Content uploaded by Azi Lev-On
Author content
All content in this area was uploaded by Azi Lev-On on Feb 25, 2016
Content may be subject to copyright.
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
1/11
The Journal of Community Informatics, Vol 9, No 3 (2013)
Communities, Crowds and Focal Sites: Fine-Tuning the
Theoretical Grounding of Collaboration Online
Azi Lev-On
Introduction
In re cen t yea rs, Internet usa ge has be en moving in an increasingly pa rticipatory and
collabo rative direction, w ith ple nty of ve nues for us ers to create and share conte nt,
generate information reposito ries, manage social networks, and orga niz e collective
action. Online communities ena ble their members to communicate over time about
themes of common concern. Membe rs also de velop relations hips a nd reposito ries of
information and advice.
From a class ical rational-actor pe rspective, the success o f e ndevo urs that rely on such
collabo rations by many participants se ems p uzzling. Analyzed through the len s of the
"logic of co llective action" (Olson, 1965), collabora tions ba sed on contributions by
many individua ls without proper ince ntives are d oo med to fail. According to this logic,
individua l a ctions dominated by narrow s elf-interes t g enerally lead pe ople to a bs tain
from contributing to su ch collabo rations , and free -riding behaviors wo uld in turn
impede their success . Indee d, o nline collabo rations often ca nno t effectively address
the proble m of free -ridership, wh ich is at the core of e very collective a ction, an d are
typically cha racterized by high percentages of lurkers a nd a highly une qual division o f
labo r (se e Lev-On a nd Hardin, 20 07; Hunt and Johnso n, 2002). Such collabora tions
typically lack fea ture s tha t may pre vent collective a ction failures, such as co ntinu ing
relations among co ntributors, exp ecta tions of future re latio nships which cas t a
"shad ow o f the future" , or e mbeddedness of contributors in a dense ly-knit ne tw ork or
community (see Axelrod, 1984; Taylor, 198 7; Hardin, 1995).
From this pe rsp ective, communities see m we ll situated to support collective actions. in
that they enable on go ing communication betwee n members a nd introduce exit costs
as w e ll as e mbedde dness in a social ne tw ork. But ca n the uniqu e be ne fits of
communities for collective a ction be reproduced in online environmen ts? If so, where
and wh y? How might online ass ociations facilitate collective action? And, to w ha t
exte nt can such as so ciatio ns reproduce th e strategic capabilities of s ome offline
communities in circumventing the "logic of collective a ction"?
After po rtraying lea ding theories of communities and collective action, I argue tha t they
are ina de quate to explain collaboration in o nline communities. I offer two ro ute s as
sub stitute s to explain the role of online communities in grounding collective action. The
first route involves communities that eithe r ge ne rate ins titutions tha t su pport collective
actions, or are e mbedde d in large r networks that make e xit o ptions e xpe ns ive. Such
communities are a ble to overcome their inhe rent limitation s to ground collective
actions. The second rou te e mphasize s the a bilities o f online communities to a ttra ct a
large number of se lf-selected pe op le with s ha red co ncerns into a fo cal lo catio n, from
wh ich members can be recruited by entrep reneurs organizing collective a ction.
Communities: offline and online
In a clas sic survey a rticle, Hillery (1955) collecte d 94 definitions of the term
"community" from the s ociolo gical litera ture. He found tha t the tw o pre valent features
attribute d to "communities" we re interactional and geographical; i.e., inter-member
communication and geographic co-location (see W ellman, 2001, p. 228).
In the 17th and 18th centuries, in para llel with economic, social and technological
transformations o f daily e xpe rience s, the use of the con cept "community" in English
expand ed to include "the idea of a g roup of people who hold so mething in common...
or w ho share a common s en se of identity e ven if they do not live in a s ingle locale"
(Cole, 2002, p. xxiii). Gradua lly, the existen ce of common p hysical mee ting -places ha s
diminished as a neces sary condition for a group to be deno te d as a "community" ;
now a da ys communities are "base d on what we do with o the rs, rathe r than where w e
live with others " (Haythornthw aite, 2002, p. 159).
The application of the labe l "community" to Internet-bas ed asso ciations has be en
popularized b y Rhe ingold's (2000) influe ntia l bo ok ab out the W ELL community in the
San Francisco ba y a rea (cf. Smith, 1992; Hafne r, 1997). Such an a pp lication explicitly
undermines one of the deep-roo ted dimension s of "communities", i.e. geog rap hical
proximity among membe rs. Such as sociations complete the s eparation of communities
from "physical" spaces , and enable a "fundamenta l libe ration from place " (We llman,
2001, p. 238). Online communities can de velop aro und a variety o f p latfo rms, such as
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
2/11
social ne tw orks, blogs, fo rums, and wikis. Intera ctions among community members
need not be limited to the Interne t; they can interact throug h a variety of offline
means, using "w hate ver means of communication is convenien t an d appro priate at the
moment" (Wellman, 2001, p. 248).
Online co mmunities differ from one another in various re sp ects, n otably the kinds a nd
levels of attachment be tw e en members (Sass enbe rg, 2002). In so me associations the
attachmen t is inte rpe rso na l, while in othe rs it is an instrumenta l ide ntification w ith the
group. But affe ction a nd stro ng ties a re not defining or eve n ess en tial fea tures o f such
communities. On line communities also vary according to othe r paramete rs, such as size
and cohe sive ne ss . The y ra nge from large-scale, w ell-functioning forms on one end of
the spe ctrum, to small, loo se and diso rga nized forms on the o the r. Some res emble
"stree t corner settings or pa rk squares," while others "e xhibit s ome propertie s of long-
lived so cial groups or communities" (Butler et al., 2002).
For many Inte rne t us ers , such associations are important so urces of goods,
information, support, and se nse of belonging (We llman, 2001). Other users dee m
online communities a s motors o f re ciprocity, collabo ration, an d civic rene w al (Conno lly,
2001). O nline communities ha ve be come a key facto r in the e xpe rience s of Inte rne t
use rs. Back in 2001, a survey by the Pe w Internet a nd American Life Project s ho wed
that 84% of use rs indicated that they ha d contacte d an o nline community, a nd 79%
identified a t lea st one community with w hich they maintaine d regular contact (Horriga n
and Rainie, 2001).
Pe op le join o nline associations fo r a variety o f reas ons, including, no ta bly, the
opportunity to discus s issues of common concern an d obta in informatio n (Horrigan and
Rainie, 2001; Ridings and Gefen, 2004). By choosing to join a community, peop le
practically se lect who to communicate w ith ab out a specific topic the y co mmonly find
wo rth pursuing.
Online communities are often categ orized according to their prime reasons for
interaction (see P orter, 2 004). For exa mple: Communities of practice a re compo sed of
profe ssionals s uch as acade mics or s oftware p rogrammers, unite d by common purs uit
of knowle dg e a nd solutions to prob lems (Wenger, 1998). Communities of interest bring
together individua ls who s ha re an interes t in a specific topic, idea , or hob by.
Communities of s upport provide information and assistance abo ut problems such as
addictions , grief counse lling, o r he alth issu es (Braina rd, 2003). P lace -bas ed
communities (i.e., rural/urban online communities) bring together individuals w ho live in
or visit the same loca lity (Cohill and Kavanaugh, 1999).
A tension exists betwee n the "ho rizonta l" dimension of communities , ofte n
chara cterized by peer-production and user-ge ne rated conte nt, an d the ir "vertical"
management dimensio n, which can be highly centralize d. Online communities ha ve
managers, organizers, moderato rs, and des igne rs w ho enable the community, an d
carry out functions tha t are ess ential for "the framew ork aroun d which community may
develop" (Kim, 20 00; Preece, 2000). The ir roles include mana ging membership (for
exa mple, re cruiting ne w members a nd manag ing member profiles); man ag ing content
(age nda se tting, facilitating and encoura ging discus sio ns, mode rating discuss ions,
preventing "flaming" and removing inapprop riate pos ts, archiving old thre ads of
discus sion, producing special e vents); framing and enforcing po licies regarding
accep ted be ha viors and sa nctions, as w ell as te chnical and financial mana gement.
But wh ere as a small numbe r of peo ple may administrate the platform us ed by many,
still the orie nta tion o f the "horizo nta l" dimens ion of online communities can be quite
egalitarian. Unlike firms or governmental bure aucracies , typically no ela bo rate and
formal chain o f command and con trol exists among us ers , nor any autho ritative division
of labor; inste ad, the emphas is is on pee rs as originato rs of con tent, and of peer-
base d pro duction, monito ring, and sanctio ning.
The pe er-bas ed a nd o ften vo luntary nature of contributions make s such communities
vulnerable to myriad collective a ction and co ordination failures. As Kollock (1999, p.
220) a rgues , "for a student of so cial o rde r, what nee ds to be explaine d is not the
amount of conflict but the grea t amou nt of sharing and coo pe ration tha t doe s occur in
online communities." Such sharing and cooperation a re manifest not jus t b y the
prese rvatio n of ord er, but a lso b y collective efforts to gathe r fun ds , crea te datab ase s
of information a nd advice, and eve n produce complex public goo ds (the no table
exa mple is the pee r-prod uction o f computer softw a re). W hy, then, do such
as so ciations succeed in mobilizing collective actio n, when the y do?
Communities and collective action
To explain community-ba sed collab ora tion online, s ome authors borrow establishe d
framew orks use d to e xplain coopera tion in offline communities , no ta bly O strom's
institution-bas ed account (se e Kollock and Smith, 1996; Kollock, 1998; C he sney, 2004;
Van W ende l de Joo de , 2004; Viéga s, Watten be rg and McKeon, 2007). I fo cus on two
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
3/11
conte mporary and complementa ry ave nues of research: Hardin (1995) emphas ize s
conventions and "coordination power" as keys for explaining community-grounded
collective action, and Ostrom (1990) p ays attention to institutional structures that
support cooperation in such communities.
According to Ha rdin (1995), the key to unde rsta nd ing the success o f ass ociations to
mobilize collective action is the ir ability to generate conventions, supporte d b y
meaningful and e ffective s anctions and ga ins. Once s uch con ventions evolve and a re
recog nized by a critical mass of community membe rs, the y create mutual expecta tions
for co mpliance, and practically regulate the intera ctions of co mmunity members . (The
origins of conventions are not importan t for our discus sio n, nor do we ha ve to as sume
that community members internalize conventions and attribute moral value to the m).
Two key facto rs suppo rt the convergence on communal con ventions: th e absence of
inexpensive exit options, and the multiplexity of th e communal experience. The
expecta tions of meaningful exit losse s (for e xample moneta ry loss es , los s of social
ties , and loss of sources of information or sup po rt) can motivate members to stay
wh en they consider leaving (Hardin, 1995). Viable exit options, on the o the r hand,
dramatically narrow the "sha do w of the future" a nd may disable mea ningful sanctions.
Instea d of voicing conce rns o r organizing prote st, membe rs can just leave , o r in the
case of online asso ciations, "unsubs cribe."
Moreover, in tightly-knit communities membe rs inte ract w ith one a no the r on a re gular
and continuous b asis, and convention s can cove r many spheres o f da ily lives (Co ok
and Hardin, 200 1). Wh en individuals are e mbedde d in a community in the se ways,
deviating from conve ntions may have gra ve consequences . Failures to comply w ith
conve ntions lead not o nly to loca l coordination failures, but a lso to los ses (for s elf and
possibly for significant others) in a variety of spheres in w hich community members a re
interconne cte d. In tightly knit a ss ociations, members w ho fail to participate in a
collective action can be sanctioned not only in the conte xt of the specific actio n, b ut in
multiple conte xts . Pe er-san ctioning is poss ible and probable-not just by a specific
member w ho bears the direct w eight o f defe ction and not only in a spe cific conte xt-but
by all membe rs and in multiple conte xts. The probability of sa nctions motivates
members to a bide by local co nventions. The common know led ge that a nti-normative
behaviors w ould be sa nctio ned is a signal for po te ntia l "de fecto rs."
Outside the communal conte xt, a nd w itho ut the communal conve ntio ns as reg ulating
principles , community members might have followed their sh ort-te rm se lf-intere st a nd
failed to co ordinate on solutions to the ir shared dilemmas. The s alience of communal
conve ntions thus ass ists membe rs in a voiding coo rdina tion failure s. In this se ns e,
communal convention s function as co ordination equilibria, and the a dhe rence of
enoug h community members to communal conventions establishes a "coordination
power" of the community. The existe nce o f a w ell-coordina te d community increa se s th e
costs of cha lleng ing de eply e ntrenched convention s, and "make s certain be ha viors o n
the part o f re leva nt othe rs less rew arding than they w ou ld be a ga inst an
uncoordinate d group" (Hardin, 199 5, p. 30 ). C ollective a ctions that are inconsistent
with communal conve ntions are almost certa in to fail.
In addition to their stra te gic advantag es, communities also sup po rt collective a ction by
solving info rmation proble ms. The thick relations hips a mong members, a nd their
repea ted and fre quent interactions , fa cilitate high ly efficient circulation of information
(for e xample, about the identities of coop erato rs and no n-coope rators). In this sense
communities fun ction a s information intermediaries.
Wh ere as Hardin's w o rk focuses on the emerge nce of loca l conventions a nd the re turns
from " coo rdina tion pow e r," Ostrom (1990) focuse s on the institutions that ground
coop eration. In her s eminal w ork Governing the Commons, ba se d on exte ns ive
fieldw ork, Ostrom iden tifies s everal key variable s th at a ffect be ha viors, de cision s, and
outcomes in communal mixed-motive scena rios . Severa l "design principles" are key for
achieving long-enduring coope rative ou tcomes in such scena rios : the ability of
as so ciation members to create and modify institutions and rules to a ddres s loca l
needs; co rres po ndence b etwee n the rules a nd no rms gove rning the use of co mmon
resources, and b etwee n local conditions ; and the a bility o f most pe ople affected b y
institutions to pa rticipate in custo mizing and trans forming the m. Thes e principle s are
not es pe cially problematic in online communitie s, and will not be further discuss ed
below ..
Three additional facto rs de serve our add itiona l attention; namely, the a bility to fashio n
arenas for low -cost co nflict reso lution, to e stab lish boundaries to the a ss ociation, and
to monitor co op era tion and s anction non coope rators. Os trom (1990) a rgues th at
boundaries must be clearly defined to pre vent individuals from ente ring th e community,
using its res ources, and exiting w itho ut contributing. In this se nse, estab lishing
boundaries is a s tructural s olution to the non-excluda bility property of p ublic goo ds .
Structural bo undary rules can involve members hip in organiza tions, personal
chara cteris tics, relationship to the res ources, or a combination of the se prope rties
(Ostro m, 1999).
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
4/11
Creating bound arie s requires the ab ility to identify ass ociation membe rs. Ide ntificatio n
distinguishes members from non-members, e stablishe s whether they have bee n
participating in collective a ction, and de te rmines w he ther they a re entitled to be ne fit
from the communal res ou rces and und er wha t conditions.
Rega rding the impo rtance of monitoring and sanctioning, s ocial control rests in large
part on th e a bility to identify individuals an d ho ld the m accountable for their actions. As
a result of monitoring failures, trust in specific as so ciation members may fail to
develop, and distrust can be ge ne ralized to the e ntire as so ciation.
A preco nditio n for monitoring is pub lic visibility, such tha t tho se w ho monitor
participation are ab le to identify contributors and free riders. Visibility also be ne fits the
community be cause it discourages de fection, due to fear of sa nctions and s ha me from
humiliation. Visibility can als o encourage coope ration. The possibility of being obse rved
makes one act publicly in w ays tha t as sociation members find praiseworthy, and
makes po ss ible the provisio n of as so ciational aw ards for acting in normatively
sup po rted ways . Sanctions can include gos sip, hostile rumors, lo ss of re lations hips
with s pe cific community membe rs, damage d rep uta tions, or sh unning by all members.
The se verity and effectivene ss of sanctioning can make formal, costly sanctioning
unne cess ary.
Ostrom's framew o rk, which is base d on res earch do ne in offline a ssociations, has be e n
use d by seve ral authors to a na lyze the pos sibilities of collabo ration in o nline
communities. But can s uch frameworks be un eq uivoca lly applied to s tud y collabo ration
in online communities? In the next se ction I argue that e xplan atory framew orks
empha sizing e xpe nsive exit and emergent conventions and institutions are, a s a
general rule, ill-suited to e xplain collaboration in such online association s.
Explaining collaboration in online communities using theories
developed for offline communities
Online communities, as explored thus far, are similar in so me res pe cts to offline
communities, but are distinct in significant re sp ects. Offline "trad itiona l" communities
are commonly chara cterized as ha ving thick and multi-layered relations among
members, an d costly exit op tions. In such e nvironments, communa l conventions ca n
become highly efficient means for so cial co ntro l. In comparison w ith such "tra ditional"
as so ciations, online communities can attract many pe ople. Membe rship is not "by
chance" but "by choice" (Galston , 1999). But online communities typically lack the thick
relations tha t chara cterize offline communities, and ha ve viable exit options . The
abilities to form conventions, monitor the performan ce of community members, and
sa nction defectors can be called into questio n in s uch environments . He nce, argua bly,
framew orks use d to explain the s uccess of offline community-bas ed co llective actio n
are ina ppropriate to exp lain much co llective actio ns tha t occur through a ssociation s
online .
Recall tha t Hardin's a ccount of the contributio n of communities to collective a ction relies
on the ir ability to genera te conventions th at regulate the interactions o f community
members. Much of their success ca n be attributed to the multiplexity o f conne ctions
among community members . But when applied to e xpla ining collabora tions in online
communities, such e xplanatory framew o rks run into a number of obsta cles . First, the
lives of members of online communitie s are typically not inte rtw ined in a variety of
spheres. Whe n members are dens ely embedde d in their communities , leaving the
community enta ils los ing the many benefits it provide s in all sphe res of membe rs' lives.
Arguably, the multiplexity of relations in tightly-bound communities is now a da ys
replaced by limite d involvement in multiple narrow communities (We llman, 2001). When
online co mmunities focus on rela tively narrow sphere s, the intera ctions b etwee n
members ten d to focus, na turally, on the se relevant sphe res. Pe op le be long to various
communities, each w ith its specific function, and the a sso ciational expe rience need not
expand beyond this function. In this sense, such online asso ciations fall short of
generating a multiplex expe rience of such "idea l-type" or "traditiona l" community life.
A seco nd obstacle to the crea tion of "coordina tion power" in online communities is the
ea se of exit. " Traditional" co mmunities se em to spontaneously e volve o ver time, a nd
their members may be bo und to the community by social, econo mic, and cultural forces .
Lea ving the community may be costly and s ometimes impossible (Komito, 1998). By
contra st, online ass ociations exist in a competitive environment, where numerous
as so ciations vie for us ers ' a tte ntion. Online communities are q uite o ften the re sult o f
"artificial" construction by peo ple or group s with a s pe cific agenda. Unprofitable
as so ciations can be d ismantled a nd disa ppea r overnight, w he n someo ne lite rally p ulls
the plug on the m. No tably, in co mmunities that e volve around pee r-prod uction of
open-source projects , viable e xit o ptions and the ability to "fork" the project are ofte n
of utmost ideolog ical importan ce, and furthe rmore "open s ou rce licenses are design ed
explicitly to empow e r exit rather than loya lty as the alternative when voice fails"
(Webe r, 2004, p. 159; see also Van W ende l de Joode, 200 4).
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
5/11
There fore , unlike so me "ide al-type " communities, that are so metimes cha racterize d as
despotic due to the punishments tha t can be inflicted on non-conformists, online
communities te nd to have the o pp os ite problem - of retaining members. Since ag en ts
usu ally join fo r a spe cific purpo se, they may not experience multi-dimensiona l los ses
wh en they e xit. Members can leave "w hen the re a re disagreements or even if things
just ge t bo ring" (Komito, 1998, p. 102). Online communities, in this reg ard , are
as so ciations of potential exite rs, who can simply leave the asso ciation wh en a sked to
contribute to collective action. As Gals ton (1999, pp. 59-60) nicely puts it,
If we are linked to others by choice rather than accident, if our interaction with them
is shaped by mutual adjustment rather than hierarchical authority, and if we can set
aside these bonds whenever they clash with our individual interests, then the lamb
of connection can lie down with the lion of autonomy.... Because they emphasize
exit as a response to discontent and dissatisfaction, they do not promote the
development of voice.
As fa r a s collab ora tion is involved, the problem w ith online communities is that th eir
"coo rdina tion power" may not be strong enoug h w he n needed. In social dilemma
situa tions, w ea k a ssociation al bo nds can simply be ove rloo ked in favo r o f s hort-term
se lf-interest. Whe n there are viable exit options, and when the embeddedne ss of
members in the a ss ociation is highly imperfect, community membe rs may fail to
participate in the crea tion o f p ublic goo ds , w ith no s ignificant a dve rse cons eque nce s.
In such cases , convention s fail to perfectly regulate the inte ractions among as so ciatio n
members, and the benefits of "co ordination po w er" are disabled.
Third, the institutional variables ide ntified by Os trom (1990) as instrumental for driving
collabo ration in a ss ociations are als o pro blematic in the context o f online communities.
Three fa ctors in pa rticular are worth clos er examination: monitoring , sanctioning, and
boundaries. De spite their structura l significance, boundaries are ra rely established and
maintained by on line communities. Membe rship is often fluid, and members can jo in
and drop o ut at w ill. O ften the re is no mea ningful distinction be tw e en members a nd
gue sts, and no clea rly state d a nd enforced members hip application procedure s or
requirements.
As regards monitoring and s anctioning, there is a continuum in the de gre e of
institutionaliza tion of online communities . Yet ofte n thes e communities do not
authenticate members ' ide ntitie s. Hence, monitoring members' a ctions, holding the m
accounta ble, and inflicting sa nctions may be imposs ible. Online communities often
focus on a single issue, and members join due to their interest in this issue. Sanctions,
then, a re re leva nt to a s ingle aspect of members ' lives -an aspect tha t was important
enoug h for a membe r to e xplo re further by joining the community, b ut still a single
as pe ct. So the p os sibilities of s anctioning se em quite limited, e ven w he n visibility is
possible.
The remaining facto rs identified b y Os trom (1990), including the ability to cons truct
institutions, rules, a nd low -cos t conflict-reso lution a renas, are not particularly
challen ging or demanding on line. Local rules and institutions are ge ne rally se ns itive to
the nee ds a nd intere sts of members , a nd in many communities ins titutions a re pee r-
prod uced. The community portal is instrumental for notifying members a bo ut rule s and
policies, for framing the n ormative environment, and for indoctrinating a nd s ocializing
newco mers. Many communities a lso include "constitutive do cuments" s uch as mission
sta te ments, gene ral policies, and lists of fre quently-aske d questio ns and an sw e rs.
Loca l no rms a re ofte n inferred fro m typical behaviors or from key attributes o f
members, and are reinfo rced ove r time. Conflict-resolution and problem-solving
mechanisms vary, and are by and large experimental in nature. They can include a
variety of to ols fo r vo ting, discuss ion, conse ns us-creation, itera ted pe er-pro duction of
solutions to communal problems, and more (see Van Wende l de Joo de , 2004).
Note tha t the obsta cles that jeo pa rdize collective action in online communities are not
the unavailability of no rms, rules, a nd conflict-resolution mecha nisms, but ra the r their
ineffectiveness. Due to the viable exit optio ns, local rules may be little more than
"pa rchment barriers " tha t canno t be effectively enforced when inte rests are
significantly non-aligne d.
The analysis th us far sugg es ts th at online communities may face major obstacles to
es tablishing "coo rdina tion power" a nd motivating members to co ntribute to non -trivial
collective actions . Online communities see m to b e more vulnera ble to coo pe ration and
coord ination failures than their o ffline counte rparts. In particular, the single -
dimens iona lity of the community experience , ea sy exit options, and the abse nce of
clear boundaries, e ffective monitoring and sanctioning, re nder the tra ditional
explana tory framew o rks of Hardin (1995) and n otably O strom (1990) impractical for
explaining why online communities induce collabo ration, w he n they do .
Routes for explaining community action online
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
6/11
So far I ha ve arg ued tha t collabo ration in online communities cannot b e an alyzed b y
"traditional" e xplanations, fo cusing on s ocial control and its de te rminants . In the ne xt
two se ctions , I propose two a lte rnative route s for explaining o nline community-
grounde d collective actions. The first emphasize s that some online communities are
"privileged" ove r others in that their institu tional structure o r netw o rk e mbedde dness
place th em in a be tte r pos ition to mobilize their members to p articipate in collective
actions. The second route e mphasizes the function ing of some online communities a s a
hub for recruiting potential contributo rs to emerging collabo rations .
As was note d a bove, online communities manifest a n inte res ting d uality be tw een the
inegalitarianism and decentra lization o f their "horizontal" dimension of pe e r-production
and use r-generate d content, and the highly ce ntra lize d contro l that can b e e xercise d
by manage rs. Thus, even w he n exit is ine xpe ns ive and multi-dimensionality is absent,
community managers can still "compensa te" fo r some of the shortages surveyed
above, by design ing a more "pro-so cial" infras tructu re for the community (Donath,
1996; Kim, 2000; Pre ece, 200 0). For example, des igne rs can b uild into online
communities:
Boundaries to differe ntiate members from guests; for exa mple, req uiring membe r
registration, cha rging member fee s, or allow ing premium acces s to certa in areas
of the community.
Visibility and authenticatio n of identities; for example, requiring unique a nd
persistent identities. The succes s of the WE LL, the first large-scale online
community, to sup pre ss "flaming" and other anti-so cial be ha viors, has been
attribute d in pa rt to a po licy of YOYOW (You Ow n Your Own Words) (Rheingo ld,
2000; Hafner, 1997; Smith, 1992).
Visual repre se ntatio n of collective actions ; for example, software applications can
emulate social pra ctices like que ue s, or provide indicators about members '
contribution levels during the production of con tinuo us pub lic goods.
Online sanctioning; for exa mple, hostile rumors and "flaming," shaming and
humiliation in public rituals, "ignoring" o r "ga gg ing" objection ab le be ha viors ,
filtering out the inpu t of specific us ers, tempo rarily or pe rmane ntly restricting
members' rights, and banishing users from particular areas of the community, or
from the community altog ether (Kollock and Smith, 1996).
Thus, the institutional variable s emphas ize d by Os trom (1990) as esse ntia l for
community-based collective action can be introduced into online communities by the ir
managers, to a certa in d eg ree . Ho w ever, the variables emphas ize d in Ha rdin (1995)
as e sse ntia l for coo pe ration-multiplexity and expensive exit-may b e lacking. From this
perspective, the communities tha t are bes t able to contro l their members ' be ha viors
are thos e wh ere the community e xpe rience is enmesh ed in a variety o f o the r spheres
of their members ' lives, online a nd particularly offline. Whe n online communities are
partially embe dd ed in offline ties, failures to comply may have conse qu en ces in a
variety of domains, o nline a s w ell a s offline. In s uch cases , sa nctio ns for und es ired
behaviors may be fea sible, and expansive exit options may increas e compliance.
For e xample, membe rs of so me co mmunities of practice a re mutua lly embe dd ed in a
profe ssional environment, are familiar w ith each others ' opinions and work, a nd may
meet face-to-face occasiona lly (see Koku, Nazer and W ellman, 2001, and Matz at, 2009
on online academic as sociations , and Ha ytho rnthwaite , 2008 o n online lea rning
communities). Such online communities of pra ctice enlarge the sca le of the
corres po nding o ffline asso ciations, improve information flows , and make the
as so ciations more cohe sive. Fee dback an d sanctions ca n be vital to members' caree rs,
es pe cially whe n carried out in public before the entire ass ociation, and w he n all
dispute s a re recorded and sto red in community archives. In s uch as sociations , a cting
in w ays that a re inconsis tent with conventions may ha ve adverse consequences for
members.
The conve ntio n-ba se d account of collective a ction may a lso s till be e ffective in some
place-based communities w hose members are not spread geo gra phically. Such group s
have opportunitie s to "meetup" and orga nize collective action. Information about
residents or local figures whos e actions are inconsis tent w ith the inte re s ts or values of
the local community is ea sily circulated a mong loca l re sidents, for w hom such actio ns
have direct implications. In this case as well, failure s to coop erate in a s pe cific action
may have ad vers e offline conse quences.
The success ful mobilizing capacities of s uch place-base d as so ciatio ns are
demonstrated in a stu dy by Hampton (2003, pp. 425-426), who repo rts on the
outsta nd ing succes s o f the residents o f "Netville" in protesting ag ains t the de velope r's
decision to re vise loca l deve lopment plans. Similarly, a nd pe rha ps more impress ively,
Mele (1999) repo rts on the collective efforts of the residents of Jervay Place (in
Wilmington, North Ca rolina ), the first public-hous ing deve lopment in the US with a
resident-mana ge d website . The re sidents organize d through the w eb to oppose the
housing a uthority's decision to exclude them from p articipating in future p lanning. To
appreciate thes e n ove l online-bas ed o rga nizatio nal cap ab ilities, keep in mind
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
7/11
Granovetter's (1973) class ic wo rk tha t demonstra te d th at the inability o f local
residents to effe ctively o rga nize against urba n renewal was a conse quence of the
abse nce of "w eak ties" th at may ha ve otherw ise be en used to mobilize res ources a nd
sup po rt (see als o Ishii and Ogasa hara, 2007 for offline embedde dness of online
networks, and Dannecker and Lechner, 2007 for embe dd edne ss o f a so cial s upport
community).
Online communities as focal sites for attracting contributors to
collective actions
So far I have explored o ne route for exp laining community-base d collective action-
"compe ns ating" for the inh ere nt limitations o f online community to s upport
collabo ration. Such compensation could involve maintaining a pro-social institutional
design or embedding the community in large r networks, w hich increa ses the
multiplexity of the community e xpe rience, a nd enha nces sa nction ing poss ibilities and
increase s e xit costs.
This se ction focuses o n a se cond route for e xplaining the succes s of collective actio n in
online communities-their functioning as meeting -place s for large numbe rs of individuals
with s hared conce rns, making the m hub s from which entrepreneurs can re cruit
potential contributo rs.
Note that d ifferen t s ocial media p roje cts demand va rying skills and inves tments from
contributors (Lev-On and Hard in, 2007; See also Brabham, 2008):
1. Some proje cts thrive on the coo pe ration of "non-experts " w ith minimal skills. In
such case s, a large number of "a mate urs " is necess ary to ge ne rate a sufficient
amount of con tributions . For exa mple, grid-computing projects, such as
SETI@home (Holoha n and Garg, 2005), slice tasks into small pieces that are se nt
to volunte ers.
2. Othe r proje cts involve contributions by agents with private information; for
exa mple, collaborative journalism or product reco mmenda tion sites. In such
cases , the large r the contributor poo l, the grea ter the pos sibility of recruiting th e
people w ith local expe rtise .
3. Lastly, complex proje cts require a division o f labo r, where different age nts
specialize in differe nt aspects of the proje ct; for example, the peer-productio n of
open sou rce softw a re or encycloped ias . Here, the more voluntee rs, the higher
the prob ab ility of finding experts a cross seve ral asp ects of the project.
Note that many such collabo rative p roje cts do no t rely o n managerial hierarchies or
markets to organize production, nor eve n on community conventions to encourag e
contributions. The last point de se rves spe cial empha sis, as peer-pro duction is
sometimes contra sted with government- and firm-based production, but conflate d w ith
community-based production (see Haythornthw aite, 2009). Still, many such projects
are base d on contributio ns by individuals without prior acqua intance, a nd with no
expecta tions of future interactions . Such massive-scale pro jects succeed in spite of the
near-abse nce o f intera ction among st contributors ; inte r-member communication is not
a key genera tor of such collective a ctions (se e Lev-On and Hardin, 2007).
Pa radoxically, then, online communities cannot generate the "coo rdina tion pow e r" that
induce s members' cooperatio n. At the sa me time, collabo ration be tw een large groups
of s trange rs can succeed without s uch as so ciations. Even in such ca se s, on line
communities can be of use to collective action, in that they function as hu bs for
recruiting potential contributo rs.
Online a ss ociations have no pra ctical limit on th eir scale. As a re sult, they can enjoy
input a nd feedba ck by many co ntributors (altho ugh, in a ll likelihood, as the sca le go es
up the cohesiveness o f the as sociation declines). W hen a member posts a query to
the asso ciation's online space , a large audien ce-including some local "experts" or
members with private information-is available to provide an answer, comments, or
references. If the comments are made pub lic b ut are deficient or incomple te, other
members can provide correctio ns and fee dback.
Online a sso ciations are also s elf-selecting. As state d a bo ve, pe op le choose to beco me
members o f a ssociation s they find a pp ealing, and can leave w he n they lose interes t.
Self-selection go es with se lf-screening: those w ho sta y a re no t forced to do so , but
rather choose to sta y; tho se who leave choose to do so-they typically a re not
ostracized. In add ition, thos e w ho stay and b eco me members are likely to be
know ledgea ble a bo ut the central the me of the a ss ociation, and motivate d to a cquire
furthe r know ledge or even be come involved in re leva nt collective actions if so aske d.
As a res ult of su ch dyna mics, online ass ociations can in time d evelop into "have ns fo r
anyo ne who re lishe s the opportunity to inte ract w ith othe rs w ho are similar" (Shapiro,
1999, p. 49). Such as so ciations often include "high proportions of pe ople who enjoy
ea ch other an d low proportions of people w ho are forced to interact with ea ch o ther
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
8/11
becaus e th ey are juxta pose d in the sa me neighbo rho od, kinship grou p, orga nizatio n,
or w orkplace" (We llman 2001, p. 235).
The emergence of large-sca le "info rmal" as so ciations of self-selecte d members o nline
is a distinct phe no meno n. Online communities can be compos ed of a large talent pool,
wh ere members choose to pa rticipate in e xplo ring an d discus sing the mes they share
an intere st in. Due to their scale and the s elf-selection of members, online communities
can turn into focal site s that attra ct many po tential contributors, who can the n be
mobilized to participate in collective a ction.
Typically such collective efforts are of interes t to large numbe rs of pe ople, but at the
abse nce of orga nizationa l infras tructure, such cause s may not attract and mobilize
enoug h support (i.e. Olso n, 1965). Fo r e xample, citizen-base d campaigns to re -
eva luate and recons ide r p ublic policies (Brabha m, 2008), orches trated demonstrations
and ra llies, or efforts to revise local de velopment plans all re quire that a ctivists.and .
sympathize rs are a ble to lea rn about the particular projects or eve nts and receive
relevant information. Successful collective efforts depend on th e existe nce of e asily
acces sed focal sites a t w hich o rga nizers, activists, and sympathize rs can converge to
coord inate their efforts. Online commun ities can support collective efforts in this sens e by
providing a channel for people with similar interests to coordinate their efforts.
For exa mple, en trepreneurs w ho focus o n ga y issues kno w that o nline asso ciations of
gays a sse mble a niche audie nce tha t can be effectively re ached and mobilize d, a nd
hence may cho os e to focus the ir efforts in s uch associations . On line ass ociations of
immigrants allow mobilization efforts in immigration-relate d the mes . Eve n fans of
sports , music, or television (Baym, 1997) can be routed into impres sive collective
efforts. For e xample, if members of a soap opera online ass ociation do no t like a
certa in twist in the plot, the y ca n (and do) orga nize a prote st a nd s end messa ge s to
the p roducers or the n etwork. The orga nized e fforts of members o f s uch communities
make a differe nce. Pro ducers care about ra tings, a nd w hen the online as so ciation of
fans is large a nd prolific enoug h, the dissa tisfaction o f membe rs can have
cons eq ue nces for th e caree rs a nd earnings of pro ducers, writers, and actors.
There a re o the r wa ys to d irect pe op le to e merge nt collective actions. Sea rch engines
also assist in routing peo ple to collab ora tive projects (Lev-On, 2008), as do online a ds
in ge ne ral-interest site s. But o nline ass ociations are uniquely situa te d to ro ute
potential contributo rs to relevan t collab ora tive projects, be cause unlike search e ngines
they include large pools o f agents w ho se lect to jo in the a ss ociation a nd ha ve some
interest or e xpe rtise rele van t to the focal the me of the community. Hence, online
communities can function as effective hub s for rou ting their members to relevant
collective e fforts .
Conclusions
In a dd res sing the mutual interactions b etwee n us ers, communities, and technologies,
I ha ve ana lyze d ho w technologies enable use rs to orga nize and collaborate , a nd how
and w hy the y make communication and communities functional and use ful for
promoting their goals.
Like the ir offline counte rpa rts, o nline co mmunities a re roote d in inte r-member
communication and member-genera ted co nte nt. But fa ctors that traditiona lly suppo rt
coop eration in o ffline communities, such as b ounda ries , monitoring oppo rtunities, and
meaningful sa nctio ns, do no t exist in many online communities . Furthermore, obsta cles
for exit are low , and the community exp erie nce is ofte n one-dimensio nal.
Hence, to understa nd collabo ration in online co mmunities w e cannot u se the
conceptual lens es of traditional framew orks. Fa ctors such as easy e xit, na rrow focus ,
fluid bounda ries , and difficult monitoring and sanctioning, limit the app licability of
framew orks s uch as O strom's (1990) a nd Hardin's (1995) to explain collabora tion in
online communities.
The key to conceptua lizing collective action in o nline communities lies , ins tead , in tw o
routes : eithe r in their abilities to compensate for their inhere nt limitations for
sup po rting collective a ction by, for e xample, deve loping institutions tha t suppo rt a nd
eve n incentivize pro-social b ehavio rs; o r by embe dding the community in large r
networks. The se cond route emphasizes the functioning of online communities as
meeting-places for many se lf-sele cted, like-minded p eo ple - "orga nizationa l hubs" from
wh ich potential contributo rs to rele van t collective actions can be recruite d.
Scholars highlight a few cons eq ue nces of the e mergence o f la rge -scale a nd narrowly
focused communities online. Ha ge l and Armstrong (1997) argue that online
communities constitute a high-quality environment for vendors to ta rge t po tential
cons umers , and comprise focus groups for market re se arch. Suns te in (200 1) argue s
that due to the convergence of a large number of like-minded pe op le convers ing
primarily amongs t the mse lves a nd hea ring e choe s of their own vo ices, s uch
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
9/11
communities a re bree ding grounds for political radicalization.
Arguably, the very same fea ture s tha t make online communitie s prime targets for
vend ors and advertisers, and bre eding ground for extremism, also make them hotbe ds
for collective actions. Such actions results when orga nizers a nd entrepreneurs tap into
the e no rmous promise of many potential collaborato rs. Such communities ca n not only
se rve a jump-sta rt to direct members to relevant collective efforts, but also establish
institutions tha t supp ort the local production of complicate d and even "impossible
public good s" (Kollock, 1999, p. 230).
References
Baym, N.K. 1997. Interpreting so ap opera s a nd creating community: Inside an
electronic fan culture. In Culture of the Internet, ed. S. Kiesler, pp. 103-142. Manhaw ,
NJ: Lawre nce Erlbaum Asso ciate s.
Brabham, D.C. 2008. Crow d so urcing a s a mode l for problem so lving: An introduction
and cases . Convergence 14: 75-90.
Brainard, L.A. 2003. Citize n organizing in cybers pa ce: Illustra tion from hea lth care
and implications for pub lic administra tion. American Review of Public Administration 33:
384-406.
Butler, B.S., L. Sproull, S. Kiesle r, and R. Kraut. 2002. Community effort in online
groups: Who does the work and why? http://opensource.mit.edu/pap ers /butler.pdf
(accesse d May 20, 2010).
Che sney, T. 2004. 'Othe r people be ne fit. I benefit from their work': Sharing guitar
tab s o nline. Journal of Computer-Mediated Communication 10(1).
Cohill, A.M., and A.L. Kavanaugh. 199 9. Community Networks: Lessons from
Blacksburg, Virginia. Boston, MA: Artech Hous e.
Cole, M. 2002. Fore w ord : Virtual communities for learning and deve lopment - A look
to the pas t an d some glimps es into the future . In Building Virtual Communities, eds. K.
A. Renninge r an d W . Shumar, pp. xxi- xxix. New York: Cambridge Univers ity Press .
Conno lly, R. 2001. The rise and persisten ce of the technological community idea l. In
Online Communities: Commerce, Community Action and the Virtual University, eds. C.
We rry and M. Mowbray, pp. 317-364. Uppe r Saddle River, NJ: Prentice Ha ll.
Cook, K.S. and R. Hardin. 200 1. Norms of coopera tivenes s a nd networks of trust. In
Social Norms, ed s M. Hechter and K. Dieter, pp. 327-347. New York: Russell Sage
Founda tion.
Dannecker, A., and U. Lechner. 2007. Online an d offline integration in virtua l
communities o f patients: An empirical an alysis. In Communities and Technologies, eds .
J. Ste infield, D. Pentland, J. Ackerman, and J. Contra ctor, pp. 151-170 . Lond on:
Procee dings of the third Co mmunites and Techno log ies Conference .
Dona th, J.S. 1996. Inhabiting the Virtual City: The Des ign of Social Environments for
Eelectron ic Communities . Ph.D. diss., Massachuse tts Institute of Technology.
Galsto n, W .A. 1999. (How) doe s th e Internet affect community? Some speculatio ns in
se arch o f evide nce. In Democracy.Com? Governance in a Networked World, eds . E. C.
Kamarck and J. S. Nye, Jr., pp. 45-61. Hollis, NH: Hollis Publishing Company.
Granovetter, M. 1973. The stre ngth of we ak tie s. American Journal of Sociology 78:
1360-1380.
Hafne r, K. 1997. The epic saga of the WELL: The wo rld's mos t influential online
community (and it's not AOL). Wired 5.05: 100-144.
Hagel, J., an d A.G. Armstro ng. 1997. Net Gain: Expanding Markets through Virtual
Communities. Bos ton, MA: Harvard Busines s School Pres s.
Hampto n, K.N. 2003. Grieving for a lost netw o rk: Collective action in a w ired suburb.
The Information Society 19: 417-428.
Hardin, R. 1995. One for All: The Logic of Group Conflict. Princeton, NJ: Princeto n
Univers ity Pre ss .
Haythornthw aite , C. 2002. Building so cial ne tw orks via computer ne tw orks. In
Building Virtual Communities, ed s. K. A. Ren ninger and W . Shumar, pp. 159-90. New
York: Cambridge University P res s.
Haythornthw aite , C. 2008. Lea rning relations and ne tw orks in web -bas ed
communities. International Journal of Web Based Communities 4(2): 140-158.
Haythornthw aite , C. 2009. Crow ds and communities: Light and heavyw e ight mod els
of pe er production. Proceedings of the 42n d Hawaii Internationa l Confere nce on
Syste m Sciences . Los Alamitos, CA: IEEE Compute r Society.
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
10/11
Hillery, G.A. Jr. 1955. Definitions of community: Areas of agreemen t. Rural Sociology
20: 111-123.
Holohan, A., and A. Garg. 2005. Collab ora tion online: The example of distribute d
computing . Journal of Computer-Mediated Commu nication 10(4).
Horrigan, J.B., and L. Rainie. 2001. Online communities: Netw orks tha t nurture long-
distance relationships and local ties . Pew Internet and American Life Project.
http://www .pew Internet.org/pdfs /PIP_Co mmunities_Report.pdf (acces se d Ma y 20,
2010).
Hunt, F., and P. Johns on. 2002. On the Pa reto distribution o f Sourceforge projects .
ww w .ifm.eng.cam.ac.uk/people/fhh10 /Sourceforge /Sourceforge %20pape r.pdf
(accesse d May 20, 2010).
Ishii, K. and Og as ahara, M. 2007. Links betwee n re al and virtua l networks: A
compara tive stud y of online communities in Japa n a nd Korea . Cyberpsychology and
Behavior 10(2): 252-257.
Kim, A.J. 2000. Community Building on the Web. Berke ley, CA: Pe achpit Pre ss .
Koku, E., N. Nazer, and B. Wellman. 2001. Netting scholars: Online and offline.
American Behavioral Scientis t 44: 1752-1774.
Kollock, P. 1998. Design principles for online communities . PC Update 15: 58-60.
Kollock, P. 1999. The economies of online coopera tion: Gifts and public goods in
cybers pa ce. In Communities in Cyberspace, eds . M. A. Smith and P. Kollock, pp. 220-
39. New York: Routle dg e.
Kollock, P., and M. Smith. 1996. Mana ging the virtual common s: Co op eration and
conflict in compute r communities. In Computer-Mediated Communication: Linguistic,
Social, and Cross -Cultural Perspectives, ed. S. Herring, pp. 109-128. Amsterdam: John
Benja mins.
Komito, L. 1998. The net as a fora ging so ciety: Flexible communities. The Information
Society 14: 97-106.
Lev-On , A. 2008. The de mocratizing e ffects of se a rch engine use: On cha nce
exposures and o rganiza tion al hubs. In Web Search: Interdisciplinary Perspectives, eds.
A. Spink and M. Zimmer, pp. 135-150. New York: Springer.
Lev-On , A., an d R. Hardin. 2007. Internet-base d collaborations and their political
significance. Journal of Information Technology and Politics 4(2): 5-27.
Mate i, S. and S.J. Bell-Rokeach. 2002. Belonging in geo gra phic, ethnic, and Inte rnet
spaces. In The Internet in Everyday Life, eds. B. W ellman and C. Haythornthw a ite, pp.
404-427. Oxford, UK: Blackw ell.
Matz at, U. 2009. The e mbed de dnes s o f academic on line group s in offline so cial
networks: Reputa tion ga in as a s timulus for o nline discus sion pa rticipation?
International Sociology 24(1): 63-92.
Mele, C. 1999. Cybers pa ce a nd Disadvantaged Communities: The Inte rnet as a Too l
for Collective Action. In Communities in Cyberspace, eds. M.A. Smith and P. Kollock, pp.
290-310. New York: Routle dge.
Olso n, M. 1965. The Logic of Collective Action. Cambridge , MA: Harvard University
Press .
Ostrom, E. 1990. Governing the Commons. New York: Cambridge Univers ity Press .
Ostrom, E. 1999. Co ping with trag edies of the commons . Annual Review of Political
Science 2: 493-535.
Porter, C.E. 2004. Typology of virtua l communities: A multi-disciplinary foundation for
future res ea rch. Journal of Computer-Mediated Communication 10(1).
Preece , J. 2000. Online Communities: Designing Usability and Supporting Sociability.
New York: John W iley.
Rheing old, H. 2000. The Virtual Community: Homesteading on the Electronic Frontier.
Rev. ed. Cambridge, MA: MIT Pre ss.
Ridings, C.M. and D. Gefen. 2004. Virtual Community Attraction: Why People Hang out
Online . Journal of Computer-Mediated Commu nication 10.
Sasse nberg, K. 2002. Common Bond and Co mmon Identity Group s o n the Inte rne t:
Attachment and No rmative Behavior in On-Topic and Off-Topic Chats. Group
Dynamics: Theory, Research, and Practice 6: 27-37.
Schuler, D. 1996. New Community Networks: Wired for Change. New York: Addison-
We sley.
Shapiro, A. 1999. The Control Revolution. New York: Public Affairs.
Smith, M.A. 1992. Voices from the WELL: The logic of the virtual commons .
5/3/13
Lev-On
ci-journal.net/index.php/ciej/rt/printerFriendly/712/1020
11/11
http://rese arch.micros oft.com/~masmith/Voices from the We ll.doc (accessed May 20,
2010).
Sunstein, C.R. 2001. Republic.com. Prince ton, NJ: Princeton Univers ity Pre ss .
Taylor, M. 1987. The Possibility of Cooperation. New York: Cambridg e Un iversity Press.
Van W ende l de Jood e, R. 2004. Co ntinuity of the commons in ope n s ou rce
communities. In Proceed ings of SSCCII-2004.
Viéga s, F.B., M. Wattenbe rg and M.M. McKeon. 2007. The hidd en order of Wikipedia.
Human-Computer Interaction 15: 445-454.
We ber, S. 2004. The Success of Open Source. Cambridg e, MA: Harvard Univers ity
Press .
We llman, B. 2001. Physical place and cybe r place: The rise of pe rsona lize d
networking. International Journal of Urban and Regional Research 25: 227-252.
We llman, B., and M. Gulia. 1999. Virtual communities as communities . In Communities
in Cyberspace, eds. M. A. Smith and P. Kollock, pp. 167-194. New York: Rutledge .
We ng er, E. 1998. Communities of Practice: Learning, Meaning, and Identity. New York:
Cambridge Univers ity Pre ss .
