open coding (Corbin & Strauss, 2008), I inductively
identified recurrent themes, defined codes based on them, then checked
these codes deductively based on these definitions. First I developed open
codes based on specific issues discussed during interviews, then I applied
them to related data in the other data types. I initially autocoded entries,
applying codes based on keywords in the interview text, then I added codes
to applicable entries that did not share the keywords. I interspersed
Spinuzzi 407

autocoding with developing open codes, which I treated as emergent and
recursive (Corbin & Strauss, 2008): Information in one data set might yield
hypotheses that I could then test by coding other data sets.
Axial coding. Finally, I performed axial coding (Corbin & Strauss, 2008)
to draw connections across starter and open codes. To develop axial codes, I
looked for codes that appeared together frequently and then used a single
code to articulate the relationship between them, developing a specific
description for that code. I then recoded all data for those axial codes, apply-
ing the respective axial code to each piece of data that fit its description.
Data reduction. Coding also allowed me to reduce the data by focusing on
heavily coded data related to key themes. Appendix B lists a selection of
these codes and descriptions.
Data Analysis
I analyzed the data through comparisons and member checks.
Comparisons. I tested relationships between codes by examining whether
they were supported by multiple data sources—across data types within the
same site and within data types across sites. I compared how the same phe-
nomenon was treated in proprietor interviews, coworker interviews, texts,
and photos at a given site. For instance, I examined whether a definition
of coworking was consistent and supported across data types at a given site.
And I compared how the same phenomenon was treated within a given data
type across sites. For instance, I examined whether a definition of cowork-
ing was shared by all proprietors across sites.
Member checks. I enacted three levels of feedback for my data. First, after
informal interviews and site walk-throughs, I wrote blog profiles of each
site (http://spinuzzi.blogspot.com/search/label/coworking). Site proprietors
reviewed and gave feedback on these profiles before posting them.
Although this method resulted in positive, somewhat promotional profiles,
it also allowed me to check the profiles’ accuracy and build trust with pro-
prietors. Second, after I transcribed the interviews, I conducted a transcrip-
tion check with each participant to gather comments and feedback. And
third, after writing a draft of this article, I gathered manuscript comments
from site proprietors, which I used in later drafts.
408 Journal of Business and Technical Communication 26(4)

Findings
As we examine how participants described the three aspects of cowork-
ing—the object (what), actors (who), and outcome (why)—we find contra-
dictions in each. In fact, if we just look at the activity system of coworking,
we might even wonder if coworking describes a coherent phenomenon at
all. The proprietors and coworkers seem to disagree at every point. I discuss
the contradictions that I found in each aspect; then in the analysis that fol-
lows, I discuss how a 4GAT approach can make sense of these contradic-
tions and lead to a more coherent understanding of coworking as a
collaborative activity.
What Is Coworking? The Object and Its Contradictions
Although proprietors and coworkers alike described themselves as being in
Austin’s coworking community, their definitions of coworking differed sig-
nificantly. Both groups offered a variety of definitions, most of which were
more specific than the definition provided by the coworking wiki. In the fol-
lowing subsections, I examine the characteristics of these definitions.
Proprietors’ Definitions. The proprietors seemed to define coworking quite
differently from each other in the site interviews and in their texts. Based
on their various definitions, we might characterize these coworking spaces
as community work spaces, unoffices, or federated spaces.
The community work space: Soma Vida and Space12. The proprietors of
Soma Vida and Space12 considered their sites to be community work spaces
and defined coworking in terms of serving their local communities.
Soma Vida was a mixed-use center located in a recently gentrified neigh-
borhood in East Austin. Its roomy interior was sectioned into various
spaces, including spaces for child care, massage therapy, acupuncture,
meetings, yoga, and coworking. Its Facebook page (T62) states that Soma
Vida is ‘‘a place for people to find peace of mind, balance and community.’’
In the coworking space, individual desks sit in front of long padded
benches. According to one of the proprietors, Sonya Davis, ‘‘cowork’’ is
short for ‘‘community work space,’’ and she worried that people might think
that ‘‘you have to actually sit down and collaborate with someone, like
there’s coprojects. But what we do i s we work within a community, and
that’s what we’re all doing as anchor tenants.’’ Soma Vida’s Web site
and collateral medi a emphasi ze t hat ‘‘Our work space allows you to
Spinuzzi 409

have dedicate d t ime to concentrate and accomplish tasks’’ (T30). In
fact, Soma Vida’s policies handout states that the coworking space is
not for conversations: Intera ctions b etween coworkers should be con-
fined to other areas, ‘‘which include our lounge area, kitchen, and out-
side garden areas’’ (T15) as we ll as a confer ence room. Rather t han
being a central focus, co-working was one of many services that Soma
Vida offers.
Space12 was also a mixed-use building. It was once a notorious East
Austin nightclub before a local church took over the site. The church now
served the community in which the building is located. ‘‘It’s kind of the oppo-
site of a typical community center,’’ codirector Sam Lee explained. ‘‘We . . .
create a space so that people could use it to do their own community initia-
tives. . . . Instead of offering services, we’re offering space.’’ Space12 had
a large open plan with a recreation area, shelves of books for a prison
ministry, a stage area f or concerts and church services, a sma ll
computer room for disadvantaged students, and a coworking space (see
Figure 1). Its Facebook page (T65) reflected the wide variety of events
held there, including concerts, church services, neighborhood barbecues
and yar d sales, and a swap-and-sew e vent, events that were reflected in
people’s Gowalla check -ins (T79). Spa ce1 2 did not have a dedicated
conference room. Like Soma Vida, Space12 had ‘‘quiet’’ rules although
they were restricted to ‘‘quiet’’ hours rather than a blanket policy: ‘‘You
have a set quiet time between nine and three, probably until the kids get
out [of school],’’ explained codirector Paul Wang. Unlike the other
coworkingspacesinthisstudy,Space12 did not charge coworkers, and
its Web site’s definition of coworking was minimal: ‘‘Co-Working
Space at Space12 is a shared office space for people from all walks
of life in our community’’ (T33; s ee al so T65).
The community work spaces defined themselves in terms of serving
local communities; the object was to work alongside, but not with, others.
Consequently, both had quiet policies in their spaces, a characteristic
unique to this category of coworking space. Both also had mixed uses, in
keeping with their larger community-oriented missions.
The Unoffice: Brainstorm, Cowork Austin, Link, and Perch. In contrast, the
proprietors of some coworking spaces saw their sites as providing office
space for those who do not work in an office but miss the interactions and
amenities of the office environment. In contrast with the community work
space, the unoffice encouraged discussions; interaction between the
coworkers was an essential feature of this coworking space. One
410 Journal of Business and Technical Communication 26(4)

proprietor, Liz Elam at Link, emphatically declared that if a space had a no-
talking policy, ‘‘then it’s not coworking.’’
Brainstorm, which was based in a Victorian house in East Austin next to the
freeway, defined coworking on its Web site as ‘‘a style of work in which inde-
pendent professionals share a working environment yet perform independent
businessactivities....Collaborationiscommonasaresultofthesocialinter-
action that naturally occurs when talented and creative people share the same
physical space’’ (T34). Brainstorm occupied the second floor of the house,
upstairs from an architecture firm; it had three rooms, including a conference
room, and a kitchen. Brainstorm sometimes rented out rooms for company
retreats and other outsider meetings, and it hosted an Imagine Austin meeting.
Cowork Austin, on the third floor of a historical building in downtown
Austin, defined coworking similarly. According to proprietor Blake Free-
burg, coworking is ‘‘kind of a low-cost business platform with shared
knowledge that amplifies your business opportunities. . . . And to a lot of
the people out there, what is coworking to them? Oh, it’s a cheap office.
. . . But you get the bonus of community and shared knowledge and maybe
avenues.’’ Like Brainstorm, Cowork Austin saw interaction as natural but
optional. Cowork Austin boasted three open-plan spaces, a kitchen, a
Figure 1. Space12, a coworking space based on the community work space model.
Spinuzzi 411

conference room, and two private offices that could be leased. Freeberg
confirmed that coworkers met clients in the conference room and also
emphasized that Cowork Austin hosted various after-hours meetings for
interest groups and organizations (e.g., a Women in Tech meeting, a Cas-
sandra hackathon, an interface-design group, a tequila-tasting event).
At Link, a modernist space in North Austin that was renovated specifi-
cally for coworking, Liz Elam defined coworking as ‘‘a membership club
that brings people together who share the need for a place to conduct their
business in an interactive space’’ (T36; see also T64). She likened it to a
gym membership. In her business plan, she emphasized that ‘‘coworking
not only provides a more desirable physical space [than a home office or
coffee shop does] but promotes collaboration, networking and incubator-
like sharing of ideas’’ (T7). But as she specified, ‘‘I am not here to form
lah-di-dah, let’s-all-sing-Kumbaya community. I want people paying me,
and I’ll provide a great space for you.’’ In particular, Elam emphasized,
Link was suitable for client meetings. It had a large open-plan room (see
Figure 2), a kitchen, a conference room, and five small meeting rooms large
enough to accommodate four people each. (Elam said she got the idea for
the meeting rooms by observing business meetings at Starbucks.) Link also
hosted networking lunches and after-hours events (e.g., Blogathon ATX,
Rock a Charity, a book club, a summer concert series).
At Perch, a modernist space ‘‘on the ground floor corner of a mixed-
use develo pment ’’ in a rec en tly gentrified area of East Austin , L isa
McTiernan defined coworking a s ‘‘a nontraditional or pa rt-time flex
office space where [inde pen dent profession als] c an t rul y ge t so me ser i-
ous work done, ne twor k with oth er s’’ in an envi ronme nt th at is ‘‘afford-
able,’’ ‘‘low-commitment,’’ ‘‘flexible, and easy.’’ Similarly, Perch’s
Facebook page specified that ‘‘our space is designed to provide a wide
variety of users with a creative, functional and affordable workplace
community’’ (T68). Like Link, Perch emphasiz ed that it provides a
space for meeting clients in a professional environment and for net-
working (T32). Perch offered an open-plan room for coworking and a
conference room, which coworkers used for client meetings. Perch also
hosted art shows and a regular yoga class.
The unoffice spaces defined themselves, then, as flexible office spaces
that allow workers to interact and to meet with clients; their object was
to recreate characteristics of the traditional office environment that inde-
pendent workers may miss. In particular, they emphasized that coworkers
can exchange ideas and get feedback from other coworkers, and they tended
412 Journal of Business and Technical Communication 26(4)

to emphasize how their meeting places are superior to the default meeting
place for independents: the coffee shop.
Two of these unoffice spaces, Cowork Austin and Link, had explicitly
considered and rejected the federated work space model.
The federated work space: Conjunctured, Cospace, GoLab Austin. Finally,
some proprietors saw the mission of their coworking spaces as fostering
more active connections between coworkers, connections that could lead
to working relationships between businesses—contracts or referrals—that
is, federations (Zuboff & Maxmin, 2004). Their focus was on entrepreneur-
ship. Like unoffice spaces, federated work spaces strongly encouraged
interaction, but they also encouraged formal collaboration.
Conjunctured, Austin’s oldest coworking space, was located in a large
refurbished house in a gentrified area of East Austin (within walking distance
of Perch and Soma Vida). Members who reviewed Conjunctured on Yelp
tended to emphasize its proximity to popular restaurants and bars (T61). In
an interview, its proprietors defined coworking as a ‘‘culture’’ and delineated
between the culture of coworking (people working together, collaborating, in
the most general terms) and a coworking space (a cafe´-like environment/
Figure 2. Link Coworking, a coworking space based on the unoffice model.
Spinuzzi 413

executive suite). ‘‘Personally, I don’t think it’s about space at all,’’ offered
coproprietor Dusty Reagan. Similarly, Conjunctured’s Web site defined cow-
orking as ‘‘a global trend where freelancers, entrepreneurs, and other mobile
workers come together to work in the same space. These mobile workers want
to remain independent, but have hit a wall working in isolation at home. Cow-
orking allows them to be part of a community of like-minded individuals with
whom they can share ideas, trade business leads, foster business partnerships,
and create friendships’’ (T37). That last point was emphasized on Conjunc-
tured’s Facebook page, where its entire self-description consists of ‘‘Conjunc-
tured loves you. Yup’’ (T63)—a sentiment that reflects Conjunctured’s
relaxed, playful atmosphere (see Figure 3).
Conjunctured’s proprietors emphasized that they saw Conjunctured as a
way to quickly form federations of contractors in order to take on projects
that were too large for individual contractors. As coproprietor John Erik
Metcalfe argued, ‘‘I’ve always thought of [taking on a project] as the white
blood cells, right? So, everybody’s kind of going along in their pipe. A proj-
ect gets dropped in, we can swarm to kill it, disseminate, and keep flow-
ing.’’ Conjunctured could offer a brand that is larger than that of an
Figure 3. In keeping with its ethos, Conjunctured offers vending machines that
serve both candy and dog treats.
414 Journal of Business and Technical Communication 26(4)

individual contractor as well as a large stable of entrepreneurs who had got-
ten to know and trust each other. It could also offer referrals: If one cowork-
er could not take on a job, that coworker could refer it to someone else in the
space.
Conjunctured had a conference room, a kitchen, and three open-plan
rooms. It also hosted frequent events and meetings of special-interest
groups (e.g., CocoaCoder meetings, Bootstrap Interactive, WordPress
Meetup, and a poker game). These events were popular with coworkers,
who described them frequently in their Gowalla check-ins (T75).
Cospace was on the second floor of an office park in North Austin. Like
Conjunctured, Cospace focused on entrepreneurs. And like Conjunctured’s
proprietors, Cospace’s proprietors drew a distinction between coworking
and a coworking space. ‘‘Coworking is an informal gathering of people who
need to accomplish a task or a project or have some work to get done, but
they want to work alongside others,’’ coproprietor Kirtus Dixon offered,
whereas a coworking space is ‘‘the more formal collection of all those peo-
ple in one designated location, with the amenities . . . [including] access to
resources, space, desk, WiFi, coffee.’’ Coproprietor Pat Ramsey empha-
sized that independent workers need a ‘‘home base’’ where they can inter-
act, but not ‘‘a full-time office where you sort of stovepipe yourself with a
bunch of people, and you see the same people every day.’’ Cospace’s Face-
book page emphasized this aspect, describing the space as a ‘‘coworking,
networking, and meeting space in North Austin’’ (T66).
Like Conjunctured, Cospace emphasized the federated aspect of the
space. As coproprietor Andrew Bushnell put it, ‘‘it gives [the coworkers]
a chance to form together into a group. . . . a group of 50 people is much
more powerful than 50 individuals.’’ Cospace had a large open-plan
room, two conference rooms, a kitchen, and two offices that could be
leased to groups. It also hosted frequent social events and meetings of
interest groups (e.g., GeekAustin, Austin Drupal Users Meetup, Word-
Press Camp, Startup Weekend) as described in their coworkers’ Gowalla
check-ins (T76).
Austin’s newest coworking space, GoLab Austin, occupied excess space
in the offices of FG Squared, an interactive marketing company. Unlike the
other spaces, GoLab Austin focused specifically on the interactive media
industry. Steve Golab, who owned FG Squared and was proprietor of
GoLab Austin, defined coworking as ‘‘basically giving people the tools that
they need to be effective in their work. Making sure that they are produc-
tive....Ithinkit’s about helping facilitate relationships between one
another and with people who are inside of the network.’’ GoLab Austin’s
Spinuzzi 415

Facebook page makes its federated model quite explicit, describing it as
‘‘more than just a shared office space’’ and promising facilitated interac-
tions and referrals (T53, T84). At the time of the interview, GoLab Austin
had not yet hosted events though Golab planned to do so.
The federated work spaces defined themselves in terms of fostering busi-
ness relationships in addition to personal ones; their object was to facilitate
collaboration with others in formal and informal relationships. That is, the
proprietors saw these spaces as comprising a collocated network of poten-
tial contractors. Although they emphasized this collaborative focus, the pro-
prietors also saw their spaces as providing the benefits of interaction that the
unoffice model provides.
But what did the coworkers think?
Coworkers’ Definitions. To get the coworkers’ perspective, I interviewed 17
coworkers at the three most populated coworking spaces—Conjunctured,
Cospace, and Link. (I also examined user-generated texts such as Yelp and
Google Places reviews and Foursquare and Gowalla check-ins, but users did
not define coworking in these texts.) Coworkers at these three spaces presented
interesting differences in their definitions (see Table 2). They believe that the
definitions they proffered capture the essence of coworking; in a later section, I
discuss their motivations, the reasons why they cowork.
Coworking as space. Of these three most populated spaces, only Link was
defined as an unoffice in the proprietor interviews and texts. But coworkers
across all three spaces defined coworking as an alternative office space. For
instance, at Conjunctured, CW01 defined coworking as ‘‘a very cheap and
easy way for me to work in an office space around people that are, I guess,
very similar to me’’ whereas CW04 defined it as ‘‘trying to recreate some
parts of the office environment . . . the parts of the office hopefully that
work better than working from home.’’ Similarly, at Cospace, CW06 said
‘‘just like in any office, you know when Monday morning comes, what
you’re getting into,’’ and CW10 called it ‘‘an office opportunity . . . It’s a
financial advantage to be able to have an office.’’ Overall, 7 of the 17 cow-
orkers explicitly defined coworking as analogous to the office whereas 15
of the 17 defined it as an office, space, or workplace.
Coworking as an inexpensive office alternative. Although coworking is dra-
matically less expensive than leasing an office space, only three coworkers
(CW1, CW2, and CW10) mentioned that fact.
416 Journal of Business and Technical Communication 26(4)

Coworking as a social hub. ‘‘I now have a water cooler,’’ stated CW15. ‘‘I
have a place where I can bounce ideas off of people.’’ Like CW15, many
coworkers across the sites emphasized interaction with other coworkers.
For instance, CW09 simply defined coworking as ‘‘social working . . .. It’s
a social environment to do your work.’’ Similarly, CW03 defined it as
‘‘combining social networking and working in a laid-back environment
where the stress is gone.’’ Overall, 6 of the 17 coworkers used some varia-
tion of the phrase ‘‘bounce ideas off of people’’ (CW15), and 9 of the 17
coworkers defined coworking in terms of a social component.
Coworking as collaboration. Surprisingly, most coworkers did not define
coworking as an opportunity to collaborate on federated projects (although
more discuss collaboration as a motivation, which I discuss later). For
instance, CW08, who worked in Cospace’s federated space, defined it as
people ‘‘working on different projects in the same shared space’’—hardly
the federated vision that Cospace’s proprietors expressed in their interviews
and texts. Overall, only 5 of the 17 coworkers defined coworking in terms of
collaborating on projects, 4 of whom worked at the unoffice rather than at
one of the federated work spaces.
Coworking as heterogeneous and homogeneous. Proprietors and coworkers
alike sometimes referred to homogeneous and heterogeneous populations.
Conjunctured (T37), Perch (T32), and GoLab Austin (T53) texts contained
the phrase ‘‘like-minded,’’ and proprietors of Perch and Soma Vida used it
in their site interviews. Like the proprietors, three coworkers (one each at
Table 2. Characteristics in Coworkers’ Definitions of Coworking
Characteristics
Coworkers at
Conjunctured
Coworkers at
Cospace
Coworkers at
Link
Space 1, 2, 4, 5 6–8, 10 11–14, 16, 17
Inexpensive office
alternative
1, 2 10
Social hub 3 6, 9, 10 11, 12, 14, 15, 17
Collaboration 4 11, 12, 15, 16
Heterogeneity 6 14, 15
Homogeneity 1 6 12
Work/home
separation
217
Spinuzzi 417

Conjunctured, Cospace, and Link) defined coworking in terms of working
with people like themselves.
At the same time, three coworkers emphasized appreciating the hetero-
geneity of workers at the space. For instance, CW15 emphasized, ‘‘I can get
really different views because I have individuals across the spectrum in
their jobs and what they do. You can get a better idea of what people think
outside of the industry that you’re in.’’ (Proprietors of Cospace, Perch, and
Space12 all emphasized heterogeneity in their site interviews, and Perch’s
Facebook page mentions it, T68.)
Coworking as work/home separation. Only two coworkers (CW2 and
CW17) mentioned work/home separation as a component of coworking.
(Others mentioned this component as a motivator for them, but not as part
of the definition.)
The Object of Coworking and Its Contradictions. What is coworking, then? That
is, what is the cyclical object of coworking activity at these sites? As we
have seen, answers vary. A coworking space is a place to get work
done—specifically, knowledge or service work that originates outside the
site in other intersecting activities. Although coworkers work together, that
work involves different, contradictory objectives, attached to and pulled by
the network of activities in which each coworker engages. These intersect-
ing activities perturbed the development of the object at each coworking
site.
Coworking proprietors defined coworking in ways that emphasized spe-
cific models. Yet, coworkers had different objects in mind as well—differ-
ent from those of the proprietors and each other. For instance, the coworkers
I interviewed tended to emphasize the unoffice model, in particular, the
combination of space and social interaction as they performed separate
projects. But beyond saying that they worked in the presence of other peo-
ple, they provided definitions that were far from unanimous.
This finding is interesting because how the participants perceive the
object of coworking affects how they coconstruct it. For instance, proprie-
tors structure, design, furnish, and run their sites based on their understand-
ing and model of coworking—and indeed they have considerable incentive
to differentiate their sites from others. But the coworkers I interviewed
seemed to understand coworking primarily based on one model, the unof-
fice. Coworkers were split on other characteristics of coworking, which
meant that they came to their spaces with different expectations about the
activity in general: Would they work in parallel or in collaboration? Would
418 Journal of Business and Technical Communication 26(4)

they socialize or find partners? Would they meet like-minded individuals or
coworkers with different views and backgrounds?
As I discuss in the next section, some of these differences appear to relate
to the individual coworkers’ businesses and fields: For instance, an indepen-
dent professional who needs to meet clients has different needs and expec-
tations from an independent contractor who needs to make business
connections and get to know potential subcontractors.
I promised that the contradictions would begin to pile up, and they cer-
tainly have. So we can be forgiven for wondering if coworking is a coherent
phenomenon rather than an uns table referent. Across proprietor s, across
coworkers, even between the proprietors and their coworkers at a given
space, participants disagree in their definitions of coworking. These con-
tradictions continue to pile up as we move to the next question: Who
coworks?
Who Coworks? The Actors and Their Contradictions
The activity systems at these coworking spaces involved various actors.
Who did coworking proprietors expect to work at their sites, and who actu-
ally did?
Who Proprietors Targeted. To determine the proprietors’ target audience, I
examined their statements in the proprietor interviews as well as the texts
I collected from each site although proprietors may have changed their tar-
get audiences after these data were collected.
All proprietors, of course, identified individuals who could choose where
to work. Those ideal coworkers had found home offices and coffee shops to
be inadequate work spaces. Additionally, proprietors tended to seek and
expect coworkers who shared their own background. For instance, Conjunc-
tured’s proprietors, who were experienced freelancers and entrepreneurs,
sought freelancers who were used to working virtually. Cospace’s proprie-
tors, who had previously owned small businesses, sought small-business
owners. Link’s proprietor, a former global account manager for Dell, sought
high-end Gen-X business travelers. Soma Vida’s proprietors, who identified
strongly as female entrepreneurs with children, sought ‘‘mamapreneurs’’
and ‘‘papapreneurs.’’ Even Space12’s proprietors, who targeted the most
diverse set of coworkers, emphasized that they wanted their church staff
working there.
In addition, the proprietors identified people with specific characteris-
tics. For instance, Conjunctured’s proprietors specifically sought people
Spinuzzi 419

seeking leads and business partnerships—in keeping with its orientation as
a federated work space—but also friendships. Cospace’s proprietors specif-
ically sought diversified small-business owners in North Austin needing a
‘‘home base’’ who could supply referrals to each other. Link’s proprietor
sought high-end independents who valued interaction with other coworkers
but were more concerned with minimizing unwanted distractions and meet-
ing clients.
These proprietors, then, envisioned different actors for their sites. But
they certainly did not rule out others as coworkers, they just did not iden-
tifythemastargets.Theyweregladtotakecoworkerswithdifferentpro-
files from what they envisioned as long as those coworkers were not
disruptive .
Who Actually Coworked. Who actually chose to work at these coworkers
spaces? The actual coworkers often did not match the targeted coworkers
whom proprietors described. The coworkers I interviewed at Conjunctured,
Cospace, and Link worked in various industries and capacities. But most of
the coworkers at the nine sites were independent workers. Based on their
interviews and Linked-In profiles, I found that of the 17 coworkers I
interviewed,

10 were small-business owners other than consultants; 6 of these were in
one-person organizations

4 were consultants; 3 of these were in one-person consultancies

1 was a dependent contractor working remotely for a large business

1 was an intern for a business in the coworking space

1 was a permanent employee of a business in the coworking space

12 had an Internet or information technology component to their busi-
ness. For instance, CW02 worked in the apparel and fashion industry for
a company that sold apparel exclusively online.
At Conjunctured and Cospace, all coworkers I interviewed had an Internet
or information technology component to their business. At Link, the co-
workers I interviewed were diverse independent professionals, all of whom
were small-business owners or consultants. Only two of the seven had an
Internet or information technology component to their business, and none
of them fit the profile of business traveler. Based on the member directories
from the three most populated coworking spaces, this breakdown appears to
reflect the membership at the three spaces as a whole.
420 Journal of Business and Technical Communication 26(4)

The Actors of Coworking and Their Contradictions. Although they continue to
pile up, the contradictions are relatively mild here because the actors are
so loosely defined. The coworkers generally represented the clientele that
proprietors expected although with important differences (e.g., none were
frequent business travelers). But they also were involved in a wide variety
of professional activities, activities that involve different ties and relation-
ships. Coworkers cowork, but freelancers freelance, consultants consult,
entrepreneurs start and grow businesses, and small-business owners run
small businesses. These activities are different, particularly because they
are exercised in different fields and disciplines and require different sorts
of work and resources—and collaborations. Consequently, the actors of
these coworking activities expected quite different things from their shared
spaces.
Why Do People Cowork? The Outcome and Its Contradictions
What did coworkers expect to get from coworking? That is, what outcomes
did they expect to cyclically achieve by coworking? To investigate this
question, I examined coworkers’ interviews as well as coworker-
generated texts such as Yelp and Google Places reviews and Foursquare and
Gowalla check-ins.
All of the coworkers reported that they had tried working from home,
and 14 of the 17 reported working from coffee shops. As Table 3 shows,
participants were unhappy with these work spaces, reporting that they expe-
rienced distractions, self-motivation problems, and feelings of isolation. For
instance, CW06 reported that when he worked at home, he would have to
take conference calls in his parked car because ‘‘you never knew when
[my dogs] were going to start barking. And it seemed like they would sadis-
tically plan to bark when I was on a call.’’ CW15 found that when she
worked from home, she would realize at noon that she was still in her paja-
mas, and she also found herself being distracted by domestic chores such as
washing dishes and doing laundry. CW08 recounted, ‘‘I got really depressed
[at home] because I didn’t talk to anybody all day long.’’ And CW03
described problems with coffee shops that many others had raised: ‘‘When
you go to a coffee shop, you are obligated to buy something. You don’t want
to spend too much time there. It’s not really conducive [as] a work space
because people are talking. People don’t think of it as a work space.’’
Given that the coworkers I interviewed were primarily small-business
owners and consultants, such problems are critical: These professionals had
to be highly motivated and focused because their livelihoods depended
Spinuzzi 421

primarily or solely on their own initiative. Yelp reviews and a Gowalla
check-in reflected how difficult coworkers found it to work in home offices
and coffee shops (T58, T59, T61) and how these coworkers welcomed qui-
eter, less distracting environs (T56, T80).
These themes carried through to what the coworkers sought from their
coworking spaces. Coworkers reported that they sought a variety o f char-
acteristics from their spaces, and the various spaces provided key differ-
entiators according to their s pac e, design, and professionalism;
flexibility; and location and the benefits that coworkers tended to receive
from each other.
Space, Design, and Professionalism. Coworkers across the three sites specifi-
cally discussed furniture and space design, but they judged these by varying
criteria. CW04, for instance, specified that ‘‘if you’re going to be sitting
somewhere for three or four hours, the chair better be comfortable.’’ Indeed,
at Conjunctured and Cospace, coworkers included mostly entrepreneurs,
freelancers, and consultants who seldom met their clients there, so these
coworkers’ focus was primarily on functionality and comfort. That is, the
space design was inward facing, focused on facilitating comfort and rela-
tionships within the coworking site. Particularly at Conjunctured, the space
design and atmosphere did not support formal meetings. CW05 recalled one
day that she met clients there: ‘‘That was the day that like three people
showed up with their dogs. Like it was 4:00, and it was loud, and the dogs
were barking and fighting. And they were laughing, and then somebody was
like drinking beer.’’ CW04, also at Conjunctured, reported that he tended to
meet clients at coffee shops instead.
Table 3. Coworkers’ Experience Working From Other Spaces
Characteristics
Coworkers at
Conjunctured
Coworkers at
Cospace
Coworkers at
Link
Had worked from home 1–5 6–10 11–17
Home: distractions 1, 4 6, 8 11, 12, 14, 15
Home: self-motivation
problems
46,811
Home: isolation 3 7–9 12, 13, 17
Had worked at coffee
shops
3–5 6–10 11, 13–17
Coffee shops:
distractions
4614
422 Journal of Business and Technical Communication 26(4)

In contrast, at Link, coworkers included mostly small-business owners
who met with their clients face-to-face, so their focus was on image and
professionalism. Six of the seven coworkers emphasized that they were
proud to meet clients there. (The seventh coworker, an artist, did not have
clients.) For instance, CW13 commented, ‘‘This place has a modern design,
but it’s organic and comfortable, and that’s so hard to pull off. It’s also
clean, and there’s no microwave-popcorn smell, and there’s no worry about
science experiments in the fridge. These things matter and especially if you
want to bring in clients or friends or whatever.’’ That is, the site was
designed to be outward facing, to facilitate professional contacts with out-
siders, to impress rather than to comfort. CW16, who had worked at both
Conjunctured and Link, also mentioned how Link, as opposed to Conjunc-
tured, had the ‘‘wow factor’’ for meeting clients.
Coworkers also mentioned how spaces supported events. For instance, in
comparing Cospace to Conjunctured, CW06 claimed that Cospace had a
greater ability to host conferences and meetings ‘‘because Conjunctured
isn’t quite as big as this place.’’
In social media, coworkers described the spaces approvingly, focusing
on the unique characteristics of each space. For instance, Link was
described on Google Places as a ‘‘high-end space’’ (T60) and in a Gowalla
check-in as ‘‘such a beautiful space!’’ (T77). Soma Vida was described in a
Foursquare check-in as ‘‘relaxing’’ (T72). And when a coworker checked in
at Cowork Austin, she suggested that others ‘‘check out the patio and terrific
conference room’’ (T74).
Going beyond specific space elements, coworkers at Cospace and Link,
when comparing their site to Conjunctured, often centered their compari-
sons around how spaces allowed them to project professionalism. For
instance, at Cospace, CW07 claimed that Conjunctured represented an
‘‘artist-commune type’’ of coworking compared to Cospace’s ‘‘businessy’’
model; similarly, CW10 claimed that Cospace was ‘‘a little bit more busi-
ness minded’’ than Conjunctured. Coworkers at Cospace (CW07) and Link
(CW13, CW16) emphasized the age gap between themselves and the
younger coworkers at Conjunctured, using terms that tied age to profes-
sional demeanor. For instance, CW13 said that Conjunctured ‘‘felt like col-
lege to me again. . . . I just wanted to have it be a place where I can turn on
professional brain and stay focused and have the peer pressure to be Profes-
sional Lady.’’
Flexibility. On the other hand, some people sought more time flexibility, par-
ticularly entrepreneurs and freelancers, who typically set their own hours
Spinuzzi 423

and do not have to meet clients regularly. These people tended to frequent
Conjunctured, which had extended hours and gave keys to trusted cowor-
kers, more than Cospace, which charged extra for keyed access, and espe-
cially more than Link, which was open only during normal business hours.
Location. Another differentiator was location. The coworkers sought cow-
orking sites that were closer to their homes, clients, or desired amenities.
The desired amenities differed from one coworker to another. For instance,
CW12 and CW13 specifically cited Link’s proximity to day care as one of
its advantages for them. Conjunctured, on the other hand, was closer to
downtown in a rapidly gentrifying area of East Austin. Using a Yelp review
to rib people at other coworking spaces, one Conjunctured coworker wrote,
‘‘I seriously feel sorry for people that go to coworking spaces in North Aus-
tin or downtown,’’ before describing Conjunctured’s proximity to popular
bars and restaurants (T61). Different locations seemed to appeal to people
in different life stages: Link’s coworkers were mostly in their 30s and older
whereas Conjunctured’s were mostly in their 20s. Finally, Cospace was at
the corner of a highway and major artery, making it more convenient for
commuters coming from suburbs north of the city (CW07).
These differentiators, of course, sometimes conflicted with each other.
CW04 articulated one such conflict: ‘‘The question is, is [coworking] going
to be community based or proximity based? That, you know, ‘do I come to
this place because I like the people here, or this happens to be the closest
place, and the people don’t piss me off?’’’ That is an important question
because people also seek coworking sites to interact with others.
Benefits From Coworkers. Beyond these benefits that the spaces themselves
provide, coworkers sought certain benefi ts from other coworkers ( see
Table 4), such as interaction, feedback, trust, lea rning, partnerships,
encouragement, and referrals.
Interaction. Many of the coworkers across the sites expressed their desire
to interact or socialize with the other coworkers. For instance, CW07 told
me that ‘‘probably I am not going to work on jobs with these people, but
I like to socialize with them and talk to them.’’ This theme of interaction
was common, especially in the Link and Cospace interviews but also in the
Yelp and Google Places reviews; for example, one Cospace coworker
enthused that ‘‘I get to see new people every day’’ (T58), a Conjunctured
coworker described how ‘‘congenial’’ people were (T61), and a Link co-
worker emphasized how much of a host the proprietor was (T59). And
424 Journal of Business and Technical Communication 26(4)

frequent after-hours events provided plenty of other opportunities to socia-
lize and network, and those events were reflected particularly in some of the
sites’ Gowalla check-ins.
Interaction also took other forms. For instance, during his interview,
CW17 remarked that ‘‘for me, it is about just the casual relationships’’ and
that he had a friendly rivalry with CW19 on Foursquare: Both competed to
be mayor of Cospace. Later, CW19 mentioned the same rivalry—then
paused the interview so that he could check in. ‘‘He’s not here,’’ CW19 told
me, then added hopefully, ‘‘I might actually get the mayorship today.’’
Feedback. People across the sites also expressed their desire for feedback
although feedback had different meanings at different sites. For instance, at
Conjunctured and Cospace, coworkers were generally in fields related to
technology or the Internet, so they tended to seek feedback on problems
from others in their field. In contrast, coworkers at Link generally worked
in more diverse, customer-contact businesses, so they tended to seek feed-
back from coworkers in different fields. As CW15 put it, ‘‘You can get a
better idea of what people think outside of the industry that you’re in.’’
Trust. Similarly, trust looked different across sites. At Conjunctured, in
accordance with its federated model, coworkers sought collaborators they
could trust as partners. As CW01 put it, ‘‘do [subcontractors] have a handle
on their time? Can they manage all the stuff that we need to get done?’’
Conjunctured provided an environment for developing such trust. Similarly,
at Cospace, CW06 emphasized that ‘‘you like to do business with people
you trust’’ and reported that ‘‘there are already trusting relationships being
built here.’’
Table 4. Desired Benefits From Coworkers
Benefits
Coworkers at
Conjunctured
Coworkers at
Cospace
Coworkers at
Link
Interaction 3 6–9 11–15, 17
Feedback 1–3, 5 6, 10 11, 15, 17
Trust 1, 4, 5 6 12–17
Learning 1, 3, 5 9 11, 15–17
Partnerships 1, 3–5 6–10 12, 13, 15–17
Encouragement 1, 5 14, 17
Referrals 5 10 16
Spinuzzi 425

On the other hand, at Link, coworkers tended to work in parallel. These
coworkers thought of trust in terms of personal possessions and in terms of
sharing ideas with people with whom they were not partnered. For instance,
CW12 told me, ‘‘I don’t have to worry about my purse and my Mac. And at
the same point, I feel like I’ve told [other coworkers] about my business
plan. They’re not going to go and tell my competition. I think they’re going
to keep it [to themselves]. I haven’t sworn them to secrecy; I haven’t made
them sign a nondisclosure agreement.’’
Learning. Learning was a consistent theme. Again, coworkers at the fed-
erated sites (Conjunctured and Cospace) tended to emphasize learning
within their fields when tackling work problems. At Conjunctured, for
instance, CW03 said, ‘‘I feel comfortable to a point now where I know if
I have difficult questions, I can ask people around here.’’ Similarly, at
Cospace, CW09 anticipated learning from a ‘‘pool of talent.’’ At Link, on
the other hand, coworkers emphasized learning about business practices
in general rather than about field-specific tools or processes. For instance,
CW17 said she had sought guidance on outsourcing accounting and using
billing practices whereas CW15 had asked for guidance on a piece of col-
lateral she had developed for her business, and CW11 intended to seek help
from someone who was ‘‘a brainiac on spreadsheets and budgets [since] I’m
really bad about keeping my receipts.’’
Partnerships. The three sites also differed in their potential for forming
business partnerships. As a federated work space, Conjunctured had tar-
geted entrepreneurs who could work virtually and who could seek business
relationships within the space. The participants whom I interviewed gener-
ally fit this profile. At Conjunctured, all of the coworkers had an Internet com-
ponent to their business, and three coworkers reported subcontracting others in
the space. For instance, CW1 said that Conjunctured was a good space to find
subcontractors because people with a certain work ethic chose to work there.
CW1, CW4, and CW5 all cited the fact that working alongside coworkers
helped them to assess whether those coworkers would be good subcontractors.
As CW5 put it, ‘‘I’ve seen them interact, you know, I’veseen them make agree-
mentsanddeliverornotdeliver....Whatbetterinterviewcouldyouget?’’
CW5 also mentioned that at Conjunctured, subcontracting opportunities
could be spontaneous: ‘‘One day we needed an editor, and I saw [a cowor-
ker] walk in . . . and I’m like, do you have three hours to edit something?
She said yeah. So she took a left instead of going into the room, and came
in here and edited this stuff that we had to deliver.’’ By offering a
426 Journal of Business and Technical Communication 26(4)

concentration of independent professionals within related industries, Con-
junctured provided the potential for nimble business collaboration.
At Cospace, although coworkers saw the potential for establishing busi-
ness partnerships with each other, at the time of my interviews, they had
not. CW7 and CW9 expressed interest in establishing such relationships
in the future. As CW9 put it, ‘‘They’re still getting started here, so there aren’t
really a lot of members yet. But one of the cool things about coworking is that
you meet other people that you could potentially work with at some point.’’
CW10 reported that he planned to refer Web clients to a coworker.
Similarly, Link’s coworkers had not yet established business relation-
ships with each other. Four of the seven (CW12, CW13, CW16, and
CW17) expressed the hope of establishing such relationships, and CW16
said that the promise of such relationships was ‘‘a big part of my choice
to be involved in coworking originally.’’ But these coworkers largely
worked in very different customer-contact fields, so the business relation-
ships that they envisioned tended to involve one-off services (e.g., buying
a house, commissioning an interior-design session): They sought customers
rather than partners.
The Outcomes of Coworking and Their Contradictions. Why did people cowork?
That is, what outcomes did they hope to achieve through the object of cow-
orking? These coworkers sought multiple outcomes, ones that were often not
shared across coworkers. They expressed overlapping concerns, partly
because different sorts of workers needed different kinds of support. Indeed,
these concerns often contradicted each other: Specifically, coworkers had dif-
ferent expectations and desires concerningspace design, feedback, trust, learn-
ing, and partnerships. One major contradiction emerged from these different
concerns: Some coworkers expected to work in parallel whereas others
expected to work in cooperation.
Parallel work as an outcome. Coworkers who expected to work in parallel
wanted to interact with each other socially, sometimes gathering feedback
from those in different fields, building a sort of neighborly trust so that they
could leave their belongings unattended or discuss business dealings with-
out having those details repeated. These coworkers often worked in sole
proprietorships in customer-contact areas (e.g., interior design, real estate)
and needed space to meet their customers as well as other amenities that
would make them look more professional. Although this outcome did not
involve direct collaboration between coworkers, these conditions for paral-
lel work definitely represented ongoing joint achievement: For these people
Spinuzzi 427

to work alone together took considerable coordination and communication.
They had to work at being good neighbors.
Cooperative work as an outcome. Coworkers who expected to work in
cooperation wanted to gather specific feedback and learning techniques from
others in theirown field, building a working trustthat could lead to partnerships
or subcontracting. Often these coworkers were freelancers or entrepreneurs
who provided services to other businesses rather than individual customers,
and sometimes they would never meet their clients face-to-face. As a result,
these coworkers focused less on image and client meetings—explaining in part
the extremely relaxed atmosphere at Conjunctured in comparison to the other
coworking sites—than on generating cross talk and camaraderie that could
lead to trusted partnerships. They had to work at being good partners.
These are contradictory outcomes, implying two different sets of ame-
nities and services. Again, the contradictions pile up. Is coworking really
even a coherent phenomenon? In the next section, I argue that it is and that
a 4GAT analysis can help us understand it systemically.
Analysis: Coworking as a Coherent 4GAT Phenomenon. Throughout this article,
I have examined the objects, actors, and outcomes of coworking, and I have
described so many contradictions that you might suspect that coworking
does not even describe a coherent phenomenon. But coworking makes more
sense if we examine it in 4GAT terms, as ‘‘collaborations and engagements
with a shared object in and for relationships of interaction between multiple
activity systems’’ (Yamazumi, 2009, p. 213)—activity systems sharing col-
laborative objects that are ‘‘bounded hubs of concentrated coordination
efforts’’ (Engestro¨m, 2009, p. 310). Doing so lets us connect some of the
isolated contradictions we have noted in the objects, actors, and outcomes
of coworking.
For instance, one coworking configuration can be described as the Good-
Neighbors configuration (see Figure 4). In this configuration, coworkers
(whose activities are represented by triangles) who regularly meet with cus-
tomers face-to-face bring their work into the coworking space (the circle)
and work on it in parallel. Their collaboration is not focused on these indi-
vidual tasks but on sustaining their neighborly relationships so that the co-
working. space can best support everyone’s parallel work. An important
aspect of this collaborative work is to provide what Goffman (1959) called
a ‘‘team performance,’’ a front stage of professionalism that is appropriate
for meeting with customers. In these coworking spaces, this good-neighbors
configuration is reflected in several areas: the proprietors’ unoffice model,
428 Journal of Business and Technical Communication 26(4)

coworkers’ focus on sociality, coworkers’ tendency to meet with customers
face-to-face, and coworkers’ desire for a parallel work outcome in a profes-
sional space where they could meet and impress customers. These aspects
of the object, actors, and outcomes make sense when we understand the good-
neighbors configuration as a nexus of otherwise unlinked external activities.
Like neighbors, these coworkers may be entirely unconnected in their work
lives but committed to sharing and improving a communal space.
In contrast, in the good-partners configuration (see Figure 5), indepen-
dent, unaffiliated specialists (whose activities are represented by triangles)
can link up inside the coworking space (represented by the circle) to attack
shared work problems. These shared problems are the objects of their
momentary collaborations, the problems that individuals from the space are
recruited to swarm; the more enduring object, however, is the networking
that facilitates these fast-forming instances of cooperative work. One such
transient team is illustrated in Figure 5: experts in Web development,
search-engine optimization, Web services, and copywriting have temporar-
ily linked up to solve a problem posed by an external business client. (The
graphic designer and retailer have been left out of this particular collabora-
tion but could be picked up on future jobs.) The good-partners configuration
supports not a front stage of professionalism but a backstage in which cow-
orkers can attack these work problems. This good-partners configuration is
reflected in the coworking spaces in several areas: the proprietors’ federated
Object:
neighborly
collaboration
Web
development
Real
estate
Interior
design
Marketing
Coworking
Art
Outcome:
parallel
work
Clients
Clients
Clients
Clients
Clients
Figure 4. The good-neighbors configuration of coworking: An outward-facing front
stage that supports individuals’ work efforts.
Spinuzzi 429

work space model, coworkers’ focus on collaboration and sociality, cowor-
kers’ business-services orientation, and coworkers’ desire for a cooperative
work outcome. These aspects of the object, actors, and outcomes make
sense when we understand the good-partners configuration as a nexus of
transient work teams composed of specialists. As partners, these coworkers
forged connections through their work lives as well as their social lives, and
they treated the coworking space as the backstage for making these connec-
tions happen.
So we can detect at least two configurations of coworking, two distinct
ways in which a coworking space can function as a nexus of networked
activities. Obviously, other configurations are possible, and other activities
impinge. For instance, coworkers preferred certain locations because they
were closer to certain amenities, suggesting that other activities (e.g., parent-
ing) also influenced coworking. Similarly, a detailed study of a coworking
space based on the community work space model might well turn up another
configuration in which other activities are networked. Nevertheless, Table 5
compares the two configurations for which we have the most evidence.
Object: team
collaboration
Web
development
Retail
Search
engine
optimization
Web
services
Copywriting
Outcome:
cooperative work
Clien
t
Graphic
design
Figure 5. The good-partners configuration of coworking: an inward-facing back-
stage that supports ongoing networking, leading to transient teams.
430 Journal of Business and Technical Communication 26(4)

As Table 5 suggests, the tensions between these two configurations
appear to explain some of the contradictions that I discussed in previous
sections. These tensions arise because the two configurations are superim-
posed in each space (see Figure 6). One configuration may be dominant in
each space—for instance, Link is dominated by the good-neighbors config-
uration whereas Conjunctured and Cospace are dominated by the good-
partners configuration—but both configurations are manifest in each space,
and the tensions between them appear as systemic, affiliated contradictions.
Implications: Who Writes the Definition of
Coworking?
In this case study, I have examined how people collectively define and interpret
an emergent collaborative activitythroughtheir talk and their manytexts. Cow-
orking is not a concrete product, like a building, but a service—in fact, a service
that proprietors provide indirectly, by providing a space where coworkers can
network their other activities by engagingin peer-to-peer interactions. This ser-
vice is now vaguely defined, allowing different configurations; consequently,
we have seen many different contradictions in the object, actors, and outcomes
of coworking. But coworking has evolved and will likely continue to evolve; as
we examine how, we can also examine implications for how we apply activity
theory to other emerging collaborative activities.
How Will the Definition of Coworking Evolve?
As we have seen, space proprietors confidently offered their definitions of
coworking in their interviews, business plans, Web sites, and collateral media,
Table 5. Two Configurations of Coworking and Their Contradictions
Good neighbors Good partners
Object (proprietors) Unoffice Federated work space
Object (coworkers) Sociality (as neighbors)
Collaboration (as neighbors)
Sociality (as potential
partners)
Collaboration (as partners)
Actors Small-business owners and
consultants providing
customer-contact services
(front stage)
Entrepreneurs and
freelancers providing
services to businesses
(backstage)
Outcomes Parallel work Cooperative work
Spinuzzi 431

and they have made decisions (e.g., space design and location, hours, rules,
pricing, and approved events) that influenced how their coworkers understood
coworking. But we have also seen that coworkers additionally define cowork-
ing in terms over which proprietors have little control: through interactions
with coworkers and the partnerships and trust that develop from these interac-
tions. In fact, because key aspects of coworking as a service are provided by
those who buy that service, proprietors have little control over the definition
of coworking. Coworking is a low-margin, monthly service with well-
established competitors such as coffee shops and home offices in addition to
other coworking sites. Indeed, coworkers freely compare spaces in their inter-
views and in social media. Much of coworking’s value rests on who else is
coworking; indeed, many of these spaces have subsequently hired or desig-
nated community managers who can structure interactions, facilitate introduc-
tions, and otherwise introduce greater social coherence—that is, managers
who can orchestrate the networking of external activities within the site.
Coworkers tended to select spaces based on various factors: whether
other coworkers did business with other businesses or with individual cli-
ents, were Gen-X or Gen-Y, were technology oriented or not. People vote
with their feet, and as they do, they change the networked configuration, the
value proposition, and thus the object of coworking at each space.
Client
Clien
t
Client
Client
Figure 6. The two configurations superimposed.
432 Journal of Business and Technical Communication 26(4)

So what is coworking? In this analysis, coworking is a superclass that
encompasses the good-neighbors and good-partners configurations as well
as other possible configurations that similarly attempt to network activities
within a given space. What configurations should a given coworking site
support? That is, how can the site serve as an effective nexus of different
work activities? As they answer these questions, proprietors and coworkers
develop their own configurations and further define what coworking means
at their own sites. Looking at these interviews and texts, we can see these
developments take place via largely decentralized discussions: in daily
interactions at each site, in Gowalla check-ins, in Yelp reviews, at after-
hours events, and in innumerable other forums.
As cities become more porous and workers become more mobile, we can
expect coworking and variations to multiply. In this early study of an emer-
ging phenomenon, we can see how critical texts and talk have been to co-
working’s definition and development. In fact, it is striking how much of
this emerging discussion has taken place over scattered genres—such as
Web sites, business plans, collateral media, guidelines, reviews, check-
ins—distributed across all who are involved in coworking.
How Will the Activity of Coworking Be Defined?
And that brings us to the implications of this study for activity theory. When
I began studying coworking, I conceived it as people simply working in the
peripheries of each other’s activities—working alone together, as some in
the coworking community say. Thus, I wondered whether activity theory
could provide an adequate framework for studying coworking. Indeed, the
term coworking seemed to gloss several objects achieved by several differ-
ent actors to yield several different outcomes. Coworking seemed not to be
a single activity at all, and certainly the participants’ understanding of co-
working rested heavily on how it intersected with their other networked
activities. These dense interconnections with other activities continually
pulled the participants’ understanding of coworking in different directions.
That pull has already caused significant differentiations in Austin cowork-
ing, resulting in at least two very different configurations, and we can
expect further differentiations as coworking becomes more common. With
such a highly collaborative, interorganizational, and fluid phenomenon, an
activity theory analysis seems difficult to apply.
To do so, I have turned to an emergent 4GAT approach that understands
internetworked activities by examining the interorganizational collabora-
tions to which they contribute, an approach that examines peer production
Spinuzzi 433

in textured activities (Engestro¨m, 2009), objects shared by activity net-
works (Yamazumi, 2009), and multiactivity interagency (Edwards, 2009).
This turn, I believe, is necessary if activity theorists are to make sense of
work that is increasingly focused on knowledge and services in interorgani-
zational and cross-disciplinary collaborations—and coworking is just one
example of such collaborations. As I have argued here, developing and
applying a 4GAT analysis allow us to better examine how these networks
of activities interact, interpenetrate, and at times contradict each other.
Examining how these people work alone together prepares us to better
apply activity theory to other examples of distributed, interorganizational,
collaborative knowledge work.
434 Journal of Business and Technical Communication 26(4)

Appendix A
Selected Texts
Text Description
T2 Conjunctured business plan
T7 Link Coworking business plan
T15 Soma Vida coworking policies
T16 Soma Vida ‘‘What is Coworking?’’
T30 Soma Vida Web definition of coworking
T32 Perch Coworking welcome page (with short definition of coworking)
T33 Space12 coworking page with coworking definition
T34 Brainstorm Coworking Web site with coworking definition
T36 Link Coworking frequently asked questions with short coworking definition
T37 Conjunctured—About page with history and coworking definition
T38 Conjunctured registration form (on Web site)
T39 Cospace members directory
T40 Link Coworking members directory
T41 Conjunctured members directory—full-time
T42 Conjunctured members directory—basic
T43 Conjunctured members directory—alumni
T53 Facebook page for FG2 ‘‘The Go Lab’’ (GoLab Austin)
T56 Yelp reviews for Space12
T58 Yelp reviews for Cospace
T59 Yelp reviews for Link Coworking
T60 Google Places reviews for Link Coworking
T61 Yelp reviews for Conjunctured
T62 Facebook page for Soma Vida
T63 Facebook page for Conjunctured
T64 Facebook page for Link Coworking
T65 Facebook page for Space12
T66 Facebook page for Cospace
T68 Facebook page for Perch Coworking
T72 Foursquare check-ins for Soma Vida
T74 Foursquare check-ins for Texas Coworking (the original name of Cowork
Austin)
T75 Gowalla check-ins for Conjunctured
T76 Gowalla check-ins for Cospace
T77 Gowalla check-ins for Link Coworking
T79 Gowalla check-ins for Space12
T80 Gowalla check-ins for Brainstorm Coworking
T84 Gowalla check-ins for GoLab Austin
Spinuzzi 435

Appendix B
Selected Codes
Code Definition
Selected starter codes
cw_definition What is coworking?
cw_history Proprietor’s history and background related to
coworking
cw_whyYou Why do you cowork?
cw_whyOthers Why do your coworkers cowork?
space_planning How did you plan your space?
space_events Do you hold events here?
reqs What are the minimum conditions for coworking?
space_collaborating How do people coordinate or collaborate here?
space_clients Do people meet clients here?
cw_evolve How will coworking evolve?
cw_challenges What challenges do you think your space will face
over the next 6 months to a year? Over the next
5 years?
cw_changes What are some critical changes that could help
coworking spaces or coworkers?
cw_diversity How diverse coworkers are re fields
cw_target Stated target market
Selected open codes
biz The coworker’s business
biz_costs Business costs, including costs of coworking space
biz_partnerships Partnerships or subcontracting in the coworking
space
biz_refer Referring business to each other in the coworking
space
biz_selfdesc Their self-description, including title, sector, or
status
coffeeshop Discussion of how people work in coffee shops
cw_community_likeminded Like-minded people in the space: similar in outlook,
profession
cw_community_managing How the space establishes and manages its
community
cw_comparison Comparing alternatives to coworking spaces
cw_comparison_spaces Comparing coworking spaces
cw_encourage Coworkers provide encouragement or support
cw_feedback Coworkers provide feedback or perspective
cw_location Location of the coworking space
(continued)
436 Journal of Business and Technical Communication 26(4)

Acknowledgments
The author thanks the proprietors of the nine Austin-area coworking sites discussed
here. They were and continue to be generous with their time. The author also thanks
the 17 coworkers who generously agreed to be interviewed. Bill Hart-Davidson and
Appendix B. (continued)
Code Definition
Selected open codes (continued)
cw_target Target market for this coworking space, including
general relationships (telecommuter, freelancer)
and specific occupations (Web developer,
architect); Differentiation from other coworking
spaces
cw_trust Coworkers can be trusted
gen_boomer Characteristics of boomers
gen_x Characteristics of Gen-X
gen_y Characteristics of millennials
learning Coworkers have to learn skills
model_mixed Mixed model
reqs_confidentiality Need for confidentiality in meeting clients,
associates, etc.
reqs_daycare Day care or child care options
self_motivation Coworker discusses ability to motivate self or
challenges to self-motivation
social_capital Social capital
space_commcenter Comparing coworking to a community center,
church, or nonprofit; interactions among these
space_furniture Furniture needs, purchases
space_pricing How they price their coworking space and
amenities
space_segment Segments for coworking
time_flex Time flexibility; degree to which space hours
accommodate flexibility for coworkers
work_home Work-home separation
work_home_distractions Distractions of working from home
Axial codes
ax_benefits What proprietors expect coworkers to get from
the site?
ax_trusted_community What sorts of trust coworkers expect in the site
ax_motivation What benefits coworkers expect to get from the
site
Spinuzzi 437

Mark Zachry provided very helpful comments on earlier versions of this article, as
did the two anonymous reviewers. Finally, JBTC editor David R. Russell provided
excellent feedback, especially in terms of refining the argument.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publi-
cation of this article.
References
Adler, P. S., & Heckscher, C. (2007). Towards collaborative community. In C.
Heckscher & P. S. Adler (Eds.), The firm as a collaborative community
(pp.11-105). New York, NY: Oxford University Press.
Benkler, Y. (2006). The wealth of networks. New Haven, CT: Yale University Press.
Bodker, S. (2009). Past experiences and recent challenges in participatory design
research. In A. Sannino, H. Daniels, & K. D. Gutierrez (Eds.), Learning and
expanding with activity theory (pp. 274-285). New York, NY: Cambridge Uni-
versity Press.
Botsman, R., & Rogers, R. (2011). What’s mine is yours. New York, NY: Collins.
Bradner, E., & Mark, G. (2002). Why distance matters: Effects on cooperation, per-
suasion and deception. In E. F. Churchill, J. McCarthy, C. Neuwirth, & T. Rod-
den (Eds.), CSCW ’02: Proceedings of the 2002 ACM conference on computer
supported cooperative work (pp. 226-235). New York, NY: ACM Press.
Burton-Jones, A. (2001). Knowledge capitalism. New York, NY: Oxford University
Press.
Castells, M. (1996). Rise of the network society. Malden, MA: Blackwell.
Castells, M. (2003). The Internet galaxy. New York, NY: Oxford University Press.
Clark, M. A. (2000). Teleworking in the countryside. Burlington, VT: Ashgate.
Corbin, J., & Strauss, A. C. (2008). Basics of qualitative research (3rd ed). Thou-
sand Oaks, CA: Sage.
Coworking. (n.d.). Retrieved May 23, 2011, from http://wiki.coworking.info/w/
page/16583831/FrontPage
De Jong, A., & Mante-Meijer, E. (2008). Teleworking behind the front door: The
patterns and meaning of telework in the everyday lives of workers. In E. Loos,
E. Mante-Meijer, & L. Haddon (Eds.), The social dynamics of information and
communication technology (pp. 171-187). Burlington, VT: Ashgate.
438 Journal of Business and Technical Communication 26(4)

Deskmag. (2011). Number of Coworking Spaces increases to 700 worldwide.
Retrieved May 23, 2011, from http://www.deskmag.com/en/number-of-
coworking-spaces-worldwide-700
Dixon, M., & Ross, P. (2011). VWork: Measuring the benefits of agility at work.
Retrieved from http://www.regus.presscentre.com/Resource-Library/VWork-
Measuring-the-Benifts-of-Agility-at-Work-Report-65d2.aspx
Dwyer, C. J. (2011). Contingent labor management: The Evolution of the Contem-
porary Contingent Workforce. Retrieved from http://www.aberdeen.com/
aberdeen-library/7023/RA-contingent-labor-workforce.aspx
Edwards, A. (2009). From the systemic to the relational: Relational theory and activ-
ity theory. In A. Sannino, H. Daniels, & K. D. Gutierrez (Eds.), Learning and
expanding with activity theory (pp. 197-211). New York, NY: Cambridge.
Ellison, N. B. (2004). Telework and social change. Westport, CT: Praeger.
Engestro¨m, Y. (2009). The future of activity theory: A rough draft. In A. Sannino,
H. Daniels, & K. Gutierrez (Eds.), Learning and expanding with activity the-
ory (pp. 303-328). New York, NY: Cambridge University Press.
Ferriss, T. (2009). The 4-hour workweek, expanded and updated. New York, NY:
Crown.
Geisler, C. (2001). Textual objects: Accounting for the role of texts in the everyday
life of complex organizations. Written Communication, 18, 296-325.
Goffman, E. (1959). The presentation of self in everyday life. New York, NY: Anchor.
Gurstein, P. (2001). Wired to the world, chained to the home. Vancouver, Canada:
University of British Columbia Press.
Gygi, K., & Zachry, M. (2010). Productive tensions and the regulatory work of gen-
res in the development of an engineering communication workshop in a transna-
tional corporation. Journal of Business and Technical Communication, 24,
358-381.
Hunt, T. (2009). The whuffie factor. New York, NY: Crown Business.
INRIX
1
national traffic scorecard. (2009). Retrieved from http://scorecard.inrix
.com/scorecard/pdf/INRIXNationalTrafficScorecardMY2009SpecialReport.pdf
Jones, D., Sundsted, T., & Bacigalupo, T. (2009). I’m outta here! How coworking is
making the office obsolete. Austin, TX: Not an MBA Press.
Kjaerulff, J. (2010). Internet and change. Arhus, Denmark: Intervention Press.
Kylin, C., & Karlsson, J. C. (2008). Re-establishing boundaries in home-based tele-
work. In A. Haunschild (Ed.), Work less, live more? (pp. 173-190). New York,
NY: Palgrave Macmillan.
Lompscher, J. (2006). The cultural-historical development of activity theory: Some
aspects of development. In P. Sawchuk, N. Duarte, & M. Elhammoumi (Eds.),
Critical perspectives on activity (pp. 35-51). New York, NY: Cambridge Univer-
sity Press.
Spinuzzi 439

Malone, T. W. (2004). The future of work. Boston, MA: Harvard Business School
Press.
Mark, G., & Su, N. M. (2010). Making infrastructure visible for nomadic work. Per-
vasive and Mobile Computing, 6, 1-12.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis. Thousand
Oaks, CA: Sage.
Mueller, M. L. (2010). Networks and states. Cambridge: MIT Press.
Paretti, M. C., McNair, L. D., & Holloway-Attaway, L. (2007). Teaching technical
communication in an era of distributed work: A case study of collaboration
between U.S. and Swedish students. Technical Communication Quarterly, 16,
327-352.
Ramsower, R. M. (1985). Telecommuting. Ann Arbor, MI: UMI Research Press.
Ruckriem, G. (2009). Digital technology and mediation: A challenge to activity the-
ory. In A. Sannino, H. Daniels, & K. D. Gutierrez (Eds.), Learning and expanding
with activity theory (pp. 88-111). New York, NY: Cambridge University Press.
Russell, D. R. (1997). Rethinking genre in school and society. Written Communica-
tion, 14, 504-554.
Smagorinsky, P. (2008). The method section as conceptual epicenter in constructing
social science research reports. Written Communication, 25, 389-411.
Spinuzzi, C. (2007). Guest editor’s introduction: Technical communication in the
age of distributed work. Technical Communication Quarterly, 16, 265-277.
Spinuzzi, C. (2011). Losing by expanding: Corralling the runaway object. Journal of
Business and Technical Communication, 25, 449-486.
Su, N., & Mark, G. (2008). Designing for nomadic work. In J. Van Der Schijff & G.
Marsden (Eds.), DIS 08: Proceedings of the 7th ACM conference on designing
interactive systems (pp. 305-314). New York, NY: ACM Press.
Toffler, A. (1970). Future shock. New York, NY: Random House.
Toffler, A. (1980). The third wave. New York, NY: Bantam Books.
U.S. Census Bureau. (2011a). Information. Retrieved July 25, 2011, from http://
www.census.gov/econ/census02/naics/sector51/51.htm
U.S. Census Bureau. (2011b). Nonemployer statistics. Retrieved July 25, 2011, from
http://censtats.census.gov/cgi-bin/nonemployer/nonsect.pl
U.S. Census Bureau. (2011c). Population estimates data sets. Retrieved July 25,
2011, from http://www.census.gov/popest/datasets.html
Yamazumi, K. (2009). Expansive agency in multi-activity collaboration. In A. San-
nino, H. Daniels, & K. D. Gutierrez (Eds.), Learning and expanding with activity
theory (pp. 212-227). New York, NY: Cambridge University Press.
Zuboff, S., & Maxmin, J. (2004). The support economy: Why corporations are fail-
ing individuals and the next episode of capitalism. New York, NY: Penguin
Books.
440 Journal of Business and Technical Communication 26(4)

Bio
Clay Spinuzzi is an associate professor of rhetoric and writing at The University of
Texas at Austin. His interests include research methods and methodology, work-
place research, and computer-mediated activity. He is the author of Tracing Genres
Through Organizations (MIT Press, 2003), which was named NCTE’s 2004 Best
Book in Technical or Scientific Communication, and Network (Cambridge Univer-
sity Press, 2008).
Spinuzzi 441