How to Define the Communication Situation: Context Measures in Present Mobile Telephony
Louise Barkhuus
react to中文The IT University of Copenhagen
Department of Design and Use of IT
Glentevej67,DK-2400Copenhagen,Denmark
barkhuus@it.edu
Abstract.Mobile telephony is rapidly becoming a wide spread mobile
technology.The technological possibilities for context-aware features are
increasing,although only few services exist in real consumer settings.In
an exploratory case study of high level mobile telephone users,the con-
text measures that are relevant to mobile telephony are studied.Start-
ing with three measures,location,time and identity,a fourth context
measure is also found,leading to the measure of social situation to be
important to users as well.The study concludes that the use of mobile
telephony is highly influenced by this limited set of context measures
but that these are not necessarily the same as the users wish to see
facilitated by technology.It also suggests that when developing context-
aware features for mobile communication,supporting these four pieces
of information should be prioritized.
Keywords
Context-aware applications,Human-computer interaction.
1Introduction
As mobile communication has become increasingly popular during the last decade, the request for more intelligent devices and services is apparent.From solely being used as voice communication tools,mobile phones have evolved into in-formation and data access terminals that support numerous tasks for the busy user.Examples of such functions are the wireless internet(WAP and i-mode), calender functions,e-mail services and text messaging.Although social behavior seems to adapt to the instant,out-of-context communication that mobile phones provide,suggestions for more aware and adaptable functions are suggested by [5]and[11]).These functions address the social problems arising when a person is available anywhere,anytime by letting the user define his/her own context to the possible communicators.
Currently visiting researcher at UC Berkeley,The Department of Engineering and Computer Science,working with Assistant Adjunct Professor Anind Dey.
2Louise Barkhuus
The technological possibilities today make context-aware services easier to implement and maintain.Intelligent features that assist the mobile device are relevant for numerous purposes,such as locating people,assessing other peo-ple’s situational context and adjusting the device itself to the cont
ext.Although researchers often consider consumer products as targets for context-awareness, few features have been implemented to real-life consumer products such as cell phones and PDAs.One exception is the‘friendfinder’implemented in Sweden by Telia,where mobile phone users can sign up for the service and then observe where their friends are located real-time.The friends,naturally,have to sign up as well[13].
Contextual information can be considered as the‘clues’that make up an overall situation in which the use of the mobile device takes place.Being part of the overall context,they are not the definition of context,because that is affected by the individual’s interpretation as well.Many researchers of context-awareness have questioned how a contextual situation can in fact be defined.The ones who focus on the actual technologies,limit their definition to the context measures that are utilized in their [2]and[?]),for example position,time and historical preferences.Others claim that a context is impossible to describe in technical terms or even in words[4].The approach to studying context in this paper then facilitates the request for“studying the expected context-of-use carefully[...]to determine what contexts are mostly emergent,contingent,and improvisatory”[4].
What this paper propose is to trace context measures by conducting an ex-ploratory case study,focusing on one area of mobile computing.It aims tofind the context information that mobile devic
e users rely on when communicating and using the device.As an example of mobile communication,mobile telephony is chosen,because of its wide spread among the population as well as its famil-iarity of use among most users.Had the study taken place just a decade ago, the users would not have been as comfortable with the type of communication as they are today.
2Context-Aware Mobile Communication
Originally considered to be a subset of ubiquitous computing,context-aware computing is increasingly viewed as essential to a truly ubiquitous computing en-vironment[6].Context-aware computing is described as“an application’s ability to detect and react to environment variables autonomously”[1].Where context-aware computing includes a large variety of applications and systems,context-aware communication limits the scope to applications that facilitate communica-tion,such as messaging and virtual whiteboard applications[9].This dimension of context-awareness that includes stationary as well as mobile applications is highly researched with the purpose of making communication morefluent and efficient.From the Olivetti Active Badge system in the early nineties[14]to the context-aware mailing list system[3],the approaches to aiding communication with situational information have been numerous.However,limiting the scope to
Context Measures in Present Mobile Telephony3 mobile communication,the research decreases to features that rely on the user’s own definition of context.Moreover the features are developed on the basis of state-of-the-art technologies and what is considered useful by the researchers themselves.Thorough user studies have yet to be made in order to clarify possi-ble other needs and present habits of communication,especially among consumer target groups1.
2.1Context as Real Life Clues
People rely on‘clues’in their everyday life in order to act according to a certain situations;they depend on personal observations in order to conclude on the situation’s higher level.One example is the person who sees a coworker’s car in the parking lot and uses this clue to infer that the coworker is in the office at the moment.These clues can be compared to the context measures that context-aware applications use to adapt according to the situation or environment.By tracing informational clues that people rely on in specific situations,the study of present context measures can add to the usability and success of future context-aware communication applications.
Although context measures are relevant for the development of context-aware applications,the notion of context is highly complex and when reviewing past literature,it shows that many researchers have at
tempted to define not only context itself,but also how situations can be perceived by different individuals. Examples of the latter include Greenberg,who claims that it is not possible to define a context in terms of sensor information because situationsfluctuate and are unpredictable in nature[4].The overall situation that a person is part of, is not solely made up of measurable context information,because the person’s own perception of the situation is part of the context.What is possible,how-ever,is to facilitate technology with functions that take certain aspects of the situation into account and thereby make the use more smooth and transparent for the user.Where early studies define context solely as context measures like for example location and the user’s task[8],more resent research realize that sensor information is only part of the definition of a certain context[3].
The actual context measures that(at least partly)create the user’s perceived situation,however,are rarely empirically explored,making most discussions ei-ther technical or theoretical.The study of present context measures is likely to provide insight into which measures are relevant to include in new technologies or applications.Moreover,it is apparent that context measures differ for different types of applications;mobile computing is highly influenced by location,where stationary applications depend more on present task.Keeping this in mind,the context information that users of mobile devices rely on can be traced as part of the relevant contribution to the context-awareness in future mobile tec
hnologies. 1Some studies have been made on the habits of mobile phone [7],however, they mostly look at social behavior in connection with the mobile communication and not at the possible need for future functions in relation to communication habits.
4Louise Barkhuus
2.2Mobile Telephony as an Example of Mobile Communication
As an example of mobile computing devices,the mobile telephone is at present time the most common mobile communication tool and therefore a good example to explore[7].Since it is an ultra-mobile tool,meaning mobile enough to walk with at the same time as using[10],its use is likely to depend on context;opposite non-mobile technologies,such as a stationary computer placed in an office,the situation changes more rapidly with a mobile phone.
The research area of mobile telephony is presently explored in relation to use and the studies are focusing on the impact that this innovated type of commu-nication has on the users(phones have been around for over a century,mobile phones for not even two decades).Examples of such studies include[5]and[7]. The features of mobile phones that are proposed and developed are often extra services relating to newly employed technologies.One example is the feature of WAP that enables nu
merous services;some intelligent services,for example messenger services,have been employed and tested,however,they are not in use among the public[12].
3Case Study Design and Method
The present study is designed as an exploratory empirical study of context mea-sures within the use of mobile phones.By interviewing high level users of mobile telephony,the users’contextual clues and habits of use are traced.The study consists of seven semi-structured qualitative interviews carried out in the partic-ipants’own environment(office,own home,school cafeteria etc.).The interview guide wasfirst tested on two participants(a miniature pilot study)and thereafter revised to betterfit the relevant context measures and to phrase the questions better.It is structured into six parts with questions regarding demographics, three measures of context,general use of their mobile phone and a presented scenario.In the end a very open question is asked,leaving the participant with the option to elaborate on subjects not discussed during the regular questions.
To limit the scope of the interviews,three context measures have been se-lected as prime focus.The three measures that are initially investigated are location,time2and identity of person communicated with.They are chosen because they are believed to be major players in a mobile telephony situation; b
y proposing these measures,the interviews become more focused than if very open questions are asked.
3.1Participants
The participants were selected from one main criteria:they had to be high level users of mobile telephony.This also made a casual restriction to age,since the 2The study distinguishes between time and relative time,that being the user’s actual time and the user or receiver’s perceived “do you have time to talk?”).
Context Measures in Present Mobile Telephony5
heavy users are easier to be found in young age groups,teenagers as well as
adults under30.The age therefore ranges between19and27,with an average
age of23.Their experience with mobile phones ranges from3years to7years
with an average of5.8years and most of the participants were using their third
phone at the time of their interview.
The participants’occupation vary from unemployment and university stu-
dent to professional consultant and military service.Their educational level range
likewise from ten years of education3to a Masters Degree in information tech-
nology.Although the variety in background is quite strong,these differences
are not considered in the study,since the emphasis is on the use of the mobile
phone in general and such consideration would have required a higher number
of participants.
Participant Age and gender Occupation Had a mobile phone for p121,female Service assistant at convenience store7years
p221,female Retail assistant at music store3years
p325,male2nd year graduate student7years
p426,female2nd year graduate student61/2years
p521,male Military service6years
p627,female Freelance consultant7years
p719,male Unemployed4years
Table1:Participant description.
3.2Collected Data
The data was collected as a digital audio recording and transcribed shortly after.
The transcription was done in several ways,a word-by-word transcription and a comprised version describing the essence of each answer.This way,comparing the
different answers is easier.Finally,groups of context measures were suggested
and the parts of the interviews that related to a certain measure(sometimes
more at the same time),were grouped together in order to compare to which
degree the measures were noted by the participants.
3.3Analysis of the Study
The data of the study was analyzed according to guidelines provided by Yin[15].
Because it is exploratory in nature and the limited number of participants,none
of the answers were quantitatively measured;the interviewees’expressions re-
garding the use and context were merely compared and analyzed for contextual
3In Denmark,high school is not obligatory and the earliest possible to leave school is
after nine years.
6Louise Barkhuus
measures.The transcriptions were analyzed for general patterns of the use of mobile telephones,in order to trace the situational context where the communi-cation takes place.
Since the study is limited to a small number of participants and is qualitative in nature,nofindings can be said to suffice as proof of factors of the particular use.The analysis of data simply provides indicatio
n of specific context measures in the use of mobil telephony.Thefindings,however,also give an insight into the social structures of mobility in an extremely mobile telephony satiated en-vironment.The study therefore also relates to overall social and communicative practices of today’s mobile device users.
Even though the questions of the study was not limited to person-to-person communication,the fact that mobile phones at present state are mainly used for this kind of communication,limits the results from for example information retrieval,which is more common in for example Japan.Context measures in relation to information search and other tasks that are not in play at present state are therefore not studied.
4Overall Findings
It is immediately evident that most of the mobile telephony is influenced highly by the situation,an observation that has been made in other studies as well[11]. Several measures and pieces of context information of both communicators are found to matter to the general use.Thefindings support the notion that people rely on clues in their decision of how to communicate through their mobile phone and that these clues are highly individual and complex.The overallfindings have been divided into three separate but related conclusions.
4.1Mobile telephony is highly dependent on situational context
measures
All participants phrase,at one point or another,that their actions in many cases depend on the situation.This non-ambiguous reply supports the hypothesis that users are constantly aware of their own situation,as well as the situation of the person they are communicating with,and act accordingly.One example of dependent use is a participant who,without hesitation,says that he greets virtually every caller differently except of course when the display shows no number or caller[participant7,question3b].Almost half of the participants have defined different ring tones for different people.That way they know if they should“run to the phone or can just relax”[participant5,question3c].
This dependency is sometimes even made into a‘necessity’for the partic-ipants;several of them actually call their friends just to infer of the friends’situation,for example where they are and what they are doing.The three par-ticipants who do this all state that it would be efficient with technology facili-tated features that could give this kind of information.“I often meet with my friends Saturday afternoon,but it is always a casual thing,so I call them up to

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。