THPSiS

.L
31003

5440753:

LIBRARY

Michigan State a

University

This is to certify that the
thesis entitled

TEMPLES OF TAIPEI
A HANDHELD TOUR GUIDE FOR POCKET PC

presented by

CHIA—YU TAI

has been accepted towards fulfillment
of the requirements for the

Master of degree in Department of
Arts Telecommunication

 

 

WW@

 

Major Professor’s Signature

5/6/03

Date

 

MSU is an Affirmative ActiorVEqual Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:iCIRC/DateDuo.p65-p.15

 

 

TEMPLES OF TAIPEI
A HANDHELD TOUR GUIDE FOR POCKET PC
By

Chia-Yu Tai

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

MASTER OF ARTS

Department of Telecommunication

2003

ABSTRACT

TEMPLES OF TAIPEI
A HANDHELD TOUR GUIDE FOR POCKET PC
By

Chia-Yu Tai

"Combining open standards-based applications (XML, Java, Macromedia Flash)
with everyday tasks, pervasive computing removes the current complexity of new
technologies and enables us to be more efficient in an tasks." writes Phillip M.

Torrone in his article: The Rise of Pervasive Computing.

This production thesis project focuses on using Macromedia Flash as the
authoring tool to turn the idea of pervasive computing into a real-world application.
This tour guide on a handheld computer (Pocket PC) provides users with interactive
maps, immersive QTVR-like panorama movies using Flash. And it also provides a
step-by-step walking tour with visual navigational aids and panoramic views and
short Windows Media video clips of interesting temple activities. Furthermore, there
is an English-Chinese “phrase book”. These useful phrases with Chinese
pronunciation are intended to help English-speaking tourists communicate with
people in Taiwan. Users’ experiences of the tour guide in the field are analyzed.

Finally, the future of handheld tour guides is explored.

ACKNOWLEDGEMENTS

It’s 2003. I sit down and start thinking about what to write in this acknowledgement,
a thought suddenly pops up in my head; then I spend some time trying to find an Old
magazine in my bookshelf. I finally find what I am looking for. It’s the New Media
magazine (already out of business now), February edition of 1994. There’s an article in
this issue called: Learning New Media. In page 48, there’s a resource guide listed some
colleges and universities which provide multimedia-related courses or degrees. Michigan
State University/D. of Telecommunications is one of them. There is a brief description
about MSU/T C: ofl‘ers course in hypermedia, has a new multimedia lab under

construction.

It was 1993. I just finished my BA in Advertising and started working as an assistant
producer for a TV commercial company. During that time, I started reading some new
ideas about new media, convergence, interactive TV in some magazines. And I thought:
this could be the area I would like to devote to for my whole life. So I kept looking for
learning resources: books, CD-ROMs, and magazines. But I still could not get the whole
picture. And then I read the article in the magazine: New Media. It was the first time I
knew there were some graduate schools in the US providing new media-related master
degrees. After some research, I found the MSU/T C was the only feasible choice for me. (I

only applied for MSU.)

In the fall of 1995, I came to the US for the first time in my life. In 1997, I went

iii

back to Taiwan. And then the intemet business started booming in Taiwan. I stopped

working on my thesis and joined the “roller coaster ride” of dotcom business.

I always consider myself as an idealist and “new media” or “the Internet” as a digital
revolution. “Talk about the revolution, well, you know, we all like to change the world,”
say the Beatles. I used to believe that the intemet is going to create a new world in “seven
days”. So when the dotcom business crashed I felt like it was the end of the world. And
then I realize that the digital revolution might come to an end, but the digital evolution

continues. . .

I would like to take this opportunity to thank Dr. Carrie Heeter for the foundation of
Digital Media Arts program several years ago. Without her effort and the Digital Media
Arts program, a guy from the Far East like me would not attend a university in the
Midwest. And I also want to thank her for her encouragement and wholehearted support
throughout the project. Gratitude is also extended to Professor Brian Winn for his

kindness and the inspiration in class. And finally, I would like to thank my parents for

everything they give me.

iv

TABLE OF CONTENTS

LIST OF FIGURES ........................................................................................................... vii

CHAPTER 1

INTRODUCTION ............................................................................................................... 1
Project Objectives ........................................................................................................ 1

CHAPTER 2

DESIGNING APPLICATIONS FOR POCKET PC IN FLASH .................................... 3
Fonts ............................................................................................................................. 3
Images .......................................................................................................................... 5
Interactivity/Navigation. . . . . . . . . ................................................................................... 5
Pixel Dimensions for a Flash Movie in Flash Player ................................................... 5

CHAPTER 3

PRODUCTIONPROCESS............... ............................................................................. 8
Photography ................................................................................................................. 8
Video .......................................................................................................................... 11
Audio........ ............................................................................................................... 14

CHAPTER 4

RESEARCH METHOD ..................................................................................................... 16

CHAPTER 5

RESEARCH RESULTS ..................................................................................................... 18

Observation Results ................................................................................................... 18
Questionnaire Results ................................................................................................ 20
Discussion .................................................................................................................. 22
CHAPTER 6
The Future of Tour Guides on Pocket PC .......................................................................... 26
Location-based Advertising and the Walking Tour ................................................... 26
The Future of Phrase Book ........................................................................................ 27

vi

LIST OF FIGURES

Images in this thesis are presented in color.

Figure 2.1 Screen shot of text property panel in Flash ....................................................... 4
Figure 2.2 Enlarged screen shot of blurred text .................................................................. 4
Figure 2.3 Pixel Dimensions for a Flash Movie in Flash Player ........................................ 6
Figure 3.1 Example of a long telephoto shot ...................................................................... 9
Figure 3.2 Bigger version of a festive event picture ......................................................... 10
Figure 3.3 Cropped version of a festive event picture ...................................................... 11
Figure 3.4 Screen shot of creating new audience in Windows Media Encoder ................ 13
Figure 3.5 Screen shot of common stream settings in Windows Media Encoder ............ 14
Figure 3. 6 Screen shot ofthe Phrase Book... 15
Figure 5.1 Map ofTamshui walking tour.. .. 23
Figure 5.2 Non-linear self-guided tour and linear guided tour.. 23
Figure 5.3 Buttons on the picture which brings up more information .............................. 25
Figure 6.1 Screenshot of Babel Fish translation service ................................................... 28

vfi

Chapter 1

Introduction

Pocket PC has gradually become an important player in the PDA (Personal Digital
Assistant) or mobile device world. And in the PC world, Flash has emerged as a de-facto
development tool for web-based multimedia. This production thesis tries to combine both

to create a handheld tour guide of the Temples of Taipei.

In the beginning, my plan was to create a tour guide for the city of Taipei. I realized
that the project scope was unmanageable for one author. It also was too broad to appeal to
one audience. Afier discussions with some American friends, the topic: Temples of Taipei
was chosen because temples are popular scenic spots for most tourists in Taipei. It was
appropriate and manageable for a one-person team. And it let me carefully define a target

user and design the product to meet their needs.

Project Objectives

The objective is to design a handheld tour guide which helps English-speaking
vacationing tourists aged 18 to 40 to travel on their own and makes their tour in Taipei’s
temple easier and more informative. Many special events at the temples occur only during
one day or one week of the year. Without technology only those tourists who happen to be
there at the right time experience those events. There are mysteries and histories,
explanation and interesting facts about architecture, artifacts, history, and culture most

tourists never learn because they lack a personal guide to point things out. This handheld

l

tour guide is designed to be played on Pocket PC, but a computer-based simulation is also
available online. The design challenge also allows the opportunity to explore the
possibility of multimedia on mobile devices such as Pocket PC. The much smaller screen
size, limited RAM and hard drive space, relatively poor speaker quality, and lack of a
physical keyboard present different design constraints than creating for a desktop PC.
Stylus-based computing as well as the mobility of the device are unique and positive yet

unexplored design possibilities.

Chapter 2

Designing Applications for Pocket PC in Flash

Designing for a Pocket PC involves much more severe constraints in terms of screen
size, RAM, processor speed, screen display quality, and file size. Pen versus mouse
interactivity also impacts design. This chapter discusses some of the important design

decisions to accommodate the Pocket PC format.

Fonts

 

Because of limitation of small screen of Flash on Pocket PC, it is necessary to use
smaller font sizes such as 10 point type in order to squeeze more textual content into the
tiny screen. Regular fonts at smaller font sizes such as 10 or 12 points look blurred on
Pocket PC and present serious legibility issues. Anti-aliasing will smooth the edges of
fonts and only works fine for text at larger font sizes such as 16 or 24 points in Flash on
Pocket PC. In order to solve the legibility problem, I use “pixel font” which is designed to

fit precisely to the Pocket PC’s screen's pixel grid.

When using pixel font in Flash for Pocket PC, all text should be at exact X, Y
positions on the screen or the text will get blurred. To be readable, Flash movies can only
be viewed at 100%. This is not currently an issue for Flash on Pocket PC because the Flash
player only shows Flash movies at 100% on Pocket PC. The font size should appear at the
designed size or multiple of it (for example, 10 points for Tenacity Condensed which I used
for the Temples of Taipei project). The text alignment should be set to left and all tracking

and kerning should be set to zero. Below is the screen shot of text property panel in Flash

3

 

which is correctly set.

 

 

 

 

r JEAhmmmmm LHTEH1,EIIEUEQHEE

'A‘y [El—:I‘ a! anrmd v] meam :‘ij Format... 5
45.?" .j l7 ' ' ' will]! 55": :- mama. _ '
Tglzao in [ Tanetl _V‘

 

Fig 2.1 Screen shot of text property panel in Flash

When creating the text label under each button for the main page of Temples of Taipei,
I initially used middle alignment for the text label of walking tour since it is more
aesthetically pleasing. But the word “tour” became blurred because the pixel font only
works when text alignment set to lefi. (See Fig 2.2) Afier some trial-and-error procedures, I
found the solution is to use separate text box for each word and set the text alignment of
each word to left. And then adjust each text box to the correct position which makes the
text look crisp and aesthetically pleasing. The text label of “photo viewer” in Fig 2.2 is an

example of the result of this technique.

P h CI 1’. III F' .3 Fl III r' a m a

'-.-' i e I.-'-.I e r“

 

 

Fig 2.2 Enlarged screen shot of blurred text

Images

It is much easier for Pocket PC to display a bitmap image than a complicated vector
image because the rendering of complicated vector image needs more processing power.
Performance issue is crucial for developing Flash for Pocket PC because Pocket PC has
slower processor and less memory than desktop PC. So I design the interface based on
bitmap images, and only used vector image in static text (when exported as a SWF, static
text is converted to a vector image) and simple interface elements such as the arrow buttons

and the red navigational bar.

Because the quality of Pocket PC’s screen is not as good as displays of notebooks and
desktops, it is better to increase images’ color saturation and contrast in Photoshop before
importing them into Flash so the images are more vibrant on the small and low-quality

screen of Pocket PC.

Interactivity/Navigation

Mobile devices such as Pocket PC use stylus-based input, no actual cursor exists on
screen. This means there is no rollover or rollout effect for buttons. On the small screen, it
is challenging to design buttons that explain themselves because there is not much space
and no rollover effect for a “button label”. If the designer creates many large buttons, there
is not much space left for the content. If buttons are too small, users might have trouble
tapping or recognizing them. So it is important to balance between size and legibility of

button and other media elements when it comes to interface design for Pocket PC.

Pixel Dimensions for a Flash Movie in Flash Player

The screen size of Pocket PC is 240 by 320 pixels. Unless you use the Standalone
Macromedia Flash Player 6 for Pocket PC which supports full screen playback outside of
the browser, you can not take advantage of the entire screen. The resolution of Pocket
Internet Explorer is 240 by 320 pixels. The caption bar on the top and the menu bar on the
bottom take 26 pixels each off the vertical resolution. These two areas can not used for

content, as shown in Fig. 2.3.

 

/ . ’. . . .I
. ‘fv'llnternfl Explorer 4: 5:25 0! . .i ‘ Height: 26 Pixels

Height: 263 Pixels Height: 320 Pixels

 

Height: 5 Pixels

  

 

  

l

.- ———I—
View Tools 0 Q a Q ”a HI I V Height: 26 Pixels ,

 

 

 

, 3 1305335964 run. uwa‘

 

Fig 2.3 Pixel Dimensions for a Flash Movie in Flash Player
When designing content for a Pocket PC that fits on one HTML page, the content of
that HTML page must not exceed 240 by 268 pixels or the vertical scroll bar of Pocket

Internet Explorer will appear.

When it comes to HTML pages embedded with Flash movie files, it is a different story.
For some reasons, the pixel dimensions of that embedded Flash movie file have to be 240
by 263 pixels or the annoying scroll bar will show up. The “missing five pixels” are right

between the menu bar and the embedded Flash movie.

 

Chapter 3
Production Process
Photography
The designated target audience is the so-called MTV generation. Tire MTV generation
consists of Generation X (born 1961 — 1981) and Generation Y (born after 1982). They
were raised on fast-paced TV program, MTV and video games and they are the most likely

users of handheld devices.

Kathleen Hessert suggests keeping content short and using audio-visual aids to
increase both understanding and comprehension when communicating the MTV

generation since they have shorter attention spans and demand concise, dynamic and

high-impact idea packaging.l

It makes sense to keep content short since it is not easy to squeeze a lot of content into

Pocket PC’s small screen. Anna Marie Pises suggestszz

Be succinct. Think Hemingway. Succinct and directly focused copy is vital, not only
because of limited screen size, but because of limited time and attention spans of the
mobile lifestyle and customer. The average line of copy on a Pocket PC, for example,

is approximately 6 to 11 words in length.

 

' Kathleen Hessert, Getting Through to the ”MTV" Generation,
http://www.bbhighway.com/Falk/Interactive_Coach/CommCorner/getting_flmugh_mw_generafion.asp
2 Anna Marie Pises, Editorial Style: Writing and Editing for Wireless Devices, Flash Design for Mobile
Devices.

I also think pictures (visual aids) are important because the target audience (MTV
generation) loves to watch instead of read. Furthermore there is an old Chinese saying: a
picture worth a thousand words. So I took more than 5,000 pictures using a digital camera
in order to get more quality shots. I feel blessed to be able to shoot pictures using a digital
camera since it saves me huge amount of time and money on film processing compared to
traditional cameras. Digital photography allows me to shoot many more photos than I

actually use, letting me choose the very best to include.

In order to compress several objects into a small and tight visual composition, I used a
long telephoto lens of around 300 mm. You will see an example below which I shot using a
long telephoto lens because I wanted to compress the dragon on the roof , the pillar in the
foreground, and the tower and mountains in the background into the small screen of Flash

on Pocket PC (240 by 263 pixels).

fjg'lnternet Explorer 4' 637" 0

    

maroon-Eases) 5..

Fig. 3.1 Example of a long telephoto shot

 

 

 

 

 

And I also found that cropping a bigger picture into a small one worked well. Fig. 3.2
is the original and bigger version picture of a noisy festive event. If I simply resized it to
240 by 263, users could not understand it well because of loss of details. So I decided to
crop the picture and Fig 3.3 is the final picture which visually conveys the message: the

festive event is deafeningly noisy.

 

Fig 3.2 Bigger version of a festive event picture

 

 

'Z— Inlmlwl Explorer 4% 0:26 0

f

‘m

    

c .)' ::'

vie-rook oaaea file

Fig 3.3 Cropped version of a festive event picture

Video

The Temples in Taipei could become very active during a festival, so I decided it
might be a good idea to incorporate some short video clips of interesting activities or
ceremonies into the tour guide. There are two ways to integrate video into Flash
applications for Pocket PC. One is launching Windows Media Player for Pocket PC file

from Flash by linking to a video file. The other is to play video inside the Flash application.

Windows Media Player is a built-in application on almost all Pocket 2002 devices.
Windows Media does a really good job on video compression. Afier some trial-and-error
procedures, I managed to encode the video clips at 15 fiames per second with video size of
208 by 160 pixels. The encoded video clips play smoothly on most Pocket 2002 devices

and look great in terms of video quality. The file size of a two-minute video clip is about

35th. In the final product, I included five video clips. The total file size of these five
video clips is 15MB, which makes up 75% of the size of the entire application which is just
under 20MB including video, graphics, interface, and sound. The current generations of
Pocket PCs have around 16MB of internal storage, most of which is taken up by programs
that ship with the Pocket PC or other programs installed by users. Therefore running
“Temples of Taipei” requires an add-on memory card to run on Pocket PCs. 20MB seemed

like a reasonable compromise between small file size and interesting content.

Before Macromedia released Flash Player 6 for Pocket PC in April of 2003, only
Flash Player 5 was available on Pocket PC. In order to play video inside the Flash
application using Flash Player 5, I could only import a sequence of bitmap images into a
series of key frames to simulate video. Using this technique results in file size problems.
Since Flash 5 does not provide any in-fi'ame compression, the imported bitmap images
make the exported Flash file terribly big. Now the Flash Player 6 for Pocket PC is available
and it provides some in-frame video compression using the built-in Quicanne Sorenson
Spark codec. But there are still some performance issues on Pocket PC if the flame size
inside the Flash application is bigger than 160 by 120 pixels. Furthermore, the file size of
the simulated “video-like” Flash file is also bigger than the same video clips encoded by

Windows Media.

Using Windows Media to deliver video clips does not have the level of control that
Flash has. Besides, users have to switch back and forth between Pocket Internet Explorer
and Windows Media Player. But I still decide to use Windows Media because it provides

bigger frame size (208 by 160 pixels vs. 160 by 120 pixels) and smoother playback on

12

Pocket PC.

I downloaded the Windows Media Encoder for free from Microsoft. There are two
versions available: Windows Media Encoder 7.1 and Windows Media Encoder 9.0. I used
Windows Media Encoder 7.1 instead of the newer Windows Media Encoder 9.0 because
the 9.0 is quite unstable when encoding video. Besides, it was not surprising for me to find
out that Windows Media Encoder does not import video files fiom the QuickTime format. Ir
Actually, Windows Media Encoder is quite picky about the codec and format of imported “ls..-
video files. For example, it does not support AVI files using Microsoft DV codec, either. So
I exported the edited video as AVI at 320 by 240 pixels in Cinepak format fi'om Adobe

Premiere. Then I imported this video into Windows Media Encoder for encoding.

When encoding video for Pocket PC, I started by creating a “New Audience” with a

maximum bit rate of 250 kilobits per second (see Figure 3.3.).

13

 

 

m Enter the name and maximum bit rate for the audience.

 

audience name: lPocketPc_250K Ll

Maximum bit rate: IZEDKI bps

Note: Type K or M to specify kilobits ct
megabits, respectively.

Cancel . 1

Fig 3. 4 Screen shot of creating new audience in Windows Media Encoder

 

 

 

And then I chose the setting for audio using Windows Media Audio 9 codec and 22
kilobits per seconds with 22 khz sampling rate in stereo. For the video, I used Windows

Media Video V7 codec with video size of 208 by 160 pixels.

 

,P' ‘\

"'T—‘r

 

Choose the settings that are common for all streams generated by this profile.
The audo and script bit rates affect the remaining bit rate available to the

 

 

 

 

 

 

 

 

 

video streams.
{Audio—m .,___ ?
l _C_odec thows Media Audio 9 L] .
i _Ecrmat: 2215:3322 kHz. stereo M 1533i?!” 3;]

 

 

 

codec: I'Windcws Media Video w
Eideo size: Izos 3160

 

___LLIL.‘

advanced...

 

___--_.__._.._.-_-_....... . _.

 

_. .....-_.. ...._.. ... ......—...—....—......—.-.,..--

ax—eerl lr—e@>] see 1 are I are |

 

 

 

 

 

 

 

Fig 3.5 Screen shot of common stream settings in Windows Media Encoder

Audio

In Temples of Taipei there is a phrase book intended to help English-speaking tourists
communicate with people in Taiwan. There are six categories of phrases for different
situations, including basic, emergency, food, hotel, shopping, and transportation. There are

about ten useful sentences in each category and users can click on the “play” button of each

15

 

 

phrase to play the Chinese pronunciation. If they are near the person with the handheld,
people in Taiwan could hear the pronunciation through the speaker of Pocket PC. I also put
the exact sentence in Chinese characters on the right in case the Chinese pronunciation
could not be heard or understood clearly because of loud environmental noise levels or the
hardware limitations of speakers on Pocket PC. Furthermore, users who are learning
Chinese could also use this phrase book to learn some Chinese pronunciation and

characters.

D E:-:|:u:e me. I'm lair
D Unlher e am l new-'7‘

D ls‘alhere': the MRT

:tatlclrfl'

D 'v'r‘here': the cadre.“ f; C‘

pilllce station”

D flag | UK: from
Flfll:|l'l‘-.“"

 

Fig 3. 6 Screen shot of the Phrase Book
In order to achieve a good balance between quality, file size and performance on Pocket PC,
I use MP3 compression in Mono, 32kpr, and Fast compression for these Chinese
pronunciation audio files. Most Pocket PCs only have one speaker, so it’s quite logical to
use compression in Mono. I decide to use 32kpr and Fast compression based on
information I get from the Internet and some trail-and-error procedures.

16

Chapter 4

Research Method

Two adult English speaking visitors to Taipei interested in sightseeing at the temples
of Taipei but have not visited the temples yet, who know how to use a Pocket PC, were
recruited to use the tour guide to tour one of the temples. Because this is a small usability
study, random selection is not important. What matters is to reach potential subjects who fit
the three criteria of the target audience (English speaking, interested in but have not visited
the temples, already know how to use Pocket PC). Friends and family of the researcher
were asked whether they know any potential subjects. A list of six names was compiled and
letters of invitation written. The referring person was given the recruiting letter to give to
the potential subject. Subjects interested in participating were asked to email researcher

Chia-Yu Tai for more information.

Subjects could choose to take the tour alone or with an English-speaking fiiend. The
subjects met the researcher at the temple where they were given a Pocket PC containing the
interactive tour and shown how to use it. The self-guided tour could last between 3 and 5
hours depending on how quickly the subjects walk and how much time they spend
exploring. The researcher accompanied the subjects (one at a time) observing how ofien,
where, and for what they use the guide. At the end of the tour (or whenever the subject
chooses to stop the tour) subjects were invited to complete a seven item questionnaire

about their reactions to the handheld guide.

The first two potential subjects who responded to the recruiting email were selected to

17

participate in the study. Both subjects who volunteered were male. James, a senior of a
university in the US, was in his early 203 and Michael, an immigrant to America from
Taiwan, was in his early thirties. Michael owned a Palm Pilot. James did not own one, but
was familiar with how they worked. Each was an American living in Taiwan. James was
working for a language school and taught children in Taiwan English. Michael was
working for an international company in Taiwan. Each was a native English speaker. James
could understand very few phrases in Chinese. Michael was fairly proficient at speaking
Chinese. James elected to take the interactive tour alone, and Michael took the tour with

his wife and child.

18

Chapter 5

Research Results

Wimmer and Dominick suggest3 :
The major advantage of field experiments is their external validity: since study
conditions closely resemble natural settings, subjects usually provide a truer picture of

their normal behavior and are not influenced by the experimental situation. (p.98)

This chapter presents the results fiom the observation of the walking tour and fiom the
questionnaires filled out by two subjects afier they finished the walking tour. Since the
research was conducted in a natural setting, the results provide a truer picture of users’

normal behavior than a laboratory study would provide.

Observation Results

The self-guided tour could last between 3 and 5 hours depending on how quickly the
subjects walked and how much time they spent exploring. It took James about 2 hours and
50 minutes to finish the tour. Michael spent about 4 hours and 10 minutes taking the tour.
The researcher accompanied the subjects (one at a time) observing how often, where, and
for what they use the guide. James was eager to get started seeing the temple, so he spent
only a few minutes exploring different content areas in the Temples of Taipei before
starting the tour. He opened the first page of the tour, and then looked around the physical
world to get oriented. While he was looking up and back between the handheld and the real

world, the power to the screen dimmed on the handheld to conserve battery. The researcher

 

3 Roger D. Wimmer and Joseph R. Dominick, Mass Media Research, Third Edition.

19

had to explain to him that he needed to tap the screen anywhere to bring back the brightness.

The screen dimmed frequently while both subjects were trying to use it.

James figured out where the arrow was pointing and started walking, holding the
handheld in one hand and the stylus in the other. After walking about 50 feet, he looked
back at the handheld device and advanced to the next page of the tour. He looked around
the real world, did not recognize where he was, and decided to keep walking in the

direction of the arrow, checking the handheld every 30-60 seconds.

When James saw familiar landmarks that matched the next photo in the tour, he
paused to look around and appreciate the sights around him for about one minute, then
looked back at the handheld and continued to follow its directions. The handheld tour
seemed to strongly control James’ exploration behavior. Only once in the tour did he
deviate momentarily from the path to go look in a room that was not part of the tour. He
spent quite a lot of time checking between the handheld and the real world to verify where

he was. He almost always read the text at each new location.

Michael struggled more than James did with how to use the handheld. He was
frustrated when the power dimmed, sometimes clicking by mistake on a button and
navigating to somewhere else in the tour when he didn’t intend to do so. At one point
Michael notices a link from a location to video. After watching the video, Michael was

stuck and didn’t know how to resume the tour without assistance fiom the researcher.

Although James never got lost, Michael made one incorrect turn during the tour, and

20

in both cases eventually asked for help to get back on the path.

The handheld tour clearly takes a lot of users’ attention away from looking at the
temple itself. On the positive side, the handheld tour gives users more background
information and details than are available in the un-supplemented real world. As Michael
and James got more used to the tour, they actually used it less. They became more
confident in recognizing each hot spot, and felt less need to check so often to be sure they

were on the right track.

Furthermore, Michael is not very comfortable with carrying a somewhat bulky Pocket

PC while walking. He sometimes asked the researcher to hold the Pocket PC when he

needed to free up his hands for drink and snack food.

Questionnaire Results

Responses to the survey questions will be reported question by question.

1. Which parts of this tour guide program do you like most? And why?

James’ favorite part of the tour guide was panorama, because he has never seen

anything like it before. Michael liked the walking tour because the tour did introduce him

to many interesting sites.

2. Which parts of this tour guide program do you dislike most? And why?

21

James most disliked the limited option of walking tour. Michael most disliked the

length of walking tour since it was diffith for some members of his group.

3. If you were visiting the temples on your own, would you have explored the handheld
tour guide before you went to the temple? Do you think it is a good tool to use before,

during, or after visiting a temple?

Both James and Michael thought it would be a good idea to explore the tour guide
before visiting the temple. Both also said they would enjoy looking at other features of the
tour guide such as the photo gallery or videos now that they had walked through the temple

and appreciated the location.

4. How much did the handheld tour help you know where to turn and what to look at in the

Temple? Did you ever feel lost in the Temple? Did you ever feel lost in the handheld tour?

James thought the tour guide did a good job. Michael felt it was good and helpful to

use the handheld tour. He felt a little lost a couple of times.

5. The handheld tour included photographs, text, videos, and maps. Which elements do you

want to see more of? And why?

James said more maps or maybe more points of interest would be helpful. Michael

said he wants more video clips like an introduction to the sites visited.

22

6. How hard or easy was it to navigate through the program and find the information you

need? Do you have any suggestions to improve the navigation?

James suggested “It would be helpful to say how long a walk between points to expect,
i.e. walk 500 feet and turn. Also it would be good to be able to choose the routes of sites

visited.” Michael said that he wants a self-guided tour which has more options.

7. Did it feel natural to use a handheld tour guide as you walked around the temple? Did
using the guide enhance the experience of visiting the temple or did it get in the way of

paying attention to the physical objects in the temple?

James said the tour guide introduced him to things he might not have picked up on his own.

Michael said the guide is a good accompaniment to the tour.

8. Would you want this kind of tour guide program for other places? What kind of places

would benefit fi'om a handheld tour guide?

James said museum and theme park would benefit from a handheld tour guide. Michael
said he would like to see tour guides of Snake Alley, Shih-Lin night market, and Yang

Ming Mountain in Taipei.

Discussion
Based on observation during the field test, users want to know how long it will take

between points, especially longer walk between points like the walk between point A and

23

point B or point C and point D in Fig 5.1. When the testerwalked between pointAand
point B or point C and point D, he kept looking at the navigational aid picture on handheld
tourguidebecausehewonderedifhemissedthepointtoturnornot. Basedonthe
information, it might be a good idea to make distance between each point equal by adding
more points between Point A and Point B. Besides, it might help to indicate distance for

users to expect.

\\' Point D

\\\ \ Point c
C

 

 

 

Fig 5.1 Map of Tarnshui walking tour
The subjects expressed a desire for the tour to be nonlinear, allowing them more

fieedom to decide where to go, but still providing lots of information about interesting

24

things to notice and interesting background stories. Fig 5.2 illustrates the difference
between non-linear self-guided tour and linear guided tour. The walking tour of Temples of
Taipei is a linear guided tour which only has one option. In a non-linear self-guided tour,

users will have more options like turning right or left in each point.

Non-linear self-guided tour

 

C—I. .—— . k M

Linear guided tour

Fig 5.2 Non-linear self-guided tour and linear guided tour
In addition to wanting more choices of routes, James suggested that he would like a

choice of length Of tours such as choices of 3-hour walking tour, 2-hour walking tour etc.

25

Both users want more in-depth information about the site they visit in the walking tour.
Michael wrote: “although the tour did introduce me to many interesting sites, I would have
liked it to point out specifics within the site or special firings to look for.” James suggested
that he would like to have more buttons on the picture which brings up more information
about the location. (See Fig 5.3)

One user said he would like to

have buttons on the picture which

brings up more information about

the location. For example, the user
wants more details of this mountain.

 

   
 
  

Fig 5.3 Buttons on the picture which brings up more information
J arnes thought perhaps there should be a greater emphasis on maps rather than
pictures: “I wonder if this map won’t be the larger one and then the picture would be
smaller.” Ideally, the user could choose to toggle the size and position of the map and the
picture whenever they wanted, allowing them to have the large view of either the map or

the picture at any time.

26

CHAPTER6

The Future of Tour Guides on Pocket PC

Location-based Advertising and the Walking Tour
Location-based services have gained a lot of attention recently. The walking tour part

of Temples of Taipei could be considered a simple location-based application.

In an article of Harvard Business Review, David Kenny and John F. Marshall make

the following remark:

Mobile devices and intemet access in a broad range of public venues will let
contextual marketers link real-life situations to virtual information and offerings.
For instance, Unilever’s mobile recipe book concept, which will be available on
digital phones in Europe, should influence consumers’ packaged-good decisions
far more than the company’s web site ever could. Intended for use while shopping,
the mobile tool suggests recipes and breaks them down into their ingredients

identified, wherever possible, by their Unilever brand.4

The walking tour could be used to identify where the user (consumer) is and
advertisements of the nearest restaurant or gift shop could be placed at appropriate spots in
the tour. This is a good example of “linking real-life situations to virtual information and
ofi‘erings.” And these advertisements could be more effective than web sites of that

restaurant or gift shop because it is so much easier for the consumer to take a few extra

 

‘ David Kenny , John F. Marshall, Contextual Marketing, Harvard Business Review, Nov/Dec 2000

27

steps and actually show up at the location being advertised. Furthermore, GPS (Global
Positioning System) on Pocket PC will enhance the tracking feature of the walking tour

and allow the creation of a much better navigation guide.

The Future of Phrase Book
The phrase book is limited because users can only use sentences which are
pre-defined and pre-recorded. There are not many sentences to choose fiom and adding

more sentences is not feasible since the storage space of Pocket PC is so limited.

But it’s already possible to design a program for Pocket PC that helps tourists and
people in Taiwan communicate with more freedom when one party uses Chinese and the
other uses English. A front-end interface designed in Flash for a connected Pocket PC
which talks via wireless network to a back-end server has the capability to translate one
language to another in real time. Here is how it works. First, English-speaking tourists
input a sentence in English using the Flash program on Pocket PC. Then the text in English
is sent to the server through wireless intemet connection. When the server finishes
translating, it sends the sentence in Chinese back to the Flash program and the sentence can
be viewed by people in Taiwan. On the other hand, people in Taiwan could also use the
Flash program to input a sentence in Chinese and have the sentence translated by the server
and then viewed by the English-speaking tourist. The main reason to use a server for
translating is that only a server has enough processing power and storage space to translate
“natural language” on the fly. Besides, it is easier to add new vocabularies, phrases,
sentences, and grammar into the database of the server and the accuracy of translating can

be improved everyday. In fact, this kind of translation service is already available on the

28

intemet. For example, I found AltaVista’s Babel Fish translation service is a very effective

translating tool for languages like Japanese, Chinese, or English.

Babel Fish Translation

Iwant one bottle of soft drinks

Search the web with this text

Chinese to English v

Fig. 6.1 Screenshot of Babel Fish translation service

 

In the near future, the Pocket PC could have speech-recognition capability to turn
English and Chinese speech into text which could be transferred and translated by the
server. The communication process could become easier because the text could be input
using voice. Furthermore, by using a text—tO-speech output program like IBM Embedded
ViaVoice Mobility Suite for Pocket PC which generates Chinese or English speech from
Chinese or English text transmitted fi'om the translation server, the quality of
communication between English-speaking tourists and people in Taiwan could be

improved greatly.

29

BIBLIOGRAPHY

30

Hessert Kathleen, Getting Through to the "MTV" GeneLation, October 1999
http://www.bbhighwavcom/Talk/Interactive Coach/CommComeL/getting_thr0ugh_mtv_

generationasp
Pises, Anna Marie. (2002) Flash Desigmfor Mobile Devices. NY: Hungry Minds

Wimmer, Roger D. and Dominick, Joseph R. (1991) Belmont, CA: Ma_ss MedjagResearch,
Third Edition. Wadsworth Publishing Company

31