Free/Libre/Open Source Software Asian Developers Online Survey (FLOSS-ASIA)

As in other parts of the world, open source software and free software (OSS/FS) is attracting a great deal of attention in Asia, primarily from a cost perspective. Unlike Japan and the West, however, where OSS/FS is used primarily for servers, in Asia there are great expectations for OSS/FS for desktop applications as well, and different language versions of the Linux OS and office suites are being developed.

However, the situation of OSS/FS in Asia is by no means uniform. China is investing enormous sums of government funds to develop a desktop Linux environment equivalent to Windows. In Korea and Taiwan, products with Linux pre-installed are actively being developed. In this way, in East Asia an IT industry has already developed, and these countries are supporting OSS/FS in order to nurture their own IT industries. In Southeast Asia, however, OSS/FS activity has only just begun. As personal computers are relatively expensive, in these countries expectations are greatest for OSS/FS as a low-cost desktop environment in an effort to eliminate the "digital divide," one of the major problems associated with information technologies.

In all of these countries, the training of OSS/FS developers is becoming a pressing national policy issue. It is also crucial for Japan to take the initiative in the OSS/FS field in Asia, and to contribute to Asian countries and achieve growth together. This survey was conducted to determine the current status of OSS/FS developers in Asia and to provide reference material for studying these policies.

This survey represents the Japanese version of the FLOSS survey* conducted in Europe in 2002, the FLOSS-US survey** conducted in the United States during the first half of 2003, and the FLOSS-JP survey*** conducted in Japan during the latter half of 2003. The FLOSS survey was implemented by the International Institute of Infonomics, University of Maastricht, Netherlands, in 2002 under a grant from the IST Program of the European Commission. The FLOSS-US survey was implemented in 2003 by Stanford University's Stanford Institute for Economic Policy Research (SIEPR) under a grant from the National Science Foundation (NSF). As these surveys were conducted in English, there was little participation by developers in Japan and the other countries of Asia. Accordingly, the authors decided to implement a FLOSS-JP survey in Japanese and a FLOSS-ASIA survey in the languages of the other countries of Asia.

1.2 Survey method

During a two-month period from December 1, 2003 through January 30, 2004, an online survey system was put up on the Mitsubishi Research Institute website (http://oss.mri.co.jp/floss-asia/). The surveys were prepared in Asian languages (Traditional/Simplified Chinese, Korean, Thai, and English for other countries), to make it easy for those developers without English proficiency to participate as well. No particular restrictions were placed on who should respond to the survey, as long as the respondents considered themselves to be open source or free software developers who applied through several specific communities.

To enable global comparison with results obtained elsewhere, most of the questions were taken directly from the European FLOSS survey and the FLOSS-US survey. For this reason, the authors worked together with the leader of the FLOSS survey, Rishab Aiyer Ghosh, in developing the questions. In general, the survey was the same as the FLOSS-JP survey, but several questions relating to skill development were added. The scope of the survey was as follows:

1.3 Tabulated results

During the two-month survey period, 138 responses were received in all. There were many responses from two economies, Korea and Thailand, which were among the countries for which local language versions of the survey had been prepared; next were Taiwan and China. The reason for the small number of responses is probably due mainly to inadequate promotion. Notices were put up on the news sites and mailing lists for the OSS/FS developer communities in each country. However, in the case of the Japanese survey, repeated announcements were made in several different media outlets, and therefore there is an undeniable sense that the publicity was insufficient. In addition, the accuracy of the translation into different languages was not necessarily high in some cases, and so the possibility that full participation could not be obtained from respondents for this reason can likewise not be ruled out.

In the sections below, the tabulated results will be presented and compared with the results of the FLOSS surveys in Europe, the United States and Japan.

2. Profile of developers

First, a profile of the individual OSS/FS developer (gender, family composition, occupation, etc.) will be presented.

2.1 Gender and family composition

As in the surveys in Europe, the United States and Japan, almost all of the respondents were male (98.5%). In terms of family composition, a high percentage (72.1%) of the respondents were single. The fact that, as will be covered later, the average age of developers is low is undoubtedly one reason for the fact that so many are unmarried.

2.2 History of OSS/FS development and present age

Compared with the European, American and Japanese surveys, the year that respondents began involvement with OSS/FS was later, around 1995, and involvement increased steadily thereafter.

Figure 3 shows the distribution of values for the age of the respondent when he/she began involvement in OSS/FS development. Starting with age 19, involvement in OSS/FS development increases rapidly, and most respondents said they were between the ages of 20 and 25 when they began development. It seems clear that respondents began their involvement in OSS/FS development when they were students or immediately after entering the work force.

Deriving the respondent's current age from the age at which he/she began development and the respondent's age at that time gives the distribution shown in Fig. 4. As this distribution shows, respondents are active mainly in their 20s.

Table 1 shows the mean values for the year that the respondent began development, the respondent's age at that time and the respondent's current age, compared with the values for the surveys in Europe, the United States and Japan. The average developer began his or her involvement in OSS/FS development around the year 2000, when he or she was around 23, and the developer's current age is around 27. Compared to Europe, the United States and Japan, developers in Asia are about the same age as those in Europe and the United States and younger than those in Japan. Their age at the time they began development is in between the age of developers in Japan and those in the West, and the year they began involvement is the latest of any of the FLOSS surveys.

2.3 Academic background

Figure 5 shows the academic background of respondents. 63.9% are college graduates. The number of respondents that have completed graduate school is lower than in the European, American and Japanese surveys, but the percent of persons in Asia who go to college and graduate school is thought to be lower than in Japan and the West. Considering this situation, it can be said that OSS/FS developers are highly educated in their respective countries.

2.4 Occupation

As in the other surveys, many Asian developers are in software-related fields, with 27.8% of them describing themselves as software engineers and 17.3% as programmers (Fig. 6). This means that about half of the developers are involved in software development not only as a hobby but in a work capacity as well. 12.0% of the respondents were affiliated with universities and research institutions. Among these are thought to be some, although few in number, who released publicly as OSS/FS the software they had developed as part of their research activities. In addition, the proportion of respondents who described themselves as information-related consultants and managers was 6.0% each. Thus people in a wide range of occupations are involved in development. The percentage of students was 15.8%, about the same as in the Japanese survey (14.4%); although the figure is not that high, approximately two-thirds of them (10.5%) are in information-related fields, which is close to the number (approximately three-fourths) in the European and American surveys.

The survey divided respondents according to information and non-information related fields. However, in statistical terms, having a small number of respondents would result in greater variations. For this reason, and to enable comparison with the other surveys, it was decided to organize respondents' occupations as shown in Table 2 for the purposes of further analysis. The results organized in this manner are shown in Fig. 7.

Software engineer	Software engineer
Student	Student (IT)
Student	Student (other sectors)
Programmer	Programmer
University / Research Institute	University / Research Institute (IT)
University / Research Institute	University / Research Institute (other sectors)
Engineer (other sectors)	Engineer (other sectors)
Other	Consultant (IT)
	Consultant (other sectors)
	Manager (IT)
	Manager (other sectors)
	Executive (IT)
	Executive (other sectors)
	Marketing / Product sales(IT)
	Marketing / Product sales(other sectors)
	Other (IT)
	Other (other sectors)

2.5 Nationality and current residence

Tables 3 and 4 show the nationality and current country of residence of the respondents. A majority of respondents listed Korea (31.6% and 30.8%, respectively) and Thailand (26.3% and 23.3%, respectively), followed by Taiwan, China and Malaysia. Rather than reflecting the degree of activity of OSS activities in these countries, it would be more appropriate to view this as reflecting the degree to which the announcement of the survey spread to developers in each country. The figures for nationality and current country of residence were almost the same, indicating that there are almost no developers who live abroad.

3. OSS and FS licensing system

3.1 OSS/FS

In this survey, respondents were asked whether they made a distinction between open source software and free software and, if so, the community with which they considered themselves to be affiliated. Most (67.7%) respondents answered that they did make a distinction between open source and free software. In addition, a majority identified themselves with the open source rather than the free software community (Fig. 10). The figures were 42.1% for open source and 25.6% for free software, almost exactly the same ratios as in the Japanese survey (43.7% and 26.7%, respectively). In the European survey, 48.0% were in the free software community and 32.6% were in the open source community. In the American survey, the percentages for free software and open source were almost exactly equal (31.4% and 31.5%).

Next, respondents were asked whether the open source community and the free software community were the same or different. As Fig. 11 shows, almost half (48.1%) of the respondents said the open source software and free software communities were "different in principle but the same in practice." Only 15.0% said there was no difference between the two communities, indicating that the majority of respondents recognize a distinction between open source software and free software. Most of the respondents who made no distinction between the communities were neutral, not allying themselves with either the open source community or the free software community. Most (94.6% open source and 94.1% free software) of the respondents who allied themselves with one of the two communities said there were differences between them (Fig. 12). The recognition of differences between the two communities was similar; slightly more respondents answered that the communities were different both in principle and in practice (46.4%) than those who answered that the communities were different in principle but the same in practice (44.1%). These results show that the respondents involved in OSS/FS development who made a distinction between the two communities did not do so out of a vague sense but with a clear understanding of the differences between free software and open source ideology.

Finally, respondents were asked which licensing system they would select if the code they created were not dependent on other code and they could select any licensing system (Fig. 13). Licensing systems tend to be considered troublesome or difficult, but only a small percentage (9.0%) of developers responded that they did not particularly care which licensing system they used. The most popular licensing system by far was GPL compatible (64.7%), followed by BSD style (15.0%). In terms of affiliation (Fig. 14), GPL compatible was the most popular licensing system for both the open source and the free software community, being chosen by approximately 70% (69.6%). In the Japanese survey, 42.0% selected GPL compatible, so the figure of 64.7% is quite high.

4. OSS/FS development activities

4.1 Development time

In Asia as well as in the other surveys, the amount of time that respondents actually spend on OSS/FS development is comparatively short. Although the time was longer than in the Japanese survey, about half (49.7%) of respondents said they spend 5 hours or less. On the other hand, 6.0% of respondents indicated that they spend 41 hours or more on OSS/FS development, and it is highly likely that these respondents are engaged in OSS/FS development as a business.

4.2 Development target

Networks and web services accounted for the largest percentages of development target fields; multimedia-related activities accounted for only a small percentage (Fig. 16). Most of the networks and web services consisted of small programs such as scripts and utilities. This is thought to be because such programs are easy to develop.

The order of the top four target fields was exactly the same as in the European and Japanese surveys. However, the percentage of respondents active in all four target fields was comparatively high at 10-15%.

4.3 Development environment

To determine the development languages and tools that developers used in OSS/FS development, respondents were asked to state their development platform, languages and tools used, and so on.

Linux (RedHat, Debian, Vine, Turbo, Plamo, Gentoo, Mandrake, SuSE, Slackware and other Linux varieties) was the most frequently used platform (71.4%) (Fig. 26). Windows was next popular at 15.0%.

Most respondents used the Linux platform when they first began programming. A high percentage also used Windows and DOS, indicating that many developers switched from Windows and DOS to Linux.

The figures were almost identical for the platform that the respondent used when he/she first began OSS/FS development and the platform most often used. Since the history of OSS/FS in Asia is brief, it would appear that there was no major change in platform except in the case of a change in distribution.

Fig.18 (Q12) Development languages/tools with which respondent has experience

Major development languages such as C, C++ and Java were the most common responses. In addition, 60.9% of respondents had experience with PHP and 55.6% had experience with SQL, indicating that many developers are engaged in creating web services using databases.

The most frequently used desktop environment was GNOME (39.1%), while KDE and X + Window Manager were cited by a combined 66.1% of respondents, indicating that developers use the OSS/FS desktop environment.

4.4 Projects

Involvement in the OSS/FS community was surveyed from the standpoint of not only time-related aspects but in terms of projects as well. The survey asked respondents to state the number of projects in which the respondent had been involved and was currently involved, the number of developers with which he or she had had contact, and the respondent's leadership experience.

Respondents had not been involved in very many projects in the past; 67.7% of respondents said that their past experience consisted of five projects or fewer (Fig. 20).

Respondents were not currently involved in many projects. The number of respondents answering that they were currently involved in one and two projects was 36.1% and 23.3%, respectively. 17.3% of respondents were not currently involved in any projects (Fig. 21). However, the figures for cumulative number of projects and current number of projects were slightly higher than the same figures in the Japanese survey.

Fig.22 (Q17) Number of projects in which respondent is currently involved as a leader

Nearly half (45.9%) of respondents had no leadership experience, but the remaining respondents had had the experience of leading a project themselves (Fig. 22). In other words, a very large number of developers have been involved in projects by starting the project themselves. The percentage of developers with leadership experience was almost exactly in the middle between the values for Japan (56.6%) and Europe (35.2%).

4.5 Community

Fig.23 (Q18) Number of members of the community with which respondents have regular contact

To determine involvement in the OSS/FS communities, the survey attempted to determine how many people the respondents had contact with on a regular basis. Although respondents were only in contact with a few people (24.1% said "no one" and a combined total of 38.4% said between one and five persons), the figures for Asian developers were slightly higher than those for Japanese developers (34.6% and 33.3%, respectively).

In the Japanese survey as well, the percentage of developers who are active in the global community was high (37.7%), but it is clear that the figures for Asia are even higher. 47.4% of respondents said they were only active in the community in their own countries, and the remaining 52.6% are active in the global community.

The English language ability of Asian respondents was higher than that of the respondents in the Japanese survey. However, it is possible that this does not necessarily indicate a high level of English language ability in the countries of Asia, for two reasons. One is that a domestic community has not developed in each of the countries of Asia, and so ultimately it is necessary for developers to be active in the global community. The other reason is the possibility that the respondents were those who have a high level of awareness and are very active in their community. As the promotion of this survey was insufficient in local communities, it is possible that the main members of the community accounted for a majority of the respondents. The trend of respondents with a higher level of English ability being more active in the global community was the same as in the Japanese survey.

Fig.26 (Q20) Major types of activity in which respondent was involved in OSS/FS development

Of the major types of activities in which respondents are involved as part of OSS/FS development, coding-related activities (developing main functions, fixing bugs, creating patches, testing) were cited by a majority of respondents. As in the Japanese survey, document preparation/translation and support activities each garnered percentages of 10-15%, revealing that many developers contributed to activities other than coding (Fig. 26). The figure for localization (18.0%) was higher than in the Japanese survey (10.8%), and this is thought to indicate that localization in these countries lags behind that in Japan.

4.6 Qualifications and education

The percentage of respondents having qualifications related to OSS/FS (13.5%) was slightly higher than in the Japanese survey (7.9%), but it was still low. Conversely, the percentage of respondents who had learned the knowledge needed for OSS/FS development by themselves was high, but lower than the figure in the Japanese survey. Many respondents had studied OSS/FS development in college (a combined total of 22.6%). Interestingly, the number of people who had learned in science and engineering-related fields (11.3%) was greater than the number who of people who had learned in information-related fields (9.8%). It is very likely that the people in information-related fields learned OSS/FS development by themselves, but the results show that OSS/FS related education is also conducted in science and engineering-related fields.

5. Motivations, expectations and views regarding OSS/FS development

5.1 Motivation for OSS/FS development

Five questions were asked of the respondents: what was the motivation for yourself/others to become involved in OSS/FS development; what are your/others' expectations for the OSS/FS community; and what is the purpose of the OSS/FS community. As in the European, American and Japanese surveys, a majority of respondents answered to acquire and share knowledge and skills (Fig. 28). Respondents also felt that others did not place as much value on skills as they did, and that proprietary software was not good; these trends were the same as in the Japanese survey. A major difference with the Japanese survey was that respondents in the Asian survey felt that in many cases other developers were involved in OSS/FS development for fame or profit.

5.2 Beginning of involvement in OSS/FS development

In the Japanese survey, approximately half of all respondents said that their involvement began with coding, such as releasing a program of their own (36.5%) or sending a patch (16.2%). In the Asian survey, most people said their involvement began with an activity other than coding. Half (50.5%) of the respondents cited exchanging emails with others (20.5%), responding to a question from a friend (15.6%), or responding to a question from a user on a mailing list (14.8%) as the beginning of their involvement. The phenomenon of involvement in OSS/FS being not limited to coding can be seen even more clearly than in the Japanese survey. However, as shown in Figure 26, most people are currently involved in coding activities. It can be inferred that many people have gained the technical skill to code through their involvement in OSS/FS development.

5.3 Importance of signature

A signature in the source code is proof that the work is one's own creation. Accordingly, developers were asked how they regarded signatures.

Awareness with regard to signatures was low in Asia as well; the results were about the same as in the Japanese survey. 88.0% of respondents said they included signatures in their work, but only 27.8% said they used their real name.

6. Compensation for OS and FS development

6.1 Compensation for OSS/FS

Approximately half of all developers said they derived monetary or indirect remuneration related to OSS/FS development activities (monetary: 45.1% / indirect: 50.4%). The level of developers receiving assistance was in line with that in the European and American surveys. In the European survey, 53.7% of developers said they received monetary compensation; in the American survey, this figure was 43.2%. In contrast, the low level in Japan (only 26.8%) is noteworthy.

Universities and public organizations were the main source of assistance. This differed from Europe and the United States, where private companies are the main source of assistance, and Japan, where there is a great deal of assistance from public organizations but very little from universities.

6.2 Recognition of OSS/FS development

On the question of whether companies (schools) were aware of the involvement of respondents in OSS/FS development, the results placed the Asian respondents in between their counterparts in the Japanese and American surveys. In other words, the figures for "involvement is known" were Japan 34.7%, U. S. 59.2% and Asia 39.1%, and the figures for "involvement is not known" were Japan 22.6%, U. S. 40.8% and Asia 33.8%, As in the Japanese survey, the developers wanted schools/companies to recognize their involvement (23.3%) and wanted to receive compensation (32.3%). Only 10.3% were satisfied with the current situation (Figure 35). What developers really want is to be involved in OSS/FS as an occupation.

7. Skills Development in OSS/FS community

The following six questions were asked only in the Asian survey. Respondents were asked what skills they acquire through their involvement in the OSS/FS community, whether these skills are actually useful, and whether this is an effective way to learn such skills.

Fig.36 (Q53) Would you say that people learn very much from participating in the OS/FS community?

86.5% of developers felt they learned a lot from the OSS/FS community. None of the respondents said they did not learn anything from their involvement.

Fig.37 (Q54) According to your personal experience, what do you learn from participating in the OS/FS community?

In addition to technical matters such as programming techniques (54.1% and system design skills (45.1%), respondents also learned interpersonal and communication skills such as how to cooperate with others to pursue work projects (38.3%) and how to accept and deal with criticism from others (26.3%).

Fig.38 (Q55) How do you learn software development skills? (outside my job/school/university)

In response to a question about how the respondent had learned software development skills apart from work and the like, more people had learned by reading the source code of others (40.6%) than by reading programming books and magazines (27.8%), indicating that the public disclosure of source code was helpful in improving software development skills.

Fig.39 (Q56) On average, how many hours per month do you spend on learning software development skills?

OSS/FS developers spent a comparatively long time studying software development skills. 19.5% of the developers said they spent 41 hours a week or more.

Fig.40 (Q57) What do you think about the skills you develop in the OS/FS community?

Respondents were asked whether their involvement in the OSS/FS community was purely because of their own interest or because it would help them in their work. 55.6% of the respondents said it was a core skill for their work, while 30.1% said that although it was not a core skill it was helpful to them in their work. Only a few respondents (14.3%) said it was not helpful to them in their work.

Fig.41 (Q58) Do you think that self-organized learning in the OS/FS community is more efficient than the more formalized ways to learn in universities or companies?

70.7% of the respondents said that learning in the OSS/FS community was more efficient than learning at a university or company. Only 5.3% said it was less effective.

From these results, it can be said that learning in the OSS/FS community is efficient and helpful in work tasks, and that, for companies as well, OSS/FS development is an activity worthy of being given some kind of remuneration or compensation to developers.

Conclusions

First, most of the respondents were developers in Korea and Thailand. Developers in Asia began OSS/FS development later than their counterparts in Europe, the United States and Japan; most began developing around the year 2000 when they were approximately 23 years of age. As in the European, American and Japanese survey, almost all (98.5%) are male and, undoubtedly due to their youth, most are single (72.1%). Most (63.9%) have graduated from university, and although the number that have completed graduate school is lower than in Europe, the United States and Japan, they seem to be highly educated when compared with the average in Asian countries. In terms of occupation, many are in software-related fields, but 15.8% are students, and of these 10.5% are in information related areas. Only in Japan were there fewer students in information-related areas than in other areas.

In Asia as well, most developers made a distinction between open source software and free software. 42.1% considered themselves to be affiliated with open source and 25.6% with free software. These percentages were about the same as in Japan. However, in terms of licenses, GPL was the overwhelming favorite (64.7%). The results provided a picture of people with an insatiable thirst for knowledge.

No particular characteristics could be ascertained with regard to development time and development target, but a high percentage of respondents were developing network and web services, and reflecting this trend many developers had experience with PHP and SQL. Linux was the most often used development and desktop environment (approximately 70%), but 15-20% used Windows, a figure that was in between the values for Japan and the West.

A higher percentage of respondents than in Japan were participants in the global community (52.6%), and their English language ability tended to be higher, but this should not necessarily lead to the conclusion that the English language ability of Japanese is poor. The high level of participation in the global community in Asia may be due to the fact that a domestic community has not developed in Asian countries because Asian developers have only recently begun OSS/FS development.

Most respondents had acquired their knowledge of OSS/FS development through self-study (54.9%), but many (22.6) had also studied in university. One distinctive characteristic was that OSS/FS-related education was being conducted not only in information-related fields (9.8%) but in science and engineering-related fields as well (11.3%).

As in the European, American and Japanese surveys, the motivation for oneself and others to participate in OSS/FS development was primarily to acquire and share knowledge and skills. However, one distinctive characteristic of the Asian survey was that many developers felt that others were involved for fame or profit. Most developers traced the beginning of their involvement in OSS/FS development to email exchanges with authors, friends and people on mailing lists, and unlike Japan, few developers traced the beginning of their involvement to coding.

The percentages of developers who derived direct and indirect remuneration related to OSS/FS development were roughly half and half (direct 45.1% and indirect 50.4%), about the same as in Europe and the United States. Noteworthy was the fact that in most cases the source of the assistance was a university or school (18.8%). Although the percentage of developers receiving assistance was about the same as in the West, the degree of recognition by companies and schools was not as high in Asia as in the West, and developers wanted increased recognition of their development activities.

As questions posed only in the Asian survey, respondents were asked about the development of skills in the OSS/FS community. In response, developers said that they felt they could learn more efficiently in the community than through seminars and other existing learning techniques (70.7%). The content of learning was wide-ranging, including not only technical matters such as programming techniques (54.1%) but also interpersonal and communications skills, and most developers (85.7%) said these were helpful in actual work activities.

Overall, for many of the questions, the responses of Asian developers placed them in between their counterparts in the West and in Japan. Although unfortunately the number of responses was not adequate, due in part to insufficient publicity, the survey did seem to reveal the active trends on the part of developers in Asia.

The most noteworthy difference between the results of the Asian and Japanese surveys is the activity on the part of the academic organizations. Teaching at academic organizations is not limited to information-related areas; moreover, since these organizations actively provide assistance to developers, young OSS/FS developers are produced one after another. In the summary to the FLOSS-JP survey, the authors listed university education as one of the problems with OSS/FS development in Japan. A study should be made of university education in not only Europe and the United States but in Asia as well, and the results reflected in Japanese education. In addition, private sector assistance is low in both Japan and Asia, indicating that OSS/FS business is still in the developing stages. Many developers said that they would like to make a living from OSS/FS business, and the establishment of OSS/FS business in Asia will undoubtedly be the key to the advancement of OSS/FS development. To achieve this, increased awareness on the part of society of OSS/FS development and improved public assistance policies for OSS/FS development will be indispensable.

It is the hope of the authors that the findings of this survey will help promote OSS/FS development in Asia and improve the circumstances of OSS/FS developers.

Acknowledgment

The FLOSS-ASIA survey is conducted with kind cooperation of Asian Open Source communities. The authors would like to offer their thanks to those developers who submitted responses and helped publicize the survey. We appreciate their great effort, especially for asian language translation and announcement in local community:

We also appreciate the European FLOSS survey team for their consent to use the questions willingly and for sending additional questions about skills development.

*For inquiries regarding this survey, please contact:
Kazuo Hiyane, Jun Iio, Hiroyuki Shimizu
Information Technology Research Department
Mitsubishi Research Institute, Inc.
2-3-6 Otemachi, Chiyoda-ku, Tokyo 100-8141
Email: floss-jp-admin@mri.co.jp
TEL (+81) 3-3277-0750 FAX (+81) 3-3277-3473

	FLOSS-ASIA		FLOSS-JP		FLOSS-EU		FLOSS-US
	Mean	Median	Mean	Median	Mean	Median	Mean	Median
Starting year	1999.4	2000	1998.4	2000	1996.7	1998	1996.6	1999
Starting age	24.3	23	26.6	26	22.9	22	-	22
Current age	27.9	27	31.2	31	27.1	26	-	27