Software in Far East Asia - A Personal Perspective -

1
Software in Far East Asia - A Personal
Perspective -

• Kouichi Kishida
• Technical Director, SRA-KTL
• Tokyo, Japan
• k2_at_sra.co.jp
• http//www.sra.co.jp/people/k2/

2
Agenda

• This talk will not give a formal or systematic
  view.
• But a personal perspective based upon speakers
  experience in Japanese software industry and also
  various international activities.

3
Self Introduction

• I have entered into computing world in first as
  a technical translator / freelance programmer in
  1960 after dropping out from University of
  Tokyo.
• My major at the university was astrophysics, but
  Ive lost my interest in science.At first, I
  wanted to be a painter.
• Here is one of my works as a young abstract
  artist.

4
(No Transcript)
5
What Im now?

• After working for a few computer-related
  companies, Ive started a small software house
  called SRA in 1967.
• Now Ive half-retired from the business and
  directing SRA-KTL, a RD subsidiary of the
  group.
• And devoting most of my time for various
  volunteer activities in domestic and
  international software community.

6
My Volunteer Activities

• Fellow of JUS Japan Unix Society (1983 )
• Secretary General of SEA
• Software Engineers Association (1985 )
• Chairman of SMSG
• Software Maintenance Study Group (1990 )
• President of JASPIC Japan SPI Consortium (2000
  )
• Connection with China (1986 )
• Annual International Symposium/Workshop
• ICSE2006 in Shanghai
• Chair of Far East Experience Track

7
My Contact Points in China

• Prof. Dehua Ju (ASTI Co., Shanhai)
• IEEE Software Editorial Member
• Prof. Xichang Zhong (SEC of CAS, Beijing)
• Beat Bill Gates with set-top box OS named Hopen
• Prof. Kegang Hao (NW Univ, Xian)
• Former president of the University
• Prof. Huawong Chen (NUDT, Changsha)
• Designed of Chinese supercomputer
• Prof. Fuqing Yang (Pekin Univ, Beijing)
• Top lady in Chinese software community

8
ICSE2006 Far East Experience Papers

• Weve got 32 submissions from 12 countries (117
  co-authors in total)
• Accepted 9 papers for regular presentation and 10
  papers for poster session.
• The topics of these papers represent current
  concerns in the Far East.

9
Titles of 9 Accepted Papers

• Development Technique
• Reengineering Standalone C Legacy Systems into
  the J2EE Partition Distributed Environment
  (China)
• UML-Based Service Robot Software Development
  (Korea)
• Analysis of the Interaction between Practices for
  Introducing XP Effectively (Japan)
• Quality / Testing
• Experiments on Quality Evaluation of Embedded
  Software (Japan)
• Procurement of Enterprise Resource Planning
  Systems (Hong Kong)
• Detecting Low Usability Web Pages Using
  Quantitative Data of Users' Behavior (Japan)
• Software Process
• Experiences of Applying SPC techniques to
  Software Development Processes (Japan)
• BSR A Statistic-based Approach for Establishing
  and Refining Software Process Performance
  Baseline (China)
• Practical Approach to Development of SPI
  Activities in a Large Organization (Japan)

10
Titles of 10 Poster Papers

• Efficiency Analysis of Model-based Review in
  Actual Software Design (Japan)
• Overseas Development for a Major U.S. eCommerce
  Website (China)
• A Series of Development Methodologies for a
  Variety of Systems (Korea)
• A Research Framework for Empirical Software
  Engineering Environment Reflected from Software
  Industry Structure (Japan)
• Estimation of Project Success Using Bayesian
  Classifier (Japan)
• Trust in Software Outsourcing Relationships An
  Analysis of Vietnamese Practitioners Views
  (Australia)
• Improving Requirements Engineering Process for
  Complex System Development through RIM
  (Thailand)
• Software Practices in Five Asian Countries An
  Exploratory Study (Philippine)
• University and Industry in Synergy A Research
  Journey Towards Product Lines (Singapore)
• An Experimental Comparison of Four Test Suite
  Reduction Techniques (China)

11
Some Keywords Observed

• Software Process Improvement
• Outsourcing
• Embedded System Development
• Free/Open Source Software
• Government

12
Keyword 1 Software Process Improvement

• Bottom-Up Movement in Japan
• JSPW in late 1980s
• SEA-SPIN from 1996
• Translation of SEI TR24 25 (1998)
• Creation of JASPIC (2000)
• Top-Down Style in Korea and China
• Why ?

13
Strange Matching between SPI Paradigm and
Confucian Principle

• Confucius said
• If names are not rectified, then language will
  not be in accord with truth.
• If language is not in accord with truth, then
  things cannot be accomplished.
• If things cannot be accomplished, then rites
  and music will not flourish.
• If rites and music do not flourish, then the
  punishment will not be just.
• If punishment is not just, then the people will
  not know how to move their hands or feet.

14
Table for Interpretation

• Names
• Language
• Things
• Rites Music
• Punishment
• People
• Hand or Feet

• Concepts
• Process Model
• Project
• Method/Tools
• Management
• Programmers
• Development Activities

15
Pitfall in Beautiful Logic

• Principle of Confucian Management has a
  beautiful logical structure. But it seems to be
  easy to collapse.
• Look at the current status of bureaucracy and
  also political situation in those countries
  based upon Confucian principle (China, Japan,
  and Korea).

16
Quotation from Sorai Ogyiu (1666 - 1728)

• The fundamental concept of Tao (Confucianism)
  is too abstract and difficult to understand.
• So, legendary sacred kings invented
  Rites-and-Musics as examples, which people easy
  to understand and practice.
• We should only inherit this spirit and think
  about to make our own version of
  Rites-and-Musics.
• Is Watts Humphrey the Sacred King for us ?

17
Keyword 2 Outsourcing

• Software outsourcing seems to be easy.
• Large number of engineers in developing countries
  and they are eager to get business.
• Gap of labor costs.
• Success stories of Indian software companies.
• But in reality, there are several difficulties.
• Language and cultural barrier.
• Rapid change of specs.
• Needs for close interaction with customers.
• High liquidity of programmers.

18
Outsourcing Case of China-Japan

• Chinese software companies are now heavily
  relying on outsourcing business.
• They are eager to have strong connection with
  Japan.
• Needs for bridge engineers to paraphrase and fix
  changing user requirement.
• Many failure stories rather than success.

19
Outsourcing SRAs case

• Not for cost saving, but to develop prototypes
  for advanced software engineering tools.
• Project tracking system to trace troubles in
  development.
• Open Web New development environment using open
  source component tools.
• SOA system architecture prototype.

20
Problem of High Liquidity of Chinese Human
Resource

• Observed tendency of young Chinese software
  engineers
• Always looking for high salary.
• Also want to do something technically new.
• Switching job at every few months.
• So, it is necessary to have project-oriented
  management.

21
Keyword 3 Embedded System Development

• It is one of the hot topics in Japan and Korea
  because governments are eager to push.
• Growing demands for new real-time applications
  in telecommunication, vehicle control, etc.
• Needs for appropriate methodology for
  hardware-software co-design development.
• Too many troubles caused by ad-hoc technical
  management.

22
Two Interesting Movement in Japan

• TEF Testing Engineers Forum
• Active discussion over the mailing list
• Annual Symposium JaSST since 2003
• SESSAME Society of Embedded Software Skill
  Acquisition for Managers and Engineers
• Providing training materials and semimars
• Annual Workshop since 2003

23
Promotion of S/W Industry in Korea

• Embedded S/W Engineering Center (KIPA)
• Linux Engineering Center (KIPA)
• Qplus Embedded Software Toolkit (ETRI)
• Best Practice (MOST Korea Embedded S/W
  Industry Council)
• Support Program for Linux Migration Service
  Development using Linux for Public Organizations

24
Situation in China

• Chinese people are very much concerned about the
  certification of software testing engineers or
  organizations.
• Maybe because of historical tradition of
  examination system for selecting talents.
• So, seminar business is now growing up, and some
  Indian organizations are trying to enter this
  market.

25
Growing Concern aboutProduct Line Software
Engineering

• New approach for modularization, reuse, and
  domain engineering
• Driving force is
• Prof. KC Kang of Postech (Korea)
• Conducted a number of University-Industry joint
  projects in Korea and also in Japan
• SRA-KTL are cooperating in Japan.

26
PLSE and SRA-KTL

• We have been cooperating with Prof. Kang to
  produce an environment for embedded software
  development called ASADAL-OBJ.
• It is a combination of Prof.Kangs system called
  ASADAL and SRA-KTLs 3D Graphics library called
  Jun.
• http//selab.postech.ac.kr/ASADAL-Simple_Overview
  .pdf

27
Keyword 4Free / Open Source Software

• Market is still dominated by Microsoft Windows,
  but there are growing concerns and also growing
  market for Free/OSS.
• In Japan, there are many active grass-root
  movements such as FSIJ, etc., but international
  contribution from Japan is still low level.
• Government is now somehow pushing the movement.
• Many companies including SRA are now trying to
  make business but not yet successful.

28
Some Industry Statistics in Japan

• Linux Server Shipment
• 2004 68,420 servers (13.9 share)
• 2007 (estimation) 140,000 servers (20 share)
• Distribution for server market (2004)
• Red Hat Linux 53
• Miracle Linux 22
• Turbolinux 13

29
Eclipse Downloads in China
30
SRA and Free/OSS

• The first contact was a telephone call from
  Richard Stallman to me in 1987.
• The trigger was an interview article on Unix
  Review Magazine, which was set by Les Belady.
• I was just in the political trouble with
  government about SIGMA project and winning the
  battle with SONY News workstation.
• I had a meeting with Richard and started to
  support his GNU project.

31
SRA and Free/OSS (continued)

• We were the first commercial user of Unix in
  Japan. So, made a bid contribution to the
  community by providing many useful free tools.
• Two major product we are now providing are
• Jun Multimedia application library in Smalltalk
  and Java http//www.srainc.com/Jun/Main_e.h
  tm
• http//www.sra.co.jp/public/sra/technical/jun/li
  nk.html (Japanese)
• PostgreSQL especially multi-byte application
  features
• http//www.postgresql.org/
• http//www.postgresql.jp/ (Japanese)

32
Free / OSSDifferent Concerns of Government

• Japan Enterprise Server
• Mission Critical Server
• Korea Embedded Linux
• China Desktop Linux

33
Keyword 5 Government

• We had a working group discussion at IWFST last
  year.
• We discussed about What is the desirable
  government action?
• Here are my presentation slides

34
My Own Closed Encounterwith Government

• PPDS - 1975-80 (CASE Development)
• SMEF - 1981-85 (Unix Environment)
• Walking out from SIGMA (1986)
• CMM - Long debate about What is SPI?
• DynC First Experience with MOE

35
Basic Tendency ofGovernment Officials

• No Long-Term Policy
• Ad hoc style of decision making
• Weak interest in technology
• 3 Major Concerns
• Budget
• Chairs in satellite organizations
• Relation with big companys

36
Tentative Strategy in My Hands

• Make effective use of budget
• Use at least 10 for real technology development
• Strong push for some hidden (real) project behind
  the scene

37
Long Term Strategy

• Needs for Global Perspective
• Do not to be too much sensitive about
  international competition
• But to focus on global needs for technology
• Be careful about rapid evolution of social
  environment

38
What We Need Now

• Long term funding for
• - Very fundamental research in software
  engineering for future
• Targeting 5-10 years ahead
• Technical infrastructure for future society
• Conceptual foundation of software engineering in
  relation to sociology and/or linguistics

39
Neuraths Boat

• We are like sailors, who on the open sea, must
  reconstruct their boat,but are never able to
  start afresh from the bottom.

40
What Kind of Boat We Are Now on ?

• OK.
• Time flows in direction to the past on the same
  speed I hour per hour.
• Lets looking back into the history of human
  thinking.

41
The Book of Great Learning edited by Zhu-Xi
(16th century)

• If you want to conquer the world,
• at first you must govern your state properly.
• To govern a state,
• at first you must regulate your family.
• To regulate your family,
• at first you must cultivate yourself.
• To cultivate yourself,
• at first you must rectify your mind.

42
Strange Matching withCurrent SPI Paradigm

• World
• State
• Family
• Ones Self

• ISO Standard
• CMM/CMMI
• Team Software Process
• Personal Software Process

43
Great Learning(Continued)

• To rectify your mind,
• at first you must make your will sincere.
• To make your will sincere,
• at first you must expand your knowledge.
• To expand your knowledge,
• at first you must study various objects around
  you.
• Object Orientation!

44
What is Software Engineering?

• Is it a version of Confucianism adapted to
  computing ?
• If so, Far East Countries have a conceptual
  advantage !?
• But, Western philosophy has a same kind of frame
  of thinking also.
• For example Max Weber.

45
Who Is the Father of OO? Pointed out by
Prof.Reenskaug (Norway)

• Ideal bureaucracy of Max Weber
• (1) emphasis on form,
• (2) concept of hierarchy,
• (3) specialization of tasks,
• (4) specified sphere of competence, and
• (5) established norm of conduct for each layer .
  
• This is also the spirit of OO!

46
One Step Beyond

• Nakamoto Tominaga (1715 - 1746)
• Front runner of Kaitokudo scholars.
• A private school on philosophy in Osaka,18th
  century
• He was a son of the one of the founding members
  of the school.
• Died very young, wrote a number of philosophical
  notes, but most were lost, only two of them
  remain.
• Very good bibliographer, but rather
  controversial.
• His name was long time forgotten, re-evaluated as
  a unique thinker recently (200 years after his
  death).

47
Add-on Principle

• To authorize their own theory, people add-on some
  new features to attack weak-points in fore-going
  theories.
• Nakamoto observed this principle thorough his
  careful study of Buddhism scriptures and
  Confucian classics.
• It is a useful tool for bibliographical study of
  human discourses.
• It is also applicable when we look into various
  theories/methods in software engineering.

48
3 Characteristics of Human Discourses

• Trends of the time
• Ethnic taste
• Style of representation
• Nakamoto found these 3 characteristics in his
  study in history of oriental religion or
  philosophical discourses.
• It will be also useful when we look into
  software engineering theories or methods.

49
Taste of Discourse

• Nakamotos Observation
• Buddhism has a taste of magic, Confucianism has
  taste of rhetoric, and Japanese Shinto has a
  taste of simplicity
• My Observation
• Structured method has a flavor of abstract art,
  OO has a taste of poetical metaphor, and formal
  people prefer beauty of mathematical notation.
• It is Just a matter of taste. No strong
  relationship to the contents of discourses.

50
From Thinking to Practice

• Nakamoto died very young. His dying message in
  his last book Okina-no-Fumi was as follows
• - Tao (the Way) is the way of sincere
  practice, that's all.
• This message overrides Sorai. There is no
  sacred king. It is important to pursue your own
  way of sincere practice.

51
Thank you!