Approaches to Support Proximate Communities

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 
153 154 155 156 

We defined “community” in the Community and Online Community section

as a set of individuals with partially overlapping personal social networks,

which collectively provide a sense of belonging. Further, we noted that

“proximate communities” are communities built around individuals co-located

in a physical location. They can also be understood as communities where the

personal social network ties of members are associated with a particular

geographic area, place, or region. Proximate communities may be residential

communities or may result from a shared physical work environment such as a

university campus or central business district. In this section we will explore

how P3-Systems can be used to support interactions in proximate communi-

Table 2. P3-systems and key design approaches

* Systems approaches dependent on both synchronous and asynchronous use of data


Design Approaches

Synchronous Communication or

Synchronous Location Awareness

Asynchronous Communication or

Asynchronous Location Awareness




• Active Badge Find Command

• Ulocate current user location.

• IM buddies with Location Descriptors

• Location Labeled Messages

• Ulocate user location history.

• Hypothetical Group Calendar with detailed

Individual Location tracking data.

• Location Labeled Messages






• Cybiko chat

• Meme-Tags and Community Mirrors

• Proxy Lady

• LoveGety

• Hummingbird

• * Hocman motorcyclist data exchange

• * Social Net user co-location monitoring

• *RoamWare

• *FolkMusic

• * Hocman - web message board for post

proximity interactions

• * Social Net recommendations

• *RoamWare Desktop - distributes

information about collocated ad hoc


Use of


Spaces by


• ActiveCampus Explorer Buddies Map

• Active Badge – Look

• ActiveMap

• CMUSky

• Online Room Calendars

• *FolkMusic




in Matching



• Digital Cities And Community Network


• Map based Instant Messaging of Nearby


• Wi-Fi AP Chat

• Community Network and Digital Cities


• Geonotes

• E-Graffiti

• Active Campus Graffiti

• IBM-World Board

ties. We do this because we believe that P3-Systems are by their very nature

well suited to support proximate communities, since by definition they aim to

link people-to-people-to-geographic-places.

Wellman et al. (1988) note that communities can provide various social

functions such as sociability, support, a sense of belonging, and social identity.

However, P3-Systems do not provide social functions directly to users, as

these functions are generated by people or users of a system, rather than the

mediating technological infrastructure. For this reason, the P3-System design

approaches will be examined in terms of their ability to support people-topeople-

to-geographic-place interactions. The P3-Systems support such interactions

through the provision of user location awareness, making apparent

location-related interpersonal affinities, and the geographical contextualization

of public or private interactions.

Table 1 classifies existing P3-System design approaches into eight basic

categories. In this section we will first compare the significance of the division

into synchronous and asynchronous communication/awareness. This will be

followed by an examination of the two people-centric and two place-centric

design approaches in terms of their ability to support proximate community


The largest division of design approaches described in Table 2 is into Synchronous

Communication/Synchronous Location Awareness or Asynchronous

Communication/Asynchronous Location Awareness. An examination of the

systems described in the fourth section suggests that this division appears to

closely parallel our understanding of how the level of synchronicity impacts on

computer mediated communication in general. Synchronous communication

and synchronous location awareness data provided by P3-Systems is created

with the expectation that it will be processed in near real-time, whereas

asynchronous communication and asynchronous location awareness data is

produced with the expectation of unpredictable delays between data creation

and consumption. The result is that synchronous data is more useful for

management of user actions in real-time within the physical spaces of proximate

communities. This is because synchronous system design allow users to

address issues such as, “where is somebody,” “who is there,” and “who is

around me,” which can be expanded to questions such as “who is around to eat

or chat with right now.” In other words, synchronous P3-Systems are well

suited to the task of supporting ad hoc interactions grounded in geographic

place. Asynchronous system designs, on the other hand, allow users to make

long-term plans in relation to their use of space, by the provision of information

such as “how is this space used over time,” “who uses this space like me,” “what

do others thinks about this place,” etc. The synchronous and asynchronous

design approaches are complementary with the provision of one approach, not

negating the possibility of the provision of the other, and with each supporting

different social requirements.

The most important division of the design space provided by Table 2 for our

purposes is the four basic P3-System design approach types: (1) Absolute

User Location; (2) Co-location/Proximity; (3) People’s Use of Space; and (4)

Interactions in Matching Virtual Places. These four types will be examined and

compared in terms of how systems can be used to support proximate community

interactions. This will be achieved by an examination of the ability of various

design approaches to enable the following. First, proximate community

outeraction processes (processes outside of the direct information exchange

that enables people to enhance and coordinate information exchanges). That is,

those that relate to the management of people-to-people-to-geographicalplace

interactions. Secondly, proximate community interactions through either

personal or online community interaction spaces. These outeraction and

interaction process can be done publicly through online community spaces or

privately through online personal spaces. The provision or lack thereof of such

spaces in P3-Systems has a profound impact on the formation of social ties,

privacy management, and scaling.

P3-System Designs Based on Absolute User Location

Systems under this category, such as Active Badge and Ulocate, provide clear

support for proximate community outeraction processes. This is because they

provide users with information about the location of others, so that users are

better able to coordinate their interactions. Seeing a user’s history of

movements in space can also help coordinate interactions in a manner similar

to that of a conventional day planner. The location information also allows users

to contextualize their proximate community interactions. For example, a

student might ask another student to bring him coffee if he sees that remote

student near the cafeteria. Similarly knowing the location from which a person

sent a message will inform recipients as to the environment in which the message

was composed. Further, a community system that knows of people’s collective

habits, schedules, appointments, and location could use individual’s trajectories

to better manage face-to-face community interactions. For example, if an

individual was running late, the system could use awareness of his location,

calendar data, and analysis of his trajectory to, where appropriate, semiautomatically

advise others to delay the meeting.

P3-Systems based on absolute user location focus on location awareness

rather than interactive communication. Active Badge system provides a listing

of the current location of Active Badges that are available to the community of

users. This raises some design issues in regards to scaling. For example, if the

listing of the number of users expands greatly, then it becomes harder to gain

an instantaneous feel for the whereabouts of community members. Cartographic

visualization may be able to display larger groups in a slightly better

fashion, but they too have their limits. Further, such a design approach using

cartographic visualization of all community member locations would lean

towards making the system Place Centric rather than People Centric.

Of the four P3-System design approaches we review here, this is the most

problematic in terms of management of privacy. This is because, by definition,

the design approach discussed here focuses on the provision of personal

location data of individuals to others. Monitoring or tracking somebody’s

whereabouts can be extremely invasive. To address this issue the designers of

Active Badge limited the storage of location history information to a one-hour

period, and made that data available to users through a specific user history

command (Want et al., 1992). However, the lack of location data history does

not resolve the privacy issue associated with real time monitoring of the users’

locations. The other obvious mechanism to address privacy concerns is by

simply taking off the badge when one desires not to be monitored. This however

also fails to truly address the problem because of the likely social implications

for anybody deciding to opt out of wearing the badge even if it be for a brief

time. In fact, differing perspectives on privacy and personal freedom were

noted as reasons for successful and unsuccessful implementations of the Active

Badge in early field studies (Harper, 1992).

P3-System Designs Based on Co-Location/Proximity

The P3-Systems of this type systematically support opportunistic meetings

(Kraut et al., 1990) and informal communication through outeraction processes,

such as proximate location awareness, and the finding of affinities based

on location histories. In addition, co-location data can be used to support

ongoing community interactions through the use of online community spaces.

For example, “Social Net” could be expanded to include a related online

community space, which enables affinities to be inferred not only from pair-wise

patterns of co-location but also community location histories. Another example

is Hocman where proximate exchanges could be systematically uploaded to a

public website so that users could gain a community perspective and proximate

interactions could be used as a spring-board for community discourse. Of

course, the storing and utilization of people’s interaction histories could

potentially result in significant privacy problems. To address this problem the

designers of Social Net do not make the history of proximate interactions

stored on the device known to the user, however they acknowledged that this

resulted in users complaining that they were not sure why a social recommendation

was made. Hence, for the systems examined here, there is a clear

tradeoff between overall utility and the provision of potentially private personal

location data.