(註:補助註冊費)
IEEE International Conference on Advanced Learning Technologies 是與數位學習領域相關的研討會，
ICALT 2011 於 7/6-7/8 在美國喬治亞州喬治亞大學所在地 Athens 舉行，因為執行校園用電子書專案之 緣故，原來對此研討會較為陌生的我，藉由投稿 Analysis and Interpretation of e-Reader User Logs: A Case Study of High School Students’ User Behaviors 的機會，第一次參與數位學習相關的國際會議，由 於經驗較缺乏，因此抱持學習的成分居多，也對不同領域學者的觀點與研究取徑有一番觀察與比較。
此次研討會的主題為:Educational Technology Forecast: Cloudy with a Slight Chance of Gain，有了科技 輔助，是否未來學習就一片萬里晴空，或者仍然多雲陣雨?顯然在數位化工具越來越普遍的今日，資訊 科技究竟能在教育學習扮演何種積極之角色，有再度審視與省思的空間，研究學者不再以開發新平台 或推動新工具為唯一目標，而是更客觀、忠實地評估這些工具的正面與負面衝擊，這種勇於反思的精 神，正是所有研究人員應該抱持的態度。
ICALT 2011 在喬治亞大學(University of Georgia)的會議中心舉行，該大學創校於 1785 年，至今已有兩 百多年的歷史，因此校園內建築頗為多樣化，有新穎的體育場，也有具歷史的古老建物，漫步於校園 四周，令人發思古之幽情。
在會議的安排部分，除了每天早上的 Keynote 或 invited speech 外，共分四個平行場次，每天安排到晚 上六點結束，poster 部分從下午三點才開始，因此整個會議 oral presentation 的比重較高，至於論文報 告的安排，則依照下列主題分組進行。從下面的主題列表，也可略窺目前數位學習領域研究的若干重 點。
Adaptive and Personalized Technology-enhanced Learning I,II,III,IV
Learning Systems Platforms and Architectures I,II,III,IV
Semantic Web and Ontologies for Learning Systems I,II
Computer Supported Collaborative Learning I,II
Digital Game and Intelligent Toy Enhanced Learning I,II,III
Wireless, Mobile and Ubiquitous Technologies for Learning I,II
Virtual Worlds for Academic, Organizational, and Lifelong Learning I,II
Web 2.0 and Social Computing for Learning and Knowledge Sharing I, II
Technology-enhanced Science Education I,II
Rethinking Pedagogy in Technology-enhanced Learning I,II
e-Assessment and new Assessment Theories and Methodologies
Knowledge and Competencies Management
Emerging technologies for enhancing learning processes and outcomes, and diagnostic of mental representations in learners
Open Content and Learning Technology Standards
Self-Regulated Learning in Responsive Open Learning Environments
Affective and Pervasive Computing for Learning
我 們 的 論 文 被 排 在 7/7 ( 周 四) Parallel session 4a (11:00-13:30) Wireless, Mobile and Ubiquitous Technologies for Learning II 進行口頭報告，由於我們進行研究所使用的電子書系統是自行開發，有關 系統整合、介面設計、教材轉置均是由本團隊所執行，因此除了分享高中生使用行為之觀察與 log 分 析外，也帶了幾台機器供現場聽眾試用，以提升互動效果，當場有三位聽眾提出問題，多能切中核心，
並未有太令人意外的疑問發生。
除了早上的公開演講，在口頭報告部分，我挑選的場次仍與技術較有密切相關者為主，如 Digital Game and Intelligent Toy Enhanced Learning，Wireless, Mobile and Ubiquitous Technologies for Learning，
Adaptive and Personalized Technology-enhanced Learning 等，有些論文有完整的論述，從學習理論出發，
搭配設計方法與實測結果及分析，有些則屬較前期的開發評估，概念新穎但效果仍待驗證，從這些不 同學科或方法探討中，讓我對數位學習有了更廣闊的見識。
此次參與會議的台灣學者也不少，有台科大、台師大、中央、成大、中山等研究團隊，在國科會的領 域分類屬於科教處，過去本人接觸不多，藉這次會議得以結識更多學者，也是額外的收穫。
Analysis and Interpretation of e-Reader User Logs
A Case Study of High School Students’ User Behaviors
Wen-Hung Liao, Chien-Pao Chueh Dept. of Computer Science National Chengchi University
Taipei, TAIWAN e-mail: whliao@cs.nccu.edu.tw
Abstract—This study explores the daily life user experiences of an experimental e-book reading device among high-school students, aiming to understand how well the digital natives accept the use of e-book reading devices and the potential utilities of such devices for them, either for leisure purposes or as an assistive educational tool. Toward this goal, we have custom-designed the reader user interface as well as the e-book content to suit the needs of this particular user group.
The unique opportunity of having access to the hardware device, software design and potential users creates an ideal experimental platform for us to unbiasedly investigate the role of this new technology through a long-term user behavior collection and analysis process. We anticipate that the new reading behaviors of the digital natives will provide clues for further improvements in the design and development of digital reader devices.
Keywords- e-ink display; e-book reader; user log analysis;
user experience
I. INTRODUCTION
Digital natives, the generation growing up with abundance digital technologies, have developed new ways of obtaining and consuming information. With continuing advances in information technology, especially in the area of display hardware, market for the once slow growing e-book reading devices has revived. The introduction of Kindle in 2007 by Amazon has kick-started a new wave of devices based on electronic-ink display. All the major book retailers in the U.S. have followed suit, aiming to replicate the success of Amazon.
The format of digital electronic books has been in existence for decades. The recent surge of e-book readers is mostly attributed to the reduced manufacturing cost of e-ink display. Unlike the conventional LCD, the e-ink screens reflect lights like ordinary paper. It causes less eye strain, and consumes very little power. Anecdotal evidence suggests that e-paper is more comfortable to read than conventional displays. However, current e-ink devices have only 16 levels of gray. The refresh rate is also too slow to show animation or video. These constraints have led the device makers to emphasize its portability, readability and
‘greenness’.
Although many e-book readers have been introduced in the last two years, they are mainly targeted on general users for their casual reading activities. Several field studies have been conducted to gather the initial responses of the users to
this new technology [1,2]. A few pilot programs sponsored by Amazon have been carried out in several U.S.
universities to study Kindle DX’s functionality and potential to replace paper textbooks in college. The initial responses are disapproving, due to the device’s slow performance and lack of note taking capability [3].
In this research, we are aiming at a different user group:
high school students. We also have a different perspective.
Instead of forcing the participant to use the device according to our direction, the objective is to exploit the role of this device in the student’s daily lifer user experiences through a long-term user behavior collection and analysis process. We are able to obtain the hardware from a supportive manufacturer. The device is based on a proprietary embedded OS. With access to the software development kit (SDK), we have the capability to customize the e-reader user interface, as well as the system functionalities of the device. To understand how this device is used, we asked the participants to fill out surveys. Additionally, a background process is embedded in the device to gather user log for subsequent analysis.
The rest of this paper is organized as follows. In Section 2 we list the specifications of the device and discuss some design constraints. We then present the system framework and user interface design. Section 3 is concerned with the experiment design, including subject selection, deployment process and test schedule. Section 4 presents the user log analysis. Section 5 concludes this paper with a brief summary and outlook on future work.
II. DEVICE AND USER INTERFACE DESIGN
The e-book reading device used in this experiment has dual screens, as depicted in Fig. 1. The top screen is a six-inch e-ink display panel with a 600x800 resolution and 16 levels of grayscale. It is used for displaying information.
The bottom is a LCD-based touch screen whose resolution is 360x100. This area is dedicated for user operation and interaction. The main screen shows the six functions of the e-book reading device. My textbooks item (A) contains course materials and assigned readings. Recent readings item (B) provides a short cut to recently opened documents.
My documents item (C) allows the user to access self upload pdf or word doc files. My music item (D) contains audio collections. My pictures item (E) is a place holder for png, jpg or gif files. System settings item (F) permits the 2011 11th IEEE International Conference on Advanced Learning Technologies
user to adjust the device’s settings, including data/time, refresh mode, etc.
Due to limited display area and number of gray levels, not all textbook materials are suitable to be presented using this device. Four subjects, namely, Chinese literature, English, History and Geography have been identified after consultation with high school teachers.
Figure 1. Specifications and user interface of the e-reader.
III. EXPERIMENT DESIGN
Forty 10th-grade students from the affiliated high school of National Chengchi University (AHS of NCCU) participated in the experiment after obtaining written consent from their parents. We held a discussion session to solicit input from the potential users during the design stage.
We have carefully described the characteristics of the e-ink display for the users to better understand the limitations of this device. The users are encouraged, yet not forced, to use the device as much as they can.
To understand how this device is used, we ran a log-collection process in the background to record user activity.
Specifically, all the key-press events, including hardware and virtual keys, the time stamp, and the corresponding page will be written to a log file. The format is shown in Table 1.
TABLE 1.FORMAT OF THE USER LOG Key Pressed Time Page
1 VK_Menu 1272888558 Page_ResourcesCenter
2 VK_A 1272888562 Page_MiniReader
The first phase of our experiment starts at 5/5/2010. We gathered user logs at 5/31/2010, a little more than three weeks after the initial distribution. Out of the forty machines sent out to the students, thirty-eight log files have been successfully collected. Two devices were sent back for repair due to fragility of the display panel.
The second phase of our experiment falls between 6/1/2010 and 7/2/2010, right before the summer vacation starts. For this period, we are able to recover only 34 system logs since four more devices were out of order and sent for repair. We made our final collection of machine
logs on 10/29/2010, almost 6 months after the initial distribution. We are able to record 33 user logs this time.
IV. USER LOG ANALYSIS
This section discusses the results of user log analysis.
We will first present the overall usage statistics and identify active users. We will then inspect the operation log of active users to study how the e-book readers have been used in a real-life situation.
A. Overall Statistics
Fig. 2 summarizes the usage statistics of the thirty-three students who participated in all three phases of the experiment. About 50% (17 out of 33) of the users spent less than 30 hours on the machine. Our speculation is that these are individuals who may be curious about the device, but the initial excitement wears down quickly after hands-on experience. These students are classified as ‘inactive’ users.
Follow-up questionnaire has been distributed to inquire the reasons for low utility. Ten participants whose usage ranges from 30-60 hours are classified as ‘casual’ users. The remaining six participants whose usage exceeds 60 hours are considered ‘active’ users.
The next useful information is to examine the average time between consecutive key-presses (or clicks) to deduce the possible user behavior. If the interval between two operations is short, the user may just be navigating through different menu structures. On the other hand, if the interval is long, we have reason to believe that the user is reading the material shown on the e-ink display. The average time between two operations (Tcc) can be computed by:
(1) Fig. 3 depicts the calculated Tcc for all 33 users. The average is 26.1 seconds with a standard deviation of 9 seconds.
Figure 3. Average time (in seconds) between two operations
We speculate that active users spend most of their time reading the e-books instead of jumping around different hierarchies of user menus. Consequently, their Tcc should be longer than that of other types of users. This is confirmed by the statistics shown in Fig. 4, in which all 6 active users
users, the distinction is not as clear, as can be seen from the mixed results in Fig. 4.
Figure 4. Histogram of Tcc
B. Statistics of Active Users
Certain statistics only make sense for those logs obtained from active users. As a result, the subsequent analysis will focus on the detailed usage patterns of the active users.
Fig. 5 shows the percentage of time active users spent on different type of materials. Either textbooks or assigned readings (i.e., classic Chinese or English novels) are not very well-received, totaling only 25% of the overall reading time. Instead, most active users prefer to upload their own documents. We believe this is strongly correlated to the policy we have in this phase of the experiment, i.e., we treat the e-book reader as a complementary tool and let the user decide when, where, and how to utilize the machine.
A user session is defined as the interval between system boot-up and standby or shut down. We can tell if the e-book reader is used in fragmented time intervals by examining the statistics of the user session. If we set the threshold to be 300 seconds (5 minutes), the average is 34.7% of long sessions (over 5 minutes) vs. 65.3% of short sessions (less than 5 minutes) for all active users.
Figure 5. Usage pattern: type of materials consumed
V. CONCLUSIONS
In this paper, we have described an experiment aimed at facilitating high school learning through e-book reading devices. Results from user log analysis indicate that the reader can play an assistive role in promoting reading activities. We are currently entering the second stage of design and experiment using a new hardware platform running Android OS. We are attempting to extend the e-pub format to enable the incorporation of online tests and interactive content. It is expected that with these new functionalities and integrated online services, the e-book readers will gain wider acceptance in modern classrooms.
REFERENCES
[1] Dennis T. Clark, Susan P. Goodwin, Todd Samuelson, and Catherine Coker, 2008. A qualitative assessment of the Kindle e-book reader:
results from initial focus groups, Performance Measurement and Metrics, Vol. 9 Issue: 2, pp.118 – 129, 2008.
[2] Guy Janssens and Harry Martin, 2009. The feasibility of E-Ink readers in distance learning: A field study, International Journal of Interactive Mobile Technologies, Vol. 3, No. 3, pp. 34-37, 2009.
[3] The Kindle DX pilot project at Darden,
http://www.darden.virginia.edu/html/standard.aspx?menu_id=72&st yleid=2&id=19304
Figure 2. Total usage (in hours) in three experiment phases
0 2 4 6 8
0-10 11--15 16-20 21-25 26-30 31-35 36-40 41-45
Inactive Users Casual Users Active Users
23%