The World of Computer Software/World_Of_Computer_Software-02-386-Vol-2of3.iso/v/viscii11.zip/REPORT.TX8

Text File | 1993-01-14 | 210KB | 3,846 lines

           VIETNAMESE CHARACTER ENCODING STANDARDIZATION REPORT




          VISCII AND VIQR 1.1 CHARACTER ENCODING SPECIFICATIONS
























============================================================================




Bilingual Report
Viet-Std, September 1992


























       (c) Copyright 1992, Viet-Std.  All rights reserved.  Permission in
           hereby granted for limited use, duplication, and distribution,
           in whole or in part, provided that no modifications are made.
           All documentation is provided without express or implied warranty.



           The Viet-Std Group
           1212 Somerset Drive
           San Jose, California 95132
           U.S.A.




                              TABLE OF CONTENTS
                              ~~~~~~~~~~~~~~~~~



PREFACE ...................................................................  1

TH┐ NGÖ G╝I QU═ Vÿ TRONG NG└NH ╨IÄN TO┴N ..................................  3

A UNIFIED FRAMEWORK FOR VIETNAMESE INFORMATION PROCESSING .................  7

  Abstract ................................................................  7

  1. INTRODUCTION .........................................................  7

  2. REVIEW OF CURRENT CONVENTIONS ........................................ 10

  3. VISCII: 8-BIT ENCODING SPECIFICATION FOR VIETNAMESE .................. 11
     3.1 MOTIVATION ....................................................... 11
     3.2 ENCODING RATIONALE ............................................... 13

  4. VIQR: MNEMONIC ENCODING SPECIFICATION FOR VIETNAMESE ................. 16
     4.1 MOTIVATION ....................................................... 16
     4.2 QUOTED-READABLE SPECIFICATION (VIQR) ............................. 18
         4.2.1 Implicit Composition ....................................... 18
         4.2.2 Explicit Composition ....................................... 19
         4.2.3 Literal State .............................................. 19
         4.2.4 English State .............................................. 20
         4.2.5 Vietnamese State ........................................... 20
         4.2.6 Character Literals in English and Vietnamese States ........ 20
         4.2.7 Closure .................................................... 21

  5. SPECIFIC APPLICATIONS ................................................ 21
     5.1 ELECTRONIC MAIL OVER 7-BIT CHANNELS .............................. 21
     5.2 VIETNAMESE KEYBOARDING ........................................... 22
         5.2.1 Immediate Echo for Implicit Composition .................... 22
         5.2.2 Delayed Echo for Explicit Composition ...................... 23
     5.3 ADAPTING EXISTING VIETNAMESE APPLICATIONS ........................ 24

  6 SUMMARY & CONCLUSIONS ................................................. 24

  REFERENCES .............................................................. 24

  GLOSSARY OF TERMS ....................................................... 26

  APPENDIX A: Vietnamese Characters under VISCII and VIQR
              by Collating Order .......................................... 29

  APPENDIX B: Vietnamese Characters under VISCII and VIQR
              by Encoding Order ........................................... 30


                                        iii

MôT KHU╘N KHæ THÅNG NHäT CHO VIÄC X╝ L▌ D  KIÄN VIÄT NG  .................. 32

  T≤m L▀■c ................................................................ 32

  1. LûI GIòI THIÄU ....................................................... 32

  2. DUYÄT LÇI NH NG QUY ┐òC HIÄN THûI .................................... 33

  3. VISCII: QUY ╨ÿNH M├ 8-BIT CHO VIÄT NG  ............................... 37
     3.1 ╨ôNG L╣C  ........................................................ 37
     3.2 C┴C L▌ DO BIÄN MINH VIÄC M├ H╙A .................................. 37

  4. VIQR: QUY ╨ÿNH VIÄT NG  ╨ÜC-╨┐öC-TRONG-NGOâC ......................... 41
     4.1 ╨ôNG L╣C ......................................................... 41
     4.2 QUY ╨ÿNH "╨ÜC ╨┐öC-TRONG-NGOâC" (VIQR) ........................... 43
           4.2.1 PhΘp T╒o Chµ NgÑm ........................................ 43
           4.2.1 PhΘp T╒o Chµ Ch∩ ╨╕nh .................................... 42
           4.2.3 Tr╒ng Thßi NguyΩn D╒ng ................................... 44
           4.2.4 Tr╒ng Thßi Anh Ngµ ....................................... 45
           4.2.5 Tr╒ng Thßi Vi«t Ngµ ...................................... 45
           4.2.6 NguyΩn T± Trong Tr╒ng Thßi Anh Ngµ vα Vi«t Ngµ ........... 45
           4.2.7 K² t± Hoαn Cñu ........................................... 46

  5. C┴C ║NG D₧NG ╨âC BIÄT ................................................ 46
     5.1 ╨IÄN TH┐ CHUYîN QUA MÇCH 7-BIT ................................... 46
     5.2 ╨┴NH CH  VIÄT .................................................... 47
           5.2.1 Cßch T╒o H∞nh Lºp T╤c Trong PhΘp T╒o Chµ NgÑm ............ 47
           5.2.2 Cßch T╒o H∞nh Chºm Trong PhΘp T╒o Chµ Ch∩ ╨╕nh ........... 48
     5.2 HöP TH║C H╙A ║NG D₧NG VIÄT NG  HIÄN H└NH ......................... 49

  6. T╙M TüT & KèT LUçN ................................................... 49

  T└I LIÄU THAM KH─O ...................................................... 49

  THUçT NG  ANH-VIÄT ...................................................... 51

  Ph° L°c A: Mτu T± Vi«t Li«t KΩ Theo Th╤ T± Síp Chµ ...................... 58
  Ph° L°c B: Mτu T± Vi«t Li«t KΩ Theo Th╤ T± Mπ S» ........................ 59


ANNOUNCEMENT OF VISCII-COMPLIANT SOFTWARE APPLICATIONS .................... 61

  1. FOR UNIX & X-WINDOWS ................................................. 61

  2. FOR MS-WINDOWS ....................................................... 63

  3. FOR MS-DOS ........................................................... 64

  4. WORK IN PROGRESS ..................................................... 65

  5. PROPOSED PROJECTS .................................................... 65

                                       iv

  6. GETTING SOFTWARE ..................................................... 66
     6.1 Using Anonymous ftp .............................................. 66
     6.2 Using Modem or MailServer at Saigon.COM .......................... 67
     6.3 Using Modem to Phsys.COM ......................................... 68
     6.4 Using Post Office ................................................ 68

  7. COPYRIGHT ............................................................ 69

  8. ABOUT THE TRICHLOR GROUP ............................................. 69











































                                       v

                                   PREFACE



The Vietnamese Standardization Group (Viet-Std) was formed in the fall of 1989
to promote the standardization of Vietnamese character encoding and to monitor
ongoing work of international bodies in  this  regard.   The  group  has  been
working on designing and implementing a code table that can be integrated into
existing computing environments on many platforms.  In addition, the group has
contacted the two committees on multilingual character encoding---the  Unicode
Consortium and  the  International  Standardizations  Organization  (ISO)---to
request that the Vietnamese characters be encoded in a precomposed form in the
same manner as other Latin-based European languages.   All these efforts  will
be reported fully in a special issue to be published in the near future.  This
special report collects only works concerning the 7- and  8-bit  encodings  of
the Vietnamese alphabet.

    The first article is a cover letter  in  Vietnamese  that  summarizes  the
Viet-Std Group's work in  the  area  of  7-  and  8-bit  Vietnamese  character
encoding.

    The second article covers the Vietnamese character encoding  in  7  and  8
bits.   It reviews the pros and cons of current encoding schemes and discusses
the vital need to  integrate  into  existing  computing  environments  that  a
standard must address.  It then presents the Viet-Std's 8-bit proposal for the
Vietnamese Standard Code for Information Interchange (known as VISCII)  and  a
Vietnamese  Quoted-Readable  convention   (VIQR)   to   represent   Vietnamese
characters in 7-bit ASCII.   The article also  examines  some  guidelines  and
conventions in handling  Vietnamese  electronic  mail  over  7-bit  channnels,
Vietnamese keyboarding,  and adapting existing  Vietnamese  applications.   It
includes two appendices listing both VISCII and VIQR for reference purposes.

    The third article is the Vietnamese translation of the above to serve  the
general Vietnamese community.

    The last item included in this report is the  announcement  of  the  third
release of public-domain  Vietnamese  software  by  the  TriChlor  Group.   It
summarizes software packages either developed or enhanced by TriChlor  members
and other independent developers to run on the Unix, X-Windows,  DOS,  and MS-
Windows platforms.

    Since the release of version 1.0 in January  1992,  VISCII  has  undergone
only one change---swapping the  two  characters  ╒  (a  dot-below)  and  á  (O
tilde)--- to accommodate MS-Windows. To reflect this change,  the versions  of
VISCII and VIQR published in this report are called VISCII 1.1  and  VIQR  1.1
although VIQR was unchanged.   This report therefore supercedes  all  previous
publications concerning VISCII and VIQR.

    A brief history of the Vietnamese Standardization Group is in order.   The
group was born out of the Viet-Net electronic mailing  list,  which  comprised
members at companies and universities throughout the computing world.   At the

                                         1
                                                                         2
time,  Vietnamese software for publishing existed  commercially  only  on  the
personal computer platform,  but each package  was  limited  to  its  designed
function and could not be  easily intermixed with others. No standard encoding
existed.   Developers designed their own  schemes  to  perform  the  job  they
needed. Even Viet-Net itself had invented a 7-bit mnemonic writing  style  for
Vietnamese for use in e-mail.   For example, "N▀╛c Vi«t Nam" would be typed as
"Nu+o+'c Vie^.t Nam."  Everyone agreed that some form of  standardization  was
necessary so as to promote availability of and access to Vietnamese computing,
and Viet-Net provided the forum that electronically brought together  for  the
first time a large number of individuals committed to this  goal.   With  this
specific interest in mind,  Viet-Std was formed  by  Thαnh  Vσn  Nguy¡n,   and
standardization    discussions   moved   to   the    mailing    list    "Viet-
[e-mail]"  and  later   to   "[e-mail]."    Early
contributors included  C▀╢ng  Tñn  Nguy¡n,  TΓm  Nguy¡n,  NhΓn  TrÑn,  Randall
Atkinson,  Khoa T⌠n,  KhiΩm H░,  T▀╛c L▀╜ng,  and many others too numerous  to
acknowledge properly.   Later joining the group were C▀╜ng Minh B∙i,  H≈c ╨∞nh
Ng⌠,  and others.   This special report is a testimony to the success  of  the
group.   In addition, the group has contributed its expertise to international
organizations on matters relevant to Vietnamese encoding;  the details will be
reported in a separate issue.

    It should be mentioned that TriChlor Software (a  non-profit  experimental
group) was formed by  C▀╢ng  Tñn  Nguy¡n,  C▀╜ng  Minh  B∙i,  and  Tφn  LΩ  to
independently explore and implement any encoding scheme for Vietnamese to gain
real experience with the  pros  and  cons  in  each  scheme.   It  has  helped
integrate VISCII-compliant Vietnamese  into  popular  public  domain  software
products.

    It is our dream one day to be able to read, write, and exchange Vietnamese
data of a common format on any machine, any platform, and to take advantage of
all the processing tools that have been produced by the computing  world. That
dream,  once a pure exercise in imagination,  has today come many steps closer
to realization.



Viet-Std Vietnamese Standardization Group
California, USA
September 1992







         TH┐ NGÖ G╝I QU═-Vÿ TRONG NG└NH ╨IÄN-tO┴N  C┘NG TäT-C─ QU═-Vÿ

                    QUAN-T┬M  ╨èN VIÄC M├-H╙A CH  VIÄT-NAM




Kφnh th▀a quφ v╕:


Ch·ng t⌠i lα m╡t nh≤m chuyΩn viΩn Vi«t Nam ╖ hΣi ngo╒i  c╡ng  tßc  trong  tinh
thÑn v⌠ v° l■i ≡¼ theo d⌡i vα ≡≤ng g≤p ² ki¬n chuyΩn m⌠n v½ chµ Vi«t  Nam  v╛i
cßc Vi«n ╨╕nh Chuªn Tin H≈c hoúc cßc c⌠ng ty ≡i«n toßn c≤  tÑm  v≤c  qu»c  t¬.
Trong th╢i gian qua ch·ng t⌠i ≡π vºn ≡╡ng tφch c±c v╛i t▒ ch╤c Unicode vα Vi«n
╨╕nh Chuªn ISO ≡¼ h≈  tiΩu  chuªn  h≤a  b╡  Vi«t-t±-mπ  (Vietnamese  character
encoding) trong khu⌠n kh▒ b╡ mπ chµ qu»c t¬ 16-bit vα 32-bit trong chi½u h▀╛ng
c≤ l■i nhñt.  Ngoαi ra, ch·ng t⌠i ≡π nghiΩn c╤u vα ≡½ ngh╕ b╡ Vi«t-t±-mπ 7-bit
vα 8-bit ≡¼ giΣi quy¬t nhu cÑu phßt tri¼n vα trao ≡▒i nhu li«u hi«n  nay.   B╡
t±-mπ 8-bit c≤ tΩn  Anh  ngµ  lα  Vietnamese  Standard  Code  for  Information
Interchange hay g≈i tít lα VISCII ≡¼ phΓn bi«t v╛i cßc b╡  t±-mπ  khßc.   TiΩu
chuªn 7-bit ≡▀■c  g≈i  lα  Quy-tíc  ╨≈c-╨▀■c-Trong-Ngoúc  (Vietnamese  Quoted-
Readable Specification) hay g≈i tít lα VIQR.  ╨Γy lα ≡½ tαi chφnh mα ch·ng t⌠i
mu»n thΣo luºn v╛i quφ v╕ qua lß th╜ nαy.

    Tr▀╛c h¬t ch·ng t⌠i xin tr∞nh bαy v½ b╡ t±-mπ 8-bit VISCII.   ╨i¼m qua quß
tr∞nh phßt tri¼n ngαnh ≡i«n toßn d∙ng chµ Vi«t ╖ hΣi  ngo╒i,  ch·ng  t⌠i  nhºn
thñy hÑu nh▀ m▓i c⌠ng ty hay m▓i nh≤m t± ≡út ra cho m∞nh  m╡t  b╡  Vi«t-t±-mπ,
v╛i h« quΣ tñt nhiΩn lα nhu li«u kh⌠ng th¼ trao ≡▒i d¡ dαng  v╛i  nhau. Do  ≡≤
ch·ng t⌠i ≡π quy¬t ≡╕nh nghiΩn c╤u m╡t b╡ Vi«t-t±-mπ  tiΩu  chuªn  (Vietnamese
character encoding standard) nhóm giΣi quy¬t t∞nh tr╒ng nαy.

    Nh▀ quφ v╕ ≡π r⌡,  ≡a s» nhu li«u hi«n nay ≡▀■c vi¬t d±a trΩn n½n tΣng m▓i
mτu t± ≡▀■c mπ h≤a bóng 8-bit (1 byte).  V╛i 8-bit ch·ng ta c≤ th¼ mπ h≤a ≡▀■c
256 mτu t± hoúc tφn hi«u khßc nhau.   V∞ nhµng l² do c≤ tφnh cßch l╕ch s╪,  b╡
t±-mπ ASCII cⁿa Hoa k╧ d∙ng 128 mπ s» (code point or code value) ≡Ñu  tiΩn  ≡π
tr╖ thαnh tiΩu chuªn qu»c t¬. B╡ t±-mπ nαy bao g░m  32  tφn  hi«u  ≡i½u  khi¼n
(control character) c≤ mπ s» t╫ 0 ≡¬n 31,  vα 96 mπ s» c≥n l╒i dαnh cho m╡t s»
mτu t± La tinh, dñu chñm cΓu hoúc k² hi«u.  (╨¼ ti«n vi«c thΣo luºn, ch·ng t⌠i
t╒m d╕ch control character lα ki¼m-t±,  cßc chµ c≥n l╒i lα k²-t±.   ╨╤ng  trΩn
quan ≡i¼m ≡i«n toßn, ch·ng ta phΣi ≡╕nh nghεa thΩm mπ-t± lα bñt c╤ cßi g∞ ≡▀■c
t▀■ng tr▀ng bóng m╡t mπ s»;  m▓i mπ-t± t▀╜ng ╤ng v╛i m╡t mπ s» vα ng▀■c  l╒i.)
PhÑn c≥n l╒i (128 mπ s» t╫ 128 ≡¬n 255) th▀╢ng ≡▀■c quy ≡╕nh vα  s╪  d°ng  t∙y
theo nhu cÑu cⁿa m▓i qu»c gia hoúc t╫ng b╡ nhu li«u.   Nh▀ vºy ch·ng ta c≤ th¼
tiΩu chuªn h≤a 128 mπ s» nαy cho cßc mτu t± Vi«t Nam kh⌠ng nóm trong danh sßch
128 chµ ASCII.

    TrΩn th±c t¬ vi«c mπ h≤a chµ Vi«t lα cΣ m╡t vñn ≡½ ph╤c  t╒p.   Ngoαi  cßc
ph° Γm,  ch·ng ta c≤ 12 nguyΩn Γm chφnh (A ┼ ┬ E ╩ I O ╘ ┤ U ┐ Y) vα 60 nguyΩn
Γm khßc ≡▀■c t╒o thαnh do cßc nguyΩn Γm chφnh k¬t h■p v╛i 5  dñu  gi≈ng  (síc,
huy½n, h÷i, ngπ,  núng).   Nh▀ vºy chµ Vi«t c≤ tñt cΣ 144 nguyΩn Γm v╫a th▀╢ng
v╫a hoa.  Ch·ng ta c≤ th¼ k¼ ra hai ph▀╜ng phßp chφnh ≡¼ mπ h≤a nguyΩn Γm Vi«t

                                         3
                                                                         4
Nam nh▀ sau:
    1. M▓i nguyΩn Γm chφnh vα m▓i dñu gi≈ng ≡▀■c xem nh▀ cßc mπ-t± riΩng bi«t,
       nghεa lα ch∩ cÑn 29 mπ s» ≡¼ mπ h≤a 12 nguyΩn Γm chµ th▀╢ng,  12 nguyΩn
       Γm chµ hoa vα 5 dñu gi≈ng. Thφ d°, mτu t± └ ≡▀■c xem nh▀ g░m c≤ hai mπ-
       t± "A" vα mπ-t± "`".

    2. M▓i nguyΩn Γm,  vα dñu gi≈ng n¬u c≤,  ≡▀■c xem nh▀ m╡t mτu t± duy  nhñt
       (mπ-t±),  nghεa lα phΣi cÑn ≡¬n 144 mπ s» ≡¼ mπ h≤a tñt  cΣ  nguyΩn  Γm
       Vi«t.

    Ph▀╜ng phßp ≡Ñu tiΩn, c≥n ≡▀■c g≈i lα ph▀╜ng phßp dñu r╢i, ≡π ≡▀■c ßp d°ng
╖ nhi½u qu»c gia d∙ng chµ La-tinh c≤ dñu ph° (diacritical  mark).   Nh▀ng  quß
tr∞nh th╪ nghi«m cho thñy ph▀╜ng phßp nαy c≤ khuy¬t ≡i¼m l╛n lao v½ nhi½u  mút
nh▀ t»c ≡╡ x╪ l² giΣm s·t,  nhu cÑu v½ s╤c ch╤a (storage) vα b╡  nh╛  (memory)
gia tσng,  vi«c thΣo ch▀╜ng ph╤c t╒p,  kh⌠ng th¼ tφch h■p vαo m⌠i  tr▀╢ng  nhu
li«u vα c▀╜ng li«u hi«n hµu ...   V∞ nhµng l² do nαy,  ph▀╜ng phßp dñu r╢i hÑu
nh▀ kh⌠ng c≥n ≡▀■c s╪ d°ng trong k█ ngh« ≡i«n toßn hi«n nay.

    ╨¼ c≤ th¼ h╡i nhºp vαo n½n k█ ngh« ≡i«n toßn cⁿa th¬ gi╛i,  ch·ng ta  phΣi
chñp nhºn giΣi phßp th╤ hai,  nghεa lα phΣi mπ h≤a m╡t s» l▀■ng rñt l╛n mτu t±
Vi«t.   Tr╫ m╡t s» chµ Vi«t ≡π c≤ s╟n trong b╡ t±-mπ ASCII, ch·ng ta c≤ tñt cΣ
134 mτu t±  cÑn phΣi mπ h≤a trong khi ch∩ c≤ 128 mπ  s»  c≥n  tr»ng  mα  th⌠i.
Cßch giΣi quy¬t th⌠ng th▀╢ng lα t∞m cßch thay th¬ 6 mπ-t± ASCII nαo ≡≤ bóng  6
mτu t± Vi«t.   C≤ rñt nhi½u cßch nh▀ng cßch nαo c√ng vñp phΣi  m╡t  s»  khuy¬t
≡i¼m riΩng.

    Sau khi phΓn tφch k█ cαng vñn ≡½ vα c╤u xΘt  cΣ  chi½u  h▀╛ng  phßt  tri¼n
c▀╜ng li«u vα nhu li«u trong t▀╜ng lai,  nh≤m Viet-Std ≡π giΣi quy¬t bαi  toßn
nαy d±a trΩn cσn bΣn tri«t ≡¼ bΣo toαn 96 k² t± ASCII trong v∙ng G0  (mang  mπ
s» 32 ≡¬n 127).   Quy¬t ≡╕nh nαy ≡▀■c tr∞nh bαy k█ cαng trong bαi Anh  Ngµ  in
l╒i trong tºp nαy [xem tr. 6--22],  nh▀ng trong ph╒m vi lß th╜ nαy c≤ th¼ ≡▀■c
t≤m tít nh▀ sau: bóng cßch bΣo toαn 96 k² t± ASCII ch·ng ta c≤ th¼ s╪ d°ng hÑu
h¬t nhu li«u vα c▀╜ng li«u sΣn xuñt khíp n╜i trΩn th¬ gi╛i mα kh⌠ng  phΣi  ≡Ñu
t▀ nhΓn l±c vα tαi nguyΩn vαo vi«c ≡i½u ch∩nh hoúc bi¬n ≡▒i cho thφch h■p  v╛i
chµ Vi«t Nam,  c° th¼ nh▀ b╡ d╕ch (compiler),  khi¼n  h«  (operating  system),
khung X (X windows),  v.v...   Ch∩ c≤ cßch giΣi quy¬t nαy m╛i gi·p ch·ng ta s╪
d°ng ≡▀■c nhµng thαnh quΣ k█ thuºt m╛i mδ nhñt trΩn th¬ gi╛i.

    V╛i ph▀╜ng chΓm trΩn,  ch·ng t⌠i ≡π thay th¬ 6 ki¼m-t± ASCII trong v∙ng C0
bóng 6 mτu t± Vi«t Nam , , , ,  vα .   Ngoαi ra,  d±a trΩn b╡ t±-mπ tiΩu
chuªn 8859/Latin-1 dαnh cho cßc n▀╛c TΓy ┬u, ch·ng t⌠i c√ng quy¬t ≡╕nh duy tr∞
mπ s» cⁿa tñt cΣ chµ Vi«t ≡π c≤ s╟n trong tiΩu chuªn nαy.   Tñt cΣ chi ti¬t v½
b╡ t±-mπ 8-bit VISCII ≡▀■c tr∞nh bαy r⌡ rαng trong ch▀╜ng 3 cⁿa  bαi  Anh  ngµ
[xem tr. 8-11] vα bΣn d╕ch Vi«t ngµ [xem tr. 26--29].

    Sau ≡Γy ch·ng t⌠i xin s╜ l▀■c v½ Quy-≡╕nh ╨≈c-╨▀■c-Trong-Ngoúc,  VIQR. ╨Γy
lα quy luºt vi¬t chµ Vi«t Nam bóng b╡ t±-mπ 7-bit ASCII cⁿa Hoa-K╧. ╨π t╫  lΓu
c╡ng ≡░ng ng▀╢i Vi«t ╖ hΣi ngo╒i th▀╢ng hay s╪ d°ng h∞nh th╤c nαy ≡¼ trao  ≡▒i
≡i«n th▀ bóng ti¬ng Vi«t trΩn cßc mßy vi tφnh kh⌠ng c≤ chµ Vi«t Nam.  Theo quy
≡╕nh nαy, cßc dñu ≡▀■c ≡ßnh sau cßc nguyΩn Γm; dñu síc, huy½n, h÷i, ngπ,  núng
≡▀■c thay th¬ bóng cßc k² hi«u ASCII Hoa-K╧ c≤ d╒ng t▀╜ng t± lα "'", "`", "?",
"~", ".", dñu trσng (dñu ß) ≡▀■c thay bóng "(", dñu m√ bóng "^",  dñu m≤c (nh▀
                                                                         5
trong chµ ┤, ┐) bóng "+",  vα chµ ≡ ≡▀■c thay th¬ bóng "dd".   Ch╞ng h╒n,  hai
cΓu th╜ Ki½u cⁿa c° Nguy¡n Du:

           Trσm nσm trong c⌡i ng▀╢i ta
           Chµ tαi chµ m«nh khΘo lα ghΘt nhau

s¿ hi«n ra nh▀ sau khi vi¬t theo quy ≡╕nh VIQR:

           Tra(m na(m trong co~i ngu+o+`i ta
           Chu+~ ta`i chu+~ me^.nh khe'o la` ghe't nhau

    Tuy nhiΩn m╡t s» ng▀╢i l╒i thφch d∙ng cßc k² hi«u khßc,  ch╞ng h╒n nh▀ "*"
t▀■ng tr▀ng cho dñu m≤c,  "<" thay cho dñu trσng,  "\"  thay  cho  dñu  huy½n,
v.v...   Nh▀ng v╛i vai tr≥ lα m╡t tiΩu chuªn,  VIQR phΣi ñn ≡╕nh m╡t quy  luºt
duy nhñt vα t»i thi¼u ≡¼ lαm c╜ s╖ th»ng nhñt cho  vi«c  trao  ≡▒i  nhu  li«u,
nghεa lα m▓i dñu Vi«t Nam phΣi ≡▀■c t▀■ng tr▀ng bóng m╡t  vα  ch∩  m╡t  k²  t±
ASCII mα th⌠i.   V╛i s± s╪ d°ng r╡ng rπi bαn ≡ßnh chµ ASCII  trong  gi╛i  ≡i«n
toßn Vi«t Nam,  ch·ng t⌠i quy¬t ≡╕nh ch≈n quy ≡╕nh VIQR d±a  trΩn  nguyΩn  tíc
th±c d°ng lα d¡ ≡≈c vα d¡ nh╛ cho tuy«t ≡╒i ≡a s» ng▀╢i d∙ng.   Quy luºt  VIQR
≡▀■c m⌠ tΣ chi ti¬t trong ch▀╜ng 4 cⁿa tαi li«u Anh Ngµ vα Vi«t ngµ ≡φnh kΦm.

    Ti«n ≡Γy ch·ng t⌠i c√ng xin nhñn m╒nh lα b╡ Vi«t-t±-mπ 8-bit VISCII vα quy
≡╕nh 7-bit VIQR ≡π ≡▀■c ph▒ bi¬n r╡ng rπi trΩn cßc m╒ng th⌠ng  tin  ≡i«n  toßn
(computer network) ╖ Hoa K╧ vα cßc qu»c gia tiΩn ti¬n khßc trΩn toαn th¬ gi╛i.
Cßc chuyΩn viΩn ≡i«n toßn Vi«t Nam ╖ hΣi ngo╒i ≡π d∙ng nhµng  tiΩu  chuªn  nαy
≡¼ vi¬t  nhu  li«u  ╤ng-d°ng  (software  application)  hoúc  nhu-li«u  d°ng-c°
(software tool) cho khi¼n h« DOS vα Unix.  Nhµng nhu li«u nαy ≡π ≡▀■c cΣ ng▀╢i
vi¬t lτn ng▀╢i d∙ng tríc nghi«m trΩn cßc mßy nh▀ PC, AT, 386/486, workstation,
mainframe,  v.v... trong m╡t th╢i gian khß lΓu dαi vα th±c t¬ cho thñy tñt  cΣ
≡½u vºn hαnh t»t ≡⌐p v╛i dµ li«u Vi«t ngµ. Trong ph╒m vi lß th╜ nαy, ch·ng t⌠i
xin t≤m tít nhµng sΣn phªm sau ≡Γy:

    * ch▀╜ng tr∞nh bi¬n ≡▒i dµ li«u ╖ d╒ng 7-bit VIQR sang d╒ng  8-bit  VISCII
      vα ng▀■c l╒i.

    * ╤ng d°ng ≡i½u khi¼n bαn chµ Vi«t trong m⌠i tr▀╢ng Unix, DOS,  MS-Windows
      3.1.   ║ng d°ng nαy cho phΘp ng▀╢i d∙ng ≡ßnh chµ Vi«t trΩn bαn chµ ASCII
      Hoa K╧ nh▀ng mαn Σnh s¿ hi«n ra chµ Vi«t Nam.

    * ╤ng d°ng mαn Σnh Vi«t ch╒y trong khung X (X windows) hoúc  khung  MS-DOS
      (MS-DOS windows).

    * ╤ng d°ng th▀ tφn cho phΘp ≡≈c vα vi¬t th▀ bóng 8-bit VISCII nh▀ng s¿  t±
      ≡╡ng bi¬n ≡▒i th▀ thαnh d╒ng 7-bit VIQR khi l▀u trµ hoúc g╪i ≡i.

    * ╤ng d°ng in dµ li«u 8-bit VISCII trΩn cßc  mßy  in  Laser,  ma-trºn-≡i¼m
      (dot matrix), hoúc PostScript.

    * ╤ng d°ng vi¬t bαi (editor) d∙ng 8-bit VISCII.

    * cßc nhu-li«u d°ng-c° vα nhu li«u th▀ vi«n  (library)  th▀╢ng  d∙ng  trΩn
      khi¼n h« Unix, DOS.
                                                                         6
    * ╤ng d°ng bΣng tφnh (speadsheet) ch╒y trΩn Unix.

    * ╤ng d°ng s▀ ph╒m d∙ng ≡¼ ra cΓu ≡» hoúc ≡½ thi tríc nghi«m ti¬ng Vi«t.

    * ╤ng d°ng ki¼m soßt l▓i chφnh tΣ.

    * ╤ng d°ng tr≥ ch╜i.

    * cßc b╡ ph⌠ng chµ ╖ d╒ng ≡i¼m (bitmap fonts), TrueType, PostScript.

    Ngoαi ra c≥n nhi½u sΣn phªm nµa kh⌠ng ti«n k¼ ra ≡Γy.   Mu»n bi¬t chi ti¬t
xin xem th⌠ng-bßo v½ vi«c phßt hαnh nhu-li«u Vi«t Nam ≡■t  3  [tr. 41].   Hi«n
ch·ng t⌠i ≡ang phΓn c⌠ng thi¬t k¬ thΩm cßc b╡ ph⌠ng chµ (fonts) VISCII  ≡¼  s╪
d°ng trΩn cßc mßy in laser.  V∞ Viet-Std kh⌠ng c≤ tφnh cßch th▀╜ng m╒i, tñt cΣ
cßc c⌠ng tr∞nh nghiΩn c╤u vα sΣn phªm nhu li«u cⁿa nh≤m ≡½u ≡▀■c ph▒ bi¬n mi¡n
phφ.  M╡t s» c⌠ng ty nhu li«u vα t▒ ch╤c chuyΩn gia Vi«t Nam ╖ Hoa K╧ ≡π tuyΩn
b» ⁿng h╡ b╡ t±-mπ 8-bit VISCII vα quy-≡╕nh 7-bit  VIQR  nh▀  VNU,  TIèN,  H╡i
ChuyΩn Gia Vi«t Nam, v.v...

    B╡ t±-mπ VISCII vα quy ≡╕nh VIQR lα c⌠ng tr∞nh nghiΩn  c╤u  cⁿa  ≡⌠ng  ≡Σo
chuyΩn viΩn Vi«t Nam ╖ hΣi ngo╒i thu╡c nhi½u lπnh v±c khßc nhau vα ≡π trΣi qua
quß tr∞nh th╪ nghi«m khß lΓu dαi.   Ch·ng t⌠i s╟n sαng g╪i ≡¬n quφ v╕  m╡t  s»
nhu li«u cÑn thi¬t ≡¼ quφ v╕ c≤ th¼ t± m∞nh ch╒y  vα  quan  sßt  ▀u  ≡i¼m  cⁿa
VISCII vα VIQR m╡t cßch c° th¼.   Xin quφ v╕ nghiΩn c╤u k█ l▀▐ng tαi li«u ≡φnh
kΦm vα dαnh cho ch·ng t⌠i s± ⁿng h╡ cÑn thi¬t ≡¼ b╡ Vi«t-t±-mπ 8-bit VISCII vα
Quy ≡╕nh 7-bit VIQR tr╖ thαnh tiΩu chuªn chφnh th╤c cho chµ Vi«t Nam.

    TrΓn tr≈ng kφnh chαo quφ v╕ vα mong nhºn ≡▀■c ² ki¬n cⁿa quφ v╕ trong th╢i
gian ngín nhñt.




Nh≤m NghiΩn C╤u TiΩu Chuªn Ti¬ng Vi«t
California, USA
September 1992




          A UNIFIED FRAMEWORK FOR VIETNAMESE INFORMATION PROCESSING


                 Vietnamese Standardization Working Group {1}
                              September 1992 {2}



                                   Abstract

Increasing demand for Vietnamese electronic information  processing  has  seen
answer in a wide array of  Vietnamese-capable  applications.   The  inevitable
need for integration of Vietnamese into existing environments and the exchange
of data among them point to the  necessity  of  standardization.   This  paper
presents the strategic and pragmatic technical  considerations  that  must  go
into such a standard,  and reviews  existing  conventions/proposals  in  these
important contexts.  A full description of the Viet-Std proposal is presented,
including 1)  an  8-bit,  fully  precomposed  encoding  table  for  Vietnamese
Standard Code for Information  Interchange  (known  as  VISCII),  2)  a  7-bit
Vietnamese Quoted-Readable (known as VIQR) standard for data interchange  over
7-bit channels,  with a seamless interface to the 8-bit  encoding,  and  3)  a
keyboard user-interface specification that works transparently with both 1 and
2.   Together,  these provide a  truly  unified  framework  for  a  Vietnamese
information  processing   environment   with   simplicity,   efficiency,   and
straightforward integration.   The real-world construction of  this  framework
has proven quite successful in an  array  of  compliant  applications  from  a
number of group and  individual  developers  across  a  number  of  platforms,
including Unix and its variants,  the X window system,  MS-DOS,  Windows,  and
with ongoing work elsewhere.


1. INTRODUCTION
~~~~~~~~~~~~~~~

With the  growing  Vietnamese  population  abroad  and  the  proliferation  of
computer usage within Viet Nam,  the  Vietnamese  language  has  seen  rapidly
increasing  representation  in   electronic   information   processing.    The
concomitant growth in demand for Vietnamese-capable software has  resulted  in
successful launches of myriad vendors in the U.S. and elsewhere, mainly in the
area of Vietnamese word processing.  In addition, individual and group efforts
have also been productive in providing Vietnamese-language  users  with  high-
quality  public-domain  applications.   In  Viet  Nam,  centers  such  as  the
Institute of Informatics [1] have reported impressive progress on many fronts,
among which is the Vietnamization of standard software packages.

------------------------------------------------------------------------------
    {1} Postal address: Viet-Std,  1212 Somerset  Dr.,  San  Jose,  California
95132, USA. E-mail address:  [e-mail]

    {2} Reprinted December 3, 1992. This version 1.1 supersedes version 1.0 of
January 1992.   The  only  significant  difference  is  the  exchange  of  the
positions of ╒ (a dot-below) and á (O tilde) in the 8-bit table.

                                         7
                                                                         8
    All of the above illustrate two important points:  1)  There  are  growing
market demands for Vietnamese-capable processing engines,  and 2) There is  no
shortage of technical talent to fulfill those demands.  Unfortunately, therein
lies a large problem: most existing Vietnamese applications have been designed
to operate in the exclusive framework or environment of the developer, and all
are incompatible with one another.   As long  as  this  trend  continues,  the
application base for Vietnamese can never keep reasonable  pace  with  demand.
Users want to do more with Vietnamese than mere word processing, and to expect
one single vendor to provide all potential applications across  all  platforms
is to dream  the  impossible. Technicians  providing  these  applications  are
limited to the Vietnamese tools they must themselves learn  and  develop  from
the ground up.  Standardization is necessary.  Anyone who has had to deal with
the incompatibility between ASCII and EBCDIC can try to imagine a world  where
every machine is using a different character set,  and appreciate how  limited
that world would be in its application base and how  cumbersome  in  its  data
interchange.   A uniform framework will greatly benefit both the user and  the
technician alike.

    The proposal for any Vietnamese data  standardization  must  take  several
important points in the proper  contexts.   First  and  foremost,  since  this
discussion is geared toward existing 7- and 8-bit environments, the prime goal
is  straightforward  and  direct  integration  onto  current  platforms.   The
standard must work  here  and  now.   This  implies  the  use  of  precomposed
Vietnamese characters,  because the handling of floating diacritics will never
see full or simple support outside of specific contexts.  The standard must be
designed so as to take advantage of existing applications as much as possible.
The familiar "don't reinvent the wheel" rule is not only an advantage---but  a
necessity---if a meaningful application base  is  to  be  established  in  any
reasonable length of time.   Furthermore,  it is known that overall efficiency
both in time and space is greater in processing  precomposed  character  units
when compared with the floating-diacritic approach [2].   Floating  diacritics
therefore must be limited to only where they  are  necessary  and  inevitable,
such as in keyboard entry or 7-bit data transmission.   There is no reason  to
require  that  all  applications  must  deal   with   the   complexities   and
inefficiencies of floating diacritics, for example,  in 8-bit data processing,
storage, transmission, screen rendering, or printing.

    The second major context points to the pragmatic and  vital  consideration
of existing precedents set in the Vietnamese software  base.   Standardization
necessarily requires adaptation,  but it makes  little  sense  to  propose  to
change the world so significantly  that  the  inertia  against  large  changes
greatly delays adoption of the standard.   The trend towards 16-bit and  wider
data standards for multinational character sets has gained momentum  with  the
recent works of Unicode [3] and ISO 10646 [4].  However, the need for an 8-bit
Vietnamese standard is irreplaceable  until  these  new  standards  are  fully
supported and completely dominate the computing world.   An  8-bit  Vietnamese
standard must not ignore existing software precedents  so  that  it  can  gain
speedy acceptance before it becomes obsolete.

    Thirdly,  the standard must address the issue of user  interface;  if  not
defining it,  then at least consider its possible  effects  on  the  end-user.
This  relates  primarily  to  the  7-bit  keyboarding  and  representation  of
Vietnamese---in  both  instances  diacritics  are  necessarily  floating,  and
                                                                         9
represented mnemonically by existing 7-bit characters with similar appearance.
With keyboarding,  one must preserve where possible existing practices such as
that  defined  for  the  Viet-Net  mailing  list  and  the  Usenet   newsgroup
Soc.Culture.Vietnamese,  both with  members  worldwide.   For  7-bit  readable
representation,  the keyword is "readable."  The goals here are to maintain  a
short learning time and to promote a uniform  interface  so  that  it  is  not
necessary  for  a  user  to  re-learn  the  particulars  of   every   software
installation before being able to use it effectively.

    Finally,  to every extent possible,  the standard  must  stay  within  the
framework of international standards, e.g., ISO-8859/x [5], in order to ensure
compatibility with  existing  environments.   For  example,  this  goal  means
preservation of the ASCII encoding.   It should extend also  to  the  encoding
into the same 8859/Latin-1 slots those Vietnamese characters that are  already
defined, thus ensuring that 8859/Latin-1 keyboards will work transparently for
those Vietnamese characters.   However,  there are many standards requirements
that are  obsolete  from  a  practical  viewpoint.   For  example,  in  recent
Unicode/ISO-10646 decisions, the prohibition from use of the available control
character space---those with encodings between xx00h and xx1Fh,  except for C0
itself---was discarded on the grounds that it was a waste of  encoding  space.
As will be discussed later, the encoding of Vietnamese into the existing 8-bit
space presents some well-known trade-offs.   Where trade-offs are  made,  they
must be justified with good reason---pragmatic preferred over theoretical.

    These primary requirements are summarized as follows:

    R1. Straightforward and direct integration into existing platforms.

    R2. Ease of adaptation for existing software.

    R3. User-friendly mnemonic encoding scheme and interface.

    R4. Adherence to international standards.

    R5. Trade-offs made only on practical usage considerations and  with  good
        reason.

    In the following section we present a brief review of  the  strengths  and
weaknesses of different approaches to Vietnamese  encoding.   Section  3  will
describe the proposed 8-bit  encoding  table  in  detail.   A  quoted-readable
encoding scheme encompassing 7-bit data streams, including electronic mail and
keyboard input,  is presented in Section 4.   Finally,  Section 5 outlines the
particular  rules  and  conventions  relevant  in  some   application-specific
contexts.


2. REVIEW OF CURRENT CONVENTIONS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

A review of current conventions used by software vendors reveals one  distinct
feature: virtually all realize the strengths of  a  precomposed  encoding  and
adopt it as a primary requirement.   The complications arise from  a  familiar
fact: apart from the alphabetics already  available  in  the  ASCII  standard,
                                                                        10
Vietnamese requires an additional 134 unique characters.  Of these, 128 can be
coded in the C1 and G1 areas.  The allocation of the remaining 6 characters in
the lower C0 and G0 space is handled with differing approaches:

    A1. Encode into 6 of the "least-used" G0  characters  in  the  context  of
        Vietnamese data processing.

    A2. Encode into 6 slots of the National Replacement  Character  {3}  (NRC)
        set.

    A3. Drop 6  of  the  "least-used"  {4}  Vietnamese  characters,  typically
        accented capitals such as , , , , , and .

    A4. Map accented "y" combinations  into  corresponding  "i"  combinations,
        e.g., "k█ s▀" is replaced with "kε s▀."

    A5. Encode into the ASCII control space C0.

    Approaches A1 and A2 both satisfy the typical needs of the word processing
environments in which rarely used ASCII characters can be avoided, or employed
by font shifting.   However they both eliminate prospects for  integration  of
Vietnamese into existing ASCII environments where all graphic characters in G0
are needed.  A character that already serves one purpose cannot be re-used for
another.   First, it makes rendering of the needed G0 character incorrect,  as
it would now look like a Vietnamese character.   The frequency of  use  of  G0
characters in an integrated environment is far too high for this  conflict  to
be tolerable. Second,  while font shifting may be employed to remedy  this  in
some situations,  a more serious problem occurs when the Vietnamese  character
is needed.   The environment  would  typically  have  assigned  some  specific
meaning to  the  G0  character,  particularly  with  those  in  the  NRC  set.
Consider,  for example,  using the backslash character "\"  for  a  Vietnamese
character under Unix.   The backslash is used for many escape mechanisms under
Unix so that the Vietnamese character  cannot  simply  be  used  but  must  be
escaped in one way or another.   This is more than just an  inconvenience;  it
means data interchange is complicated by the fact that  the  escape  mechanism
will not be understood on another platform,  and data integrity has  thus  not
been preserved.   A standard  employing  this  approach  fails  at  its  basic
mission: to provide cross-platform transparency.   A similar case can be  made
for the other G0 characters.

    Both A3 and A4 propose to limit Vietnamese language data  in  one  way  or
another.  Most agree that elimination of some Vietnamese characters are simply
unacceptable;  indeed,  this point is so  fundamental  that  we  have  in  the
foregoing chosen to assume it as a technical requirement without  elaboration.
However,  it must be said that A4 is not  a  proposal  without  rationale.   A
school of thought exists that believes y's existing in words as a single vowel

------------------------------------------------------------------------------
    {3} This set contains 12 country-specific  characters  at  code  positions
corresponding to ASCII characters #, $, @, [, \, ], ^, `, {, |, }, ~.

    {4} Least-used because they (a) rarely begin words and  therefore  do  not
often get capitalized, and (b) appear in fewer words.
                                                                        11
should be mapped to corresponding i's,  as  their  pronunciations  are  indeed
identical.  The concept dates as far back as 1948 [6, 7].   However, it is not
the function of an encoding standard to settle a linguistic issue,  and  hence
A4 is also a bad choice.

    The immediate objection to A5 is primarily in data communication  channels
where many C0 characters are used as  data  control.   In  addition,  it  also
presents problems for integration into environments where some  C0  characters
are used in the keyboard interface and in data format controls, similar to the
problem facing A1 and A2. However,  as will be  discussed  further,  judicious
choice of the  6  C0  characters  to  be  used  has  in  practice  been  shown
successfully to avoid characters that are significant in  data  communication.
Furthermore, most data channels provide for clean transfer of binary data, and
there is no reason to worry that arbitrary data bits cannot be  employed  over
these binary routes.

    With those particular cases where C0 is used in  the  keyboard  interface,
judicious choice as well as remapping of keys  can  minimize  conflict.   Data
format control is application-specific but is typically scattered  in  C0  and
C1.   It is therefore a  universal  problem  for  integration  because  C1  is
necessarily densely encoded, but,  again,  conflict can be avoided by studying
significant applications.   Finally,  the choice can be made for 6  least-used
Vietnamese characters so that the probability of conflict is greatly reduced.

    It should be noted here that the foregoing discussion  has  subjected  the
alternatives to the requirements of integration into existing applications and
platforms,  as outlined in Section 1.   The importance of this goal cannot  be
overstated,  and it does present complications that result  in  the  following
Pragmatism Principle: it is obviously impossible to  define  a  standard  that
would operate seamlessly with all existing applications,  therefore  pragmatic
considerations must be made to make a standard workable in as  many  important
applications and on as many platforms as possible,  with emphasis on the  word
"workable."


3. VISCII: 8-BIT ENCODING SPECIFICATION FOR VIETNAMESE
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

3.1. MOTIVATION
~~~~~~~~~~~~~~~

The available body of evidence shows that  alternative  A5  described  in  the
previous section,  encoding into 6 of the  C0  characters,  has  the  greatest
chance of success in fulfilling the requirements outlined in Section  1.   The
choice of the 6 C0 codes and the 6 least-used Vietnamese  capital  letters  to
encode,  when made carefully,  greatly reduces the probability of conflict for
all  practical  purposes. Concerns  regarding  data  communications  are  well
addressed by avoiding C0 codes that are in fact often used for  data  control.
Indeed, data communication concerns are more applicable to C1 and G1 encoding;
a prominent example is electronic mail transfer  through  7-bit  gateways  and
mail agents.  Communication failure here has in most cases been due to the use
of the eighth bit and not because of C0 encoding.   In any event,  the  option
exists for data to be sent in some "binary" mode,  or to employ the Vietnamese
                                                                        12
   Table 1:  A sampler of possible C0 usage conflicts.  Codes selected
             for this standard proposal are noted with a "+".
  ======================================================================
  CODE  COMM. CTRL-  GENERAL  PRINTER(PC)   PC        Unix     vi (Unix)
  ----------------------------------------------------------------------
    0    NUL   @     C string                       strings
    1    SOH   A
   +2    STX   B                                             back screen
    3    ETX   C     INTR                  INTR      INTR      INTR
    4    EOT   D     EOF                              EOF    back tab
   +5    ENQ   E
   +6    ACK   F                                             forw.screen
    7    BEL   G     BEL         BEL        BEL       BEL
    8    BS    H     BS          BS         BS        BS        BS
    9    HT    I     HT          HT         HT        HT        HT
   10    LF    J     LF          LF         LF        LF        LF
   11    VT    K                 VT
   12    FF    L     FF          FF         FF                 redraw
   13    CR    M     CR          CR         CR        CR        CR
   14    SO    N             wide on(IBM)
   15    SI    O             comp.on(IBM)
   16    DLE   P                          Prt.on/off
   17    DC1   Q     XOFF      XOFF         XOFF      XOFF
   18    DC2   R             comp.off(IBM)            retype
   19    DC3   S     XON       XON          XON       XON
  +20    DC4   T             wide off(IBM)                     forw.tab
   21    NAK   U             clr buf(IBM)             kill     kill
   22    SYN   V                                      literal  literal
   23    ETB   W                                      werase   werase
   24    CAN   X                                      kill
  +25    EM    Y                                      suspend
   26    SUB   Z                            EOF       suspend
   27    ESC   [     ESC     ESC sequence   ESC       ESC        ESC
   28    FS    \                                      quit
   29    GS    ]  Telnet ESC
  +30    RS    ^
   31    US    _                           Windows
  =====================================================================



  Table 2:  Vietnamese-specific characters already present in 8859/Latin-1.
        +===============================================================+
        | 0 : 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : A : B : C : D : E : F |
   +----+---------------------------------------------------------------+
   | Cx | └ : ┴ : ┬ : ├ :   :   :   :   : ╚ : ╔ : ╩ :   : ╠ : ═ :  :  : |
   | Dx | ╨ :   : ╥ : ╙ : ╘ : á :   :   :   : ┘ : ┌ :   :   : ▌ :  :  : |
   | Ex | α : ß : Γ : π :   :   :   :   : Φ : Θ : Ω :   : ∞ : φ :  :  : |
   | Fx | ≡ :   : ≥ : ≤ : ⌠ : ⌡ :   :   :   : ∙ : · :   :   : ² :  :  : |
   +====================================================================+


                                                                        13
Quoted-Readable format to be described in Section 4.

    The overwhelming advantage of this approach is  that  it  is  readily  and
easily integrated into existing environments  without  many  of  the  problems
plaguing the other alternatives,  if they can at  all  be  integrated.   As  a
testimony to the approach's successful application,  this document itself  was
prepared using the TeX system under Unix.   The text source was edited  in  an
8-bit X terminal window using a minimally modified version  of  Elvis  {5},  a
public-domain 8-bit version of Unix's Vi text editor.   Both TeX  (a  document
preparation system) and Dvi2ps (a PostScript generator) readily  accepted  and
processed Vietnamese (8-bit) data transparently.

    Many other applications including a  spreadsheet,  various  text  viewers,
PostScript and dot-matrix printing, DOS's WordPerfect, Word,  PC Tools,  etc.,
have  been  tested  and  seen  to   operate   well   with   Vietnamese   text.
Modifications,  if any,  were primarily in making  these  applications  accept
8-bit data.   An educational  teaching  tool  for  Vietnamese  has  also  been
produced using the  C  programming  language  with  8-bit  Vietnamese  strings
embedded in the  source  code. With  increasing  system  internationalization,
applications and tools are being  made  8-bit  "clean,"  further  facilitating
integration of this Vietnamese encoding.

3.2. ENCODING RATIONALE
~~~~~~~~~~~~~~~~~~~~~~~

A basic requirement is to preserve the 7-bit  ASCII  graphic  characters  (G0)
layout,  since  the  emphasis  is  on  integration.   G0  was  therefore  left
unchanged.   For the 6 C0 characters,  we first lay out  the  code  space  and
consider typical usage, a sampler of which is in Table 1.  The codes selected,
STX (2), ENQ (5), ACK (6),  DC4 (20),  EM (25),  and RS (30) present the least
possible  problems  with  data  communication  and  significant   applications
considered.   The use of ACK,  for example,  is actually context-dependent. In
those protocols we have reviewed,  it is only considered a "control" character
outside of a data frame;  within a data frame it is transfered without special
interpretation.   To reduce the probability of conflict even  further,  the  6
least-often used Vietnamese capital letters, , , , , , and , are encoded
into these slots.

    The remaining task is to encode the other 128 Vietnamese  characters  into
the extended ASCII space (C1 and G1).   Since no unique international encoding
standard exists in this region,  the philosophy is to be as much  conservative
as possible so that in the worst case the user can still use all of the  lower
case Vietnamese letters.

    The encoding of C1 is less troublesome,  although in  application-specific
contexts it has been found that some C1 characters are employed  with  special
meanings.   A review of ongoing work on 8-bit mail  transport  standardization
indicates that C1 characters will be fully  supported  as  graphic  characters
without special interpretation.   Nevertheless it is prudent  to  encode  only

------------------------------------------------------------------------------
    {5} The modifications provided the keyboard interface described  in  later
sections.
                                                                        14
upper-case characters into the C1 space.

    For G1,  the aim is to accommodate the popular PC character set (code page
850) and to adhere, if possible, to the 8859/Latin-1 mapping where Vietnamese-
specific characters are already encoded.

    Experience  in  development  of  this  encoding  on  the  MS-DOS  platform
motivates the consideration of line-drawing glyphs in the  PC  character  set.
In many situations where  both  Vietnamese  and  line-drawing  characters  are
desirable but font switching is impossible,  the best we can do is to preserve
all the lower case Vietnamese characters and all the single-  and  double-line
drawing characters.   This means that code positions occupied by  single-  and
double-line drawing characters must be  populated  with  upper  case  letters.
With this provision,  the MS-DOS user can be supplied with either  code  pages
containing all Vietnamese glyphs or code pages where a number  of  upper  case
Vietnamese  characters  are  replaced  by  PC  line-drawing  characters.   For
existing applications,  the user can choose the code page most appropriate for
her purpose.   Where the code page with line-drawing characters must be  used,
the penalty from missing Vietnamese  characters  has  been  minimized  by  the
choice of the  infrequently  used  ones.   For  new  applications,  code  page
switching can easily be done on the fly, if it is desired.

    Compatibility  with  the  8859/Latin-1  standard  is   merely   for   user
friendliness and is not mandatory.  It is natural and reasonable for a user in
France to expect that the same keystrokes producing Θ on the screen for French
will do the same for Vietnamese.  The motivation for this compatibility is the
predominant and increasing availability of  8859/Latin-1  keyboards  and  font
sets,  e.g.,  Digital's VT-terminal series,  Xterm  keymaps,  and  Microsoft's
Windows.   Table 2 lists the subset of 8859/Latin-1 characters in G1 that  are
also Vietnamese {6}.   It can be concluded that  all  8859/Latin-1  text  that
contains characters mostly from G0 (ASCII) and this  table,  French  text  for
example, is highly readable in the Vietnamese environment.

    Finally,  certain characters in G1 are  not  renderable  in  a  number  of
applications such as character codes  160  (non-breaking  space  character  in
8859/Latin-1), 202 (non-breaking space character on Macintosh),  or 255.   The
list of potentially non-graphic characters in C1 and G1 can  be  quite  large:
nearly 30 characters in MS Windows 3.0 and roughly 25 characters in MS Windows
3.1.   These positions  must  be  populated  with  upper  case  characters  in
consistence with the above philosophy.   In applications where font  switching
is allowed and upper case characters are blocked out,  a solution is to supply
fonts in pair: a normal font and a capital font.   In the capital font all the
positions that should be filled with lower case characters are actually filled
with the corresponding upper case.   When a capital letter in the normal  font
cannot be rendered, the user simply switches to the corresponding capital font
and types in the corresponding lower case character.

    With the above guidelines,  the task is then  to  lay  out  the  remaining

------------------------------------------------------------------------------
    {6} Note that the <≡> in Table 2 is actually a  similar-looking  Icelandic
"edh" in 8859/Latin-1;  the Vietnamese rendering form is better  reflected  in
8859/Latin-2.
                                                                        15
        Table 3: VISCII 8-bit Encoding Standard Proposal for Vietnamese

    +======================================================================+
    |    ||  0 : 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : A : B : C : D : E : F |
    +======================================================================+
    | 0x || NUL:SOH:  :ETX:EOT:  :  :BEL:BS :HT :LF :VT :FF :CR :SO :SI |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 1x || DLE:DC1:DC2:DC3:  :NAK:SYN:ETB:CAN:  :SUB:ESC:FS :GS :  :US |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 2x || SP : ! : " : # : $ : % : & : ' : ( : ) : * : + : , : - : . : / |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 3x ||  0 : 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : : : ; : < : = : > : ? |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 4x ||  @ : A : B : C : D : E : F : G : H : I : J : K : L : M : N : O |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 5x ||  P : Q : R : S : T : U : V : W : X : Y : Z : [ : \ : ] : ^ : _ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 6x ||  ` : a : b : c : d : e : f : g : h : i : j : k : l : m : n : o |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 7x ||  p : q : r : s : t : u : v : w : x : y : z : { : | : } : ~ :DEL|
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 8x ||  Ç : ü : é : â : ä : à : å : ç : ê : ë : è : ï : î : ì : Ä : Å |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 9x ||  É : æ : Æ : ô : ö : ò : û : ù : ÿ : Ö : Ü : ¢ : £ : ¥ : ₧ : ƒ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Ax ||  á : í : ó : ú : ñ : Ñ : ª : º : ¿ : ⌐ : ¬ : ½ : ¼ : ¡ : « : » |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Bx ||  ░ : ▒ : ▓ : │ : ┤ : ╡ : ╢ : ╖ : ╕ : ╣ : ║ : ╗ : ╝ : ╜ : ╛ : ┐ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Cx ||  └ : ┴ : ┬ : ├ : ─ : ┼ : ╞ : ╟ : ╚ : ╔ : ╩ : ╦ : ╠ : ═ : ╬ : ╧ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Dx ||  ╨ : ╤ : ╥ : ╙ : ╘ : ╒ : ╓ : ╫ : ╪ : ┘ : ┌ : █ : ▄ : ▌ : ▐ : ▀ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Ex ||  α : ß : Γ : π : Σ : σ : µ : τ : Φ : Θ : Ω : δ : ∞ : φ : ε : ∩ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Fx ||  ≡ : ± : ≥ : ≤ : ⌠ : ⌡ : ÷ : ≈ : ° : ∙ : · : √ : ⁿ : ² : ■ :   |
    +======================================================================+


Vietnamese characters in some fashion, perhaps even arbitrary.   This has been
done in such a way so as  to  provide  some  degree  of  symmetry  simply  for
aesthetics.   It turns out that all the above guidelines  can  be  adhered  to
except for compatibility with the letter á (O tilde)  in  8859/Latin-1.   Note
that the Vietnamese collating order cannot in any case be preserved,  but this
is not a major issue since collation for non-ASCII characters is well accepted
to be a table-lookup problem.

    The preceding guidelines have resulted  in  the  VISCII  8-bit  Vietnamese
encoding proposal listed in Table 3.  It is intended to be a single table that
applies  to  Vietnamese   data   handling   including   storage,   processing,
transmission,  and font encoding.   This greatly simplifies  the  integration,
implementation,  and usage processes and is indeed one of the major  strengths
of the proposal.
                                                                        16

4. VIQR: MNEMONIC ENCODING SPECIFICATION FOR VIETNAMESE
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

4.1  MOTIVATION
~~~~~~~~~~~~~~~

While the 8-bit specification attempts to standardize Vietnamese  encoding  in
8-bit  environments,   much  remains  to  be  addressed  in  important   7-bit
environments such as electronic mail transport and other 7-bit data lines,  as
well as in keyboard entry applications  where  the  interface  for  generating
Vietnamese characters needs to be standardized.

    Transporting more than 128 unique symbols over 7-bit data channels is  not
a problem specific to the Vietnamese language.   Since its proposal  in  1982,
the Internet Simple Mail Transfer Protocol ("SMTP",  [8]) has seen unrelenting
efforts to extend it to accommodate 8-bit  and  wider-word  data  in  European
Latin scripts and Oriental ideographic characters  (see,  e.g.,  [9]).   While
clean 8-bit transport is highly desirable,  all mail gateways are not going to
be converted overnight.   For the foreseeable  future  there  is  a  need  for
unambiguous transport of Vietnamese text over existing 7-bit channels.

    Indeed there is an ad-hoc standard in use on the Viet-Net mailing list and
the Usenet newsgroup Soc.Culture.Vietnamese, where mnemonic use of appropriate
characters to follow a vowel proves to be quite readable;  for example,  "Vi«t
Nam" would be written as "Vie^.t   Nam". However,  this  is  troubled  by  the
ambiguity in the multiple roles played by the mnemonic diacritical marks;  for
example, does "tha?" mean "tha?" or "thΣ"?

    The Viet-Net convention is not  far  in  concept  from  a  quoted-readable
format proposed by K. Simonsen [10,  11] which  disambiguates  such  texts  by
specifying text states  at  both  the  character  and  character  set  levels.
Unfortunately,  in its attempt to provide a  universal  solution  to  mnemonic
encoding,  the proposal does not provide a good answer  for  Vietnamese  text.
First,  it restricts the use of mnemonics to the  83  invariant  ISO-646  [12]
graphic characters,  which  is  a  good  idea  in  principle,  but  sacrifices
readability in the process.   For example, the counter-intuitive mnemonics for
hook-above (dau hoi) and tilde (dau nga) are "2"  and  "?",  respectively,  in
order to avoid "~" itself,  which is not an invariant.   The wide availability
of ASCII keyboards to the great majority of Vietnamese users  makes  this  too
unreasonable a limitation in the context of Vietnamese processing.   It should
be noted that we are in fact  arguing  in  favor  of  "readability  for  most"
against "illegibility for all."  Furthermore,  with on  progress  on  keyboard
and display internationalization, e.g., in graphical window environments where
keyboard mapping and font switching are easily implemented,  this availability
is on the increase, further obsoleting the restriction.

    The greater difficulty is that the two-character fixed-length encoding {7}
cannot provide  a  readable  or  mnemonic  representation  of  all  Vietnamese

------------------------------------------------------------------------------
    {7} The convention  is  "&xy"  where  x  is  a  literal  character  and  y
represents some combining form.
                                                                        17
characters, in particular those with 2 diacritical marks.  The variable-length
mnemonics {8} have been extended to include  all  Vietnamese  characters,  but
this scheme is so cluttered with announcers and  delimiters  that  readability
and efficiency are near nil,  keeping in mind that diacritics are heavily used
in Vietnamese.   While machine data translators will have little trouble  with
any "mnemonic" scheme, one that is directly accessible to human users, who are
in many cases typing mail messages using 7-bit editors, needs to be more user-
friendly.   A Vietnamese user will not want to learn  or  remember  among  all
possible combinations that, say, "a5" stands for "í", nor will she like typing
sequences as long as "&_a('_"  for some letter in every word.

    To satisfy  the  readability  and  flexibility  requirements,  a  separate
specification is necessary.   It is better to adopt an approach like code-page
switching under ISO-2022 [13] to switch the text into  "Vietnamese"  mode  and
optimize encoding according to the language state.  Recently, van der Poel put
forth a mnemonic proposal [14] which emphasizes language-specific  conventions
for these rea This proposal provides a means to specify  the  language  state,
each with its own (efficient) encoding  method.   Its  strength  lies  in  the
flexible specification that conformant implementations "need not  be  able  to
display all of the character sets specified";  they have the option of stating
messages such as "undisplayable Greek appeared here" for unsupported languages
(for  a  more  precise  specification,  see  [14]).   This  allows  networking
communities to determine the best approach for encoding their  own  languages.
The VIQR convention is compatible with this  approach  and  should  easily  be
incorporated into this framework.

    The  specification  here  encompasses  all  data  streams  including  text
transfer,  file I/O,  and keyboard entry.   This principle has been the  major
reason for success in operating systems such as Unix, in which device-specific
details are hidden as much  as  possible  from  the  applications  programmer,
leaving a uniform interface above which tools such as common library  routines
can be shared.   Indeed  as  the  keyboard  example  above  has  implied,  the
characters actually typed by the user are often not different  from  the  text
data that is eventually stored or transmitted.   It is therefore desirable  to
provide a common base on  which  to  build  data  interpreters  for  all  data
streams, independent of the input source.   In actual implementation, this has
greatly facilitated development of the Vietnamese-capable software base.

    In addition,  the user stands to benefit tremendously from standardization
of keyboard entry.   One does not need to learn  a  different  keyboard  entry
technique for each different Vietnamese application.  If one standard keyboard
model is fully supported by all Vietnamese software,  a user familiar with the
standard can sit down and start typing Vietnamese immediately.   This standard
defines the minimum expected behavior from compliant software;  any additional
input techniques can of course be incorporated as a superset of  the  standard
behavior.  This is discussed further in Section 5.2 on Vietnamese keyboarding.


------------------------------------------------------------------------------
    {8} The convention is "&_xxxx_" where xxxx can be  an  arbitrary  mnemonic
sequence.
                                                                        18
4.2. QUOTED-READABLE SPECIFICATION (VIQR)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The mnemonic model from Viet-Net is fully employed in the specification.   The
Vietnamese QR comprises three major states: Literal, English,  and Vietnamese.
The Literal state is intended for completely transparent handling  of  literal
data (except of course for the  escape  sequences  into  and  out  of  Literal
state).   The English and Vietnamese states are  designed  for  mixed  use  of
English and Vietnamese, with each optimized in appearance as well as data size
for texts containing mostly English and Vietnamese,  respectively.   In either
state there exist methods for composing Vietnamese-specific characters,  using
a base vowel followed by one or two diacritics.

    We first introduce the concept of implicit and explicit composition,  then
discuss how they are used in each of the states.

4.2.1. Implicit Composition
~~~~~~~~~~~~~~~~~~~~~~~~~~~

Implicit composition is useful for  data  containing  a  large  percentage  of
Vietnamese characters.

    With implicit composition,  a sequence of a base vowel followed by one  or
two diacritical marks is combined into one Vietnamese letter as long as it  is
grammatically legal.  This is best illustrated by examples:

                        a^      --->      Γ
                        o+?     --->      ╖
                        ╜?      --->      ╖
                        Vie^.t  --->      Vi«t
                        ViΩ.t   --->      Vi«t
                        la'^n   --->      lß^n (kh⌠ng phΣi lñn)
                        lß^n    --->      lß^n (kh⌠ng phΣi lñn)

    Note in the last two example that the sequence "a^'" is not  grammatically
equivalent to "a'^" or "ß^".   In general a  modifier  ("(",  "^",  "+")  must
immediately follow the appropriate vowel in order to be combined.

    The special sequence dd is composed into ≡ ; DD, dD,  and Dd all represent
╨.

    The base vowels are: a, σ, Γ, e, Ω,  i,  o,  ⌠,  ╜,  u,  ▀,  y,  and their
corresponding capitals.   The encoding values are those listed in Table 3, the
8-bit VISCII proposed standard.

    The  diacritical  marks  are  represented  by  ASCII   characters   having
correspondingly similar appearances.   Table 4 lists the  7  ASCII  characters
used as mnemonic replacements for the Vietnamese diacritics;  the first  three
are modifiers, and the remaining five are tone marks.

                                                                        19
           Table 4:  ASCII Mnemonics for Vietnamese Diacritics

    ================================================================
    Diacritic   Char  ASCII Code          Dñu        Example
    ----------------------------------------------------------------
    breve        (    0x28, left paren    trσng      ba(n khoa(n
    circumflex   ^    0x5E, caret         m√         ho^m nay
    horn         +    0x2B, plus sign     m≤c        Qui Nho+n

    acute        '    0x27, apostrophe    síc        La'i Thie^u
    grave        `    0x60, backquote     huy½n      Bi`nh Du+o+ng
    hook above   ?    0x3F, question      h÷i        Thu?  DDu+'c
    tilde        ~    0x7E, tilde         ngπ        di~ va~ng
    dot below    .    0x2E, period        núng       ho.c ta^.p
    ================================================================


4.2.2. Explicit Composition
~~~~~~~~~~~~~~~~~~~~~~~~~~~

Explicit composition is associated with the concept  of  a  leading  character
which explicitly announces the composition.   The announcer character  is  the
backslash ("\", ASCII 0x5C), known here as <COM> {*}. The subsequent combining
characters are defined in the same way as those in implicit composition.  Thus
the examples given above would appear in explicit composition mode as:

             a^         --->    Γ
             o+?        --->    ╖
             Vie^.t     --->    Vi«t

    Explicit composition is useful for data containing mainly English text, as
well as  for  maintaining  real-time  compatibility  with  keyboard  character
events, as will be discussed in Section 5.2 on Vietnamese keyboarding.

    With the composition methods described,  we are now ready to  discuss  how
they are employed in each of the three states.   The state of the data  stream
is specified by the two character sequence <COM>x ,  where  "x"  is  specified
below.

4.2.3. Literal State
~~~~~~~~~~~~~~~~~~~~

The appearance of <COM>L or <COM>l in the data stream  initiates  the  Literal
state.   This state is intended  for  near-perfect  transparent  literal  data
transfer.  Neither implicit nor explicit composition is available here, nor is
the <COM> character special,  except when it is followed by  one  of  the  six

------------------------------------------------------------------------------
    {*} In cases of ambiguity the notation <...> will be used to indicate that
the whole sequence <...> represents a byte in  memory  or  storage;  i.e.,  it
corresponds to one and only one character.   For instance,  the word  TÅT  can
also be written as <T><Å><T>.
                                                                        20
characters l, L, v, V, m or M which initiates one of the three states {9}.

4.2.4. English State
~~~~~~~~~~~~~~~~~~~~

The sequence <COM>M or <COM>m  sets  the  data  stream  state  to  English. In
English state,  only explicit composition is supported.   This means  that  in
order to generate a Vietnamese letter,  the announcer character <COM> must  be
used.    A  "composition"  sequence  not  preceded  by  <COM>  will  be   left
uninterpreted.  Examples:

        \mD\u~ng, how are you?  --->    D√ng, how are you?
        \mKho\e? kh\o^ng?       --->    Khoδ kh⌠ng?

    As noted,  the sequence "you?" above was not converted into "yoⁿ"  because
no composition was specified.

4.2.5. Vietnamese State
~~~~~~~~~~~~~~~~~~~~~~~

The data stream state is set to Vietnamese when the sequence <COM>V or  <COM>v
is encountered.   In Vietnamese mode,  both explicit and implicit compositions
are in effect.   The  following  examples  assume  that  the  data  stream  is
initially in English state:

        \vChu+~ Vie^.t          --->    Chµ Vi«t
        \vCh\u+~ Vi\e^.t        --->    Chµ Vi«t
        Chu+~ \vVie^.t          --->    Chu+~ Vi«t

    The availability of implicit composition in Vietnamese state ensures  that
the text is not cluttered with unnecessary <COM>s,  as would be  the  case  in
Vietnamese text using explicit composition. Explicit composition  is  included
to maintain compatibility with the English state so that there is no  need  to
define additional meanings for  the  <COM>  sequences.   Also,  the  real-time
keyboard compatibility mentioned previously is also  available  in  Vietnamese
state through explicit composition.

4.2.6. Character Literals in English and Vietnamese States
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Consider the following example:

      \vDu~ng, how are you?    --->    D√ng, how are yoⁿ

    In this example,  the sequence "you?" was interpreted as "yoⁿ" because the
data stream was still in Vietnamese state.   Thus it is sometimes desirable to
suppress composition altogether without having to switch states.   The literal
property of the <COM> character conveniently  accomplishes  this.   In  either

------------------------------------------------------------------------------
    {9} To effect <COM>L,  <COM>M,  and <COM>V themselves,  it is necessary to
switch to either English or Vietnamese state and  use  the  Character  Literal
feature available there.
                                                                        21
Vietnamese or English state,  whenever <COM> is followed  by  a  non-combining
character c the result is the  literal  character  c  itself.   The  <COM>  is
discarded from the data stream.   To get the <COM>  character  literally,  use
<COM><COM>.  Consider the following examples:

                \vddi dda^u?       --->        ≡i ≡Γⁿ
                \vddi dda^u\?      --->        ≡i ≡Γu?
                \m\ddi v\o^?       --->        ≡i v▒
                \m\ddi v\o^\?      --->        ≡i v⌠?
                \\                 --->        \
                \\V                --->        \V
                \\M                --->        \M
                \\L                --->        \L

4.2.7. Closure
~~~~~~~~~~~~~~

The data stream supports another special character used to  generate  explicit
closure.   The closure character is CTRL-A (ASCII 0x01),  known here as <CLS>.
When <CLS> is encountered in the data stream,  it immediately  terminates  any
ongoing composition sequence.  The <CLS> itself is always discarded, unless it
appears in the literal sequence <COM><CLS>.

    Explicit closure is useful in real-time  character  applications  such  as
keyboard entry,  when it is necessary to specify that a  composition  sequence
has in fact ended and the input engine should not stay hanging  and  wait  for
more data.


5. SPECIFIC APPLICATIONS
~~~~~~~~~~~~~~~~~~~~~~~~

This section outlines application-specific  guidelines  and  conventions  that
have evolved in the software development community.   It is intended to  be  a
live and growing documentation of  such  discussions  as  more  experience  is
gathered.   Readers are  welcome  to  participate  in  these  discussions  and
contribute to the development of these guidelines in particular,  and  to  the
standards in general.

5.1  ELECTRONIC MAIL OVER 7-BIT CHANNELS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Many of the available channels for electronic mail currently still enforce the
7-bit limitation.   The 8-bit character set defined in  Section  3  cannot  be
transported verbatim over these channels.   VIQR plays an important role here,
as it provides for 7-bit transport of Vietnamese text  without  the  ambiguity
problem of deciding what to do with the double usage of a
diacritical/punctuation mark,  e.g.,  the hook-above or  question  mark,  "?".
Because of the 7-bit nature of these communications channels, mail agents will
typically not encounter those Vietnamese-specific base vowels that are encoded
in the G1 area, namely: σ, ┼, Γ, ┬, Ω, ╩, ⌠, ╘, ╜, ┤, ▀, and ┐. However,  mail
agents designed to work with 8-bit channels are still expected to  handle  the
occurrence of these characters according  to  the  complete  VIQR,  namely  to
                                                                        22
combine base vowels and diacritical marks as appropriate, for example:

                    σ'     --->      í

    In order to  be  correctly  interpreted,  electronic  mail  messages  must
explicitly set the language state either in the headers  or  text  body.   One
cannot assume what state the receiving input engine is in at the start of  the
message,  since messages are not always read in message units,  e.g.,  when  a
file containing multiple mail messages is scanned.

    Furthermore,  if a language state specification (\L,  \V or \M) is present
in a mail message,  it is highly recommended  that  the  message  end  in  the
Literal state.   This helps applications reading multiple mail messages in one
data stream,  such as a terminal  application.   It  is  useful  because  mail
headers do not adhere to the VIQR,  and they are more adversely affected  when
interpreted in non-Literal states.

5.2. VIETNAMESE KEYBOARDING
~~~~~~~~~~~~~~~~~~~~~~~~~~~

Keyboards are becoming increasingly internationalized.   As mentioned  in  the
8-bit specification,  this is  the  major  reason  for  using  the  same  code
positions for those Vietnamese characters already present in ISO 8859/Latin-1.
A Vietnamese keyboard driver designed to work in  the  7-bit-only  environment
can assume that it will not encounter Vietnamese base vowels residing  in  G1.
Keyboard drivers for the 8-bit environments, like 8-bit electronic mail agents
(Section 5.1),  must be prepared to accept any  base  vowel,  including  those
encoded in G1.

    The real-time  echoing  behavior  of  keyboard  input  during  composition
requires further specification.   The options are to report the character only
after the composition sequence has finished,  or to  report  all  intermediate
forms and backspacing over them.  Each has its own useful context as described
below.

5.2.1.  Immediate Echo for Implicit Composition
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Implicit composition is designed to be convenient for a user  processing  data
that is mostly Vietnamese.   As such it is desirable for the keyboarding  user
to get immediate  feedback  on  typed  keys. With  implicit  composition,  the
keyboard works in immediate-echo mode.   Keypresses immediately  generate  key
events.   If a character is subsequently composed with a diacritical  mark,  a
backspace (typically BS,  ASCII 0x08) is sent followed  by  the  new  composed
character.   This cycle continues as  long  as  composition  is  possible. The
sequence of events for the key sequence "a^'n" under immediate echo is:

       1. user types a, a is sent to the application
       2. user types ^, BS and Γ are sent
       3. user types ', BS and ñ are sent
       4. user types n, the single key n is sent

    The actual backspace character code may  vary  depending  on  the  system,
                                                                        23
application,  and user  settings.   The  keyboard  interface  should  use  the
appropriate code,  and/or allow the user to specify  the  preferred  backspace
character.

5.2.2. Delayed Echo for Explicit Composition
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

When a composition sequence is started,  the keyboard interface must not  send
any key events to the application expecting keyboard input until the  sequence
is terminated.   Composition may  end  either  naturally  when  the  interface
receives a character that cannot be composed into the sequence,  or  when  the
closure character <CLS> is received.   A single key  event  for  the  composed
character is then sent to the application above.   Subsequent  processing  can
proceed naturally.   Consider what happens when the user  types  the  sequence
"\a^'n"  under delayed echo:

       1. user types \, no key is sent to the application
       2. user types a, no key is sent
       3. user types ^, no key is sent
       4. user types ', the single key ñ is sent
       5. user types n, the single key n is sent

    Or an example involving closure, t\o+<CLS>:

       1. user types t, the key t  is sent
       2. user types \, no key is sent
       3. user types o, no key is sent
       4. user types +, no key is sent
       5. user types CTRL-A, the single key ╜ is sent

    Note that without the closure key the keyboard interface  would  still  be
left hanging after the "+" key has been pressed,  because the user  can  still
enter a tone mark as part of the composition sequence.

    This delayed-echo behavior for explicit composition is designed to  ensure
compatibility  with  applications  expecting  single  key  events   for   each
character,  particularly in the English state where only explicit  composition
is available.

    While  it  is  certainly  possible  to  have  immediate-echo  in  explicit
composition or delayed-echo in implicit composition,  these  options  are  not
useful and serve only to confuse the user learning how  to  use  a  Vietnamese
keyboard.   It is therefore simplest to associate delayed-echo  with  explicit
composition, and immediate-echo with implicit composition.  These options make
natural sense.

    This standard defines the minimal  "look-and-feel"  behavior  a  user  can
expect from a compliant Vietnamese software package.  A standardized interface
decreases the required learning time for each new application.   This standard
does not preclude other input mechanisms to improve  user-friendliness,  e.g.,
intelligent menu-driven diacritics,  or  to  assist  in  speed  typing,  e.g.,
through the use of CONTROL or FUNCTION keys.   Any  enhancement  in  compliant
applications is a bonus for the user,  so long as  such  enhancements  do  not
                                                                        24
adversely conflict with the minimum expected behavior described here.

5.3. ADAPTING EXISTING VIETNAMESE APPLICATIONS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

A realistic approach to  standardization  provides  for  the  inertia  against
change in existing software applications.   While it  is  desirable  that  the
standard 8-bit encoding described here  be  fully  supported,  an  alternative
exists which is more amenable to  rapid  adoption.   All  applications  should
provide a means for importing and exporting  data  encoded  using  the  VISCII
8-bit encoding table.  At the same time, the VIQR keyboard interface should be
implemented,  at least as an optional entry method.   Such  moves  are  highly
desirable both for the user and the vendor alike.   The user will be  able  to
use the software immediately because of the  uniform  keyboard  interface,  as
well as process the same data  in  different  applications  and  on  different
platforms,  with increased productivity and interactivity among  users.   This
ease of use means greater acceptance and  a  correspondingly  larger  customer
base for the vendor.


6. SUMMARY & CONCLUSIONS
~~~~~~~~~~~~~~~~~~~~~~~~

This  paper  has  presented  a  proposal  for  standardization  of  Vietnamese
information  processing.    A  case  has  been  made  for  the  necessity   of
standardization;  we hope to have encouraged vendors and users  of  Vietnamese
alike to work together toward this goal to benefit everyone involved.  Various
encoding approaches were discussed,  leading to the choice of the VISCII 8-bit
encoding proposal.   A single encoding table was presented that has been shown
in actual practice to work well for Vietnamese including editing,  processing,
storage, transfer, font encoding, and printing.  Where 8-bit data handling was
not available or reliable,  e.g.,  electronic mail transport,  the  Vietnamese
Quote-Readable specification (VIQR)  was  introduced  to  provide  a  seamless
filtering gateway.   VIQR was defined to be input-source-independent and hence
has been designed to be applicable to Vietnamese keyboard  input  as  well  as
machine data filters.   All of this was shown to  have  been  integrated  into
existing  environments  facilitating   the   use   of   existing   tools   and
applications---a   major   strength   of   the   encoding.    Finally,   these
specifications have been linked together seamlessly to include every point  in
the input-process/transfer-output cycle of data handling  and  provide  for  a
truly unified framework for Vietnamese information processing.


References
~~~~~~~~~~

[1]  B╒ch H▀ng Khang.  "Institute of Informatics."  Hα N╡i,
     Vi«t Nam, Feb. 1991.

[2]  B. Jerman-Blazic, "Will the Multi-octet Standard Character Set
     Code Solve the World Coding Problems for Information Interchange?"
     Computer Standards & Interfaces, vol. 8, pages 127-136, 1988.

                                                                        25
[3]  The Unicode Consortium.  The Unicode Standard: Worldwide
     Character Encoding Version 1.0.  Addison-Wesley, Reading, MA,
     first edition, Oct. 1991.

[4]  ISO Technical Committee, "Universal Multiple-Octet Coded
     Character Set (UCS), ISO/IEC DIS 10646-1.2," Draft standard,
     International Organization for Standardization, 1992.

[5]  International Organization for Standardization, ISO 8859/x:
     8-bit Internatonal Code Sets.  ISO, 1977.

[6]  Famjxuaen Thais.  Vi«t Ngµ CΣi Cßch.  T╤ HΣi, Hα N╡i,
     Vi«t Nam, Mar. 1948.

[7]  Ph╒m XuΓn Thßi.  Chµ Vi«t H■p Lφ, Tφn-╨╤c Th▀ Xπ,
     Sαi G≥n, Vi«t Nam, Apr. 1958.

[8]  J. Postel, "Simple Mail Transfer Protocol," RFC 822, USC
     Information Sciences Institute, Aug. 1982.

[9]  J. C. Klensin et al., "SMTP Extensions for Transport of
     Text-Based Messages Containing 8-bit Characters," Internet draft,
     Massachusetts Institute of Technology, July 1991.

[10] K. Simonsen, "Character Mnemonics & Character Sets," Internet
     Draft, Danish Unix Users Group, Jan. 1992.

[11] K. Simonsen, "Mnemonic Text Format," Internet draft, Danish
     Unix Users Group, Aug. 1991.

[12] International Organization for Standardization.  ISO 646:
     7-bit Coded Character Set for Information Interchange.  ISO,
     third edition, 1991.

[13] International Organization for Standardization.  ISO 2022:
     7-bit and 8-bit Coded Character Sets--Code Extension Techniques.
     ISO, third edition, 1986.

[14] E. M. van der Poel, "Multilingual Character Encoding for Internet
     Messages," Internet draft, Software Research Associates,
     Japan, Jan. 1992.

[15] IBM.  System/370 Reference Summary-GX20-1850-5, sixth edition, 1984.

[16] C.E. Mackenzie.  Coded-Character Sets: History and Development.
     Addison-Wesley, Reading, MA, 1980.

[17] D.E. Knuth.  The TeXbook.  Addison-Wesley, Reading, MA, 1984.


                                                                        26
Glossary of Terms
~~~~~~~~~~~~~~~~~

Announcer: A character or sequence of characters appearing in  the  data  that
    signifies the start of some special sequence.   In this text, it announces
    a Vietnamese composition sequence.

ASCII:  American Standard Code for Information  Interchange,  a  128-character
    code  used  almost  universally  by   computers   for   representing   and
    transmitting characters data,  in which each character  corresponds  to  a
    decimal number between 0 and 127. Eight- or nine-bit codes  of  which  the
    first 128 characters correspond to ASCII are called  Extended  ASCII;  the
    additional characters are used to provide  graphic  characters  for  roman
    alphabets with diacritics,  non-roman alphabets,  special screen  effects,
    etc.

Base Vowel: In this text, the unaccented Vietnamese vowels: a, σ, Γ, e, Ω,  i,
    o, ⌠, ╜, u, ▀, y (and their capitals).  Contrast this with Vowel.

C0 Space: "Control characters" at code positions with hex  values  00  through
    1F.

C1 Space: "Control characters" at code positions with hex  values  80  through
    9F.

Code: In data communication,  the numeric or  internal  representation  for  a
    character, e.g., in ASCII.

Code Page: Name used to denote glyph sets on the IBM PC. Abbreviated as CP. CP
    850 is the multilingual code page,  CP860 is for Portugal,  CP863  is  for
    French Canada, CP865 is for Norway.

Control Character: An ASCII character  in  the  range  0  to  31,  plus  ASCII
    character 127,  contrasted with the printable,  or graphic,  characters in
    the range 32 to 126. It is produced on an ASCII terminal by  holding  down
    the CTRL key and typing the desired character.

EBCDIC: Extended Binary Coded Decimal  Interchange  Code. The  character  code
    used on IBM mainframes.  Not covered by any formal standards but described
    definitively in [15] and discussed at length in [16].

Floating Diacritics: A multiple-unit encoding  approach  for  Vietnamese  that
    treats the vowel and its diacritics as separate units. The diacritics  may
    either precede or follow the vowel,  or even the word. Contrast this  with
    Precomposed Character.

Glyph: The physical appearance of a character as displayed on  the  screen  or
    printed on paper.

G0 Space: "Graphic characters" at code positions with hex  values  20  through
    7F.

G1 Space: "Graphic characters" at code positions with hex  values  A0  through
                                                                        27
    FF.

ISO:   International   Organization   for   Standardization.     A   voluntary
    international  group  of  national  standards  organizations  that  issues
    standards in all areas, including computers,  information processing,  and
    character sets.

ISO 646: The standard 7-bit code set, equivalent to ASCII [12].

ISO Standard 8859: An ISO standard  specifying  a  series  of  8-bit  computer
    character sets that include characters from many languages. These  include
    ISO Latin Alphabets 1-9,  which cover most of the written languages  based
    on Roman letters, plus special character sets for Cyrillic, Greek, Arabic,
    and Hebrew [5].

ISO 8859/1: ISO Standard 8859 Latin Alphabet Number 1. Supports at  least  the
    following languages: Latin, Danish,  Dutch,  English,  Faeroese,  Finnish,
    French, German, Icelandic, Irish, Italian, Norwegian, Portuguese, Spanish,
    and Swedish [5].

ISO 2022 and ISO 4873: ISO standards for switching code pages [13].

ISO DIS 10646: The prospective 16- and  32-bit  Universal  Coded  Set,  (Draft
    International Standard) [4].

Latin: Referring to the Latin, or Roman, alphabet,  comprised of the letters A
    through Z, or to any alphabet based upon it.

MS-DOS: Microsoft's Disk Operating System  for  microcomputers  based  on  the
    Intel 80x86 family of CPU chips.

Modifier: A phonetic diacritical mark.   The Vietnamese modifiers  are:  breve
    (dñu trσng or dñu ß), circumflex (dñu m√, ^),  horn (dñu m≤c).

PC: Personal Computer.   In this text, the term PC refers to the entire IBM PC
    and PS/2 families and compatibles, which includes the AT,  286,  386,  and
    486 PC's.

PostScript: A page description language with  graphics  capabilities  designed
    for  electronic  printing. The  description  is  high-level  and   device-
    independent. PostScript is a trademark of Adobe Systems Incorporated.

Precomposed Characters: An encoding approach for Vietnamese  that  treats  all
    vowel  combinations  as  single   units. Contrast   this   with   Floating
    Diacritics.

TeX:  A computerized  typesetting  system  developed  by  Donald  Knuth  [17],
    providing  nearly  everything  needed  for  high-quality  typesetting   of
    mathematical notations as well as of ordinary text. TeX is a trademark  of
    the American Mathematical Society.

Tone Mark: A tonal diacritical  mark  that  indicates  the  tone/accent.   The
    Vietnamese tone marks are: acute (síc),  grave (huy½n),  hook above (h÷i),
                                                                        28
    tilde (ngπ), dot below (núng).

Unicode:  A  16-bit  multilingual  character  code  proposed  by  the  Unicode
    Consortium [3].

Unix: A popular operating system developed at AT&T Bell Laboratories and noted
    for its portability.

Usenet: A worldwide network available to users for sending messages (or  "news
    articles") that can be read and responded to by other users. Participating
    in Usenet is like subscribing to a  collection  of  electronic  magazines.
    These "magazines," called newsgroups,  are devoted to  particular  topics.
    The  "Soc.Culture.Vietnamese"  newsgroup  is  very  popular   among   both
    Vietnamese and non-Vietnamese worldwide.

Viet-Std: A non-profit group of overseas Vietnamese  and  other  professionals
    working on software & hardware  standards  for  the  Vietnamese  language.
    Members of the group exchange ideas via electronic mail and meetings.

Vowel: In this text,  a generic term applying to  all  Vietnamese  vowels  and
    their various combining forms, e.g., a, σ, and í. See Base Vowel.

                                                                        29
  Appendix A: Vietnamese Characters under VISCII and VIQR by Collating Order
+=================++==================++==================++=================+
|Chr: VIQR:VISCII || Chr: VIQR:VISCII || Chr: VIQR:VISCII || Chr: VIQR:VISCII|
+---:-----:-------++----:-----:-------++----:-----:-------++----:-----:------+
| A : A   :  065  ||  N : N   :  078  ||  a : a   :  097  ||  n : n   :  110 |
| ┴ : A'  :  193  ||  O : O   :  079  ||  ß : a'  :  225  ||  o : o   :  111 |
| └ : A`  :  192  ||  ╙ : O'  :  211  ||  α : a`  :  224  ||  ≤ : o'  :  243 |
| ─ : A?  :  196  ||  ╥ : O`  :  210  ||  Σ : a?  :  228  ||  ≥ : o`  :  242 |
| ├ : A~  :  195  ||  Ö : O?  :  153  ||  π : a~  :  227  ||  ÷ : o?  :  246 |
| Ç : A.  :  128  ||  ╒ : O~  :  213  ||  á : a.  :  160  ||  ⌡ : o~  :  245 |
| ┼ : A(  :  197  ||  Ü : O.  :  154  ||  σ : a(  :  229  ||  ≈ : o.  :  247 |
| ü : A(' :  129  ||  ╘ : O^  :  212  ||  í : a(' :  161  ||  ⌠ : o^  :  244 |
| é : A(` :  130  ||  Å : O^' :  143  ||  ó : a(` :  162  ||  » : o^' :  175 |
|  : A(? :  002  ||  É : O^` :  144  ||  ╞ : a(? :  198  ||  ░ : o^` :  176 |
|  : A(~ :  005  ||  æ : O^? :  145  ||  ╟ : a(~ :  199  ||  ▒ : o^? :  177 |
| â : A(. :  131  ||  Æ : O^~ :  146  ||  ú : a(. :  163  ||  ▓ : o^~ :  178 |
| ┬ : A^  :  194  ||  ô : O^. :  147  ||  Γ : a^  :  226  ||  ╡ : o^. :  181 |
| ä : A^' :  132  ||  ┤ : O+  :  180  ||  ñ : a^' :  164  ||  ╜ : o+  :  189 |
| à : A^` :  133  ||  ò : O+' :  149  ||  Ñ : a^` :  165  ||  ╛ : o+' :  190 |
| å : A^? :  134  ||  û : O+` :  150  ||  ª : a^? :  166  ||  ╢ : o+` :  182 |
|  : A^~ :  006  ||  ù : O+? :  151  ||  τ : a^~ :  231  ||  ╖ : o+? :  183 |
| ç : A^. :  135  ||  │ : O+~ :  179  ||  º : a^. :  167  ||  ▐ : o+~ :  222 |
| B : B   :  066  ||  ö : O+. :  148  ||  b : b   :  098  ||  ■ : o+. :  254 |
| C : C   :  067  ||  P : P   :  080  ||  c : c   :  099  ||  p : p   :  112 |
| D : D   :  068  ||  Q : Q   :  081  ||  d : d   :  100  ||  q : q   :  113 |
| ╨ : DD  :  208* ||  R : R   :  082  ||  ≡ : dd  :  240  ||  r : r   :  114 |
| E : E   :  069  ||  S : S   :  083  ||  e : e   :  101  ||  s : s   :  115 |
| ╔ : E'  :  201  ||  T : T   :  084  ||  Θ : e'  :  233  ||  t : t   :  116 |
| ╚ : E`  :  200  ||  U : U   :  085  ||  Φ : e`  :  232  ||  u : u   :  117 |
| ╦ : E?  :  203  ||  ┌ : U'  :  218  ||  δ : e?  :  235  ||  · : u'  :  250 |
| ê : E~  :  136  ||  ┘ : U`  :  217  ||  ¿ : e~  :  168  ||  ∙ : u`  :  249 |
| ë : E.  :  137  ||  £ : U?  :  156  ||  ⌐ : e.  :  169  ||  ⁿ : u?  :  252 |
| ╩ : E^  :  202  ||  ¥ : U~  :  157  ||  Ω : e^  :  234  ||  √ : u~  :  251 |
| è : E^' :  138  ||  ₧ : U.  :  158  ||  ¬ : e^' :  170  ||  ° : u.  :  248 |
| ï : E^` :  139  ||  ┐ : U+  :  191  ||  ½ : e^` :  171  ||  ▀ : u+  :  223 |
| î : E^? :  140  ||  ║ : U+' :  186  ||  ¼ : e^? :  172  ||  ╤ : u+' :  209 |
| ì : E^~ :  141  ||  ╗ : U+` :  187  ||  ¡ : e^~ :  173  ||  ╫ : u+` :  215 |
| Ä : E^. :  142  ||  ╝ : U+? :  188  ||  « : e^. :  174  ||  ╪ : u+? :  216 |
| F : F   :  070  ||    : U+~ :  255  ||  f : f   :  102  ||  µ : u+~ :  230 |
| G : G   :  071  ||  ╣ : U+. :  185  ||  g : g   :  103  ||  ± : u+. :  241 |
| H : H   :  072  ||  V : V   :  086  ||  h : h   :  104  ||  v : v   :  118 |
| I : I   :  073  ||  W : W   :  087  ||  i : i   :  105  ||  w : w   :  119 |
| ═ : I'  :  205  ||  X : X   :  088  ||  φ : i'  :  237  ||  x : x   :  120 |
| ╠ : I`  :  204  ||  Y : Y   :  089  ||  ∞ : i`  :  236  ||  y : y   :  121 |
| ¢ : I?  :  155  ||  ▌ : Y'  :  221  ||  ∩ : i?  :  239  ||  ² : y'  :  253 |
| ╬ : I~  :  206  ||  ƒ : Y`  :  159  ||  ε : i~  :  238  ||  ╧ : y`  :  207 |
| ÿ : I.  :  152  ||   : Y?  :  020  ||  ╕ : i.  :  184  ||  ╓ : y?  :  214 |
| J : J   :  074  ||   : Y~  :  025  ||  j : j   :  106  ||  █ : y~  :  219 |
| K : K   :  075  ||   : Y.  :  030  ||  k : k   :  107  ||  ▄ : y.  :  220 |
| L : L   :  076  ||  Z : Z   :  090  ||  l : l   :  108  ||  z : z   :  122 |
| M : M   :  077  ||    :     :       ||  m : m   :  109  ||    :     :      |
+=================++==================++==================++=================+

         * VIQR also allows "╨" to be represented by "Dd" or "dD".
                                                                        30
  Appendix B: Vietnamese Characters under VISCII and VIQR by Encoding Order
+===================================+========================================+
|VISCII:Chr:VIQR : Descriptive Name |VISCII:Chr:VIQR : Descriptive Name      |
+------:---:-----:------------------+------:---:-----:-----------------------+
|  002 :  : A(? :A breve hook-above|  112 : p : p   : p                     |
|  005 :  : A(~ :A breve tilde     |  113 : q : q   : q                     |
|  006 :  : A^~ :A circumflex tilde|  114 : r : r   : r                     |
|  020 :  : Y?  :Y hook-above      |  115 : s : s   : s                     |
|  025 :  : Y~  :Y tilde           |  116 : t : t   : t                     |
|  030 :  : Y.  :Y dot-below       |  117 : u : u   : u                     |
|  065 : A : A   : A                |  118 : v : v   : v                     |
|  066 : B : B   : B                |  119 : w : w   : w                     |
|  067 : C : C   : C                |  120 : x : x   : x                     |
|  068 : D : D   : D                |  121 : y : y   : y                     |
|  069 : E : E   : E                |  122 : z : z   : z                     |
|  070 : F : F   : F                |  128 : Ç : A.  : A dot-below           |
|  071 : G : G   : G                |  129 : ü : A(' : A breve acute         |
|  072 : H : H   : H                |  130 : é : A(` : A breve grave         |
|  073 : I : I   : I                |  131 : â : A(. : A breve dot-below     |
|  074 : J : J   : J                |  132 : ä : A^' : A circumflex acute    |
|  075 : K : K   : K                |  133 : à : A^` : A circumflex grave    |
|  076 : L : L   : L                |  134 : å : A^? :A circumflex hook-above|
|  077 : M : M   : M                |  135 : ç : A^. :A circumflex dot-below |
|  078 : N : N   : N                |  136 : ê : E~  :E tilde                |
|  079 : O : O   : O                |  137 : ë : E.  :E dot-below            |
|  080 : P : P   : P                |  138 : è : E^' :E circumflex acute     |
|  081 : Q : Q   : Q                |  139 : ï : E^` :E circumflex grave     |
|  082 : R : R   : R                |  140 : î : E^? :E circumflex hook-above|
|  083 : S : S   : S                |  141 : ì : E^~ :E circumflex tilde     |
|  084 : T : T   : T                |  142 : Ä : E^. :E circumflex dot-below |
|  085 : U : U   : U                |  143 : Å : O^' :O circumflex acute     |
|  086 : V : V   : V                |  144 : É : O^` :O circumflex grave     |
|  087 : W : W   : W                |  145 : æ : O^? :O circumflex hook-above|
|  088 : X : X   : X                |  146 : Æ : O^~ :O circumflex tilde     |
|  089 : Y : Y   : Y                |  147 : ô : O^. :O circumflex dot-below |
|  090 : Z : Z   : Z                |  148 : ö : O+. : O horn dot-below      |
|  097 : a : a   : a                |  149 : ò : O+' : O horn acute          |
|  098 : b : b   : b                |  150 : û : O+` : O horn grave          |
|  099 : c : c   : c                |  151 : ù : O+? : O horn hook-above     |
|  100 : d : d   : d                |  152 : ÿ : I.  : I dot-below           |
|  101 : e : e   : e                |  153 : Ö : O?  : O hook-above          |
|  102 : f : f   : f                |  154 : Ü : O.  : O dot-below           |
|  103 : g : g   : g                |  155 : ¢ : I?  : I hook-above          |
|  104 : h : h   : h                |  156 : £ : U?  : U hook-above          |
|  105 : i : i   : i                |  157 : ¥ : U~  : U tilde               |
|  106 : j : j   : j                |  158 : ₧ : U.  : U dot-below           |
|  107 : k : k   : k                |  159 : ƒ : Y`  : Y grave               |
|  108 : l : l   : l                |  160 : á : a.  : a dot-below           |
|  109 : m : m   : m                |  161 : í : a(' : a breve acute         |
|  110 : n : n   : n                |  162 : ó : a(` : a breve grave         |
|  111 : o : o   : o                |  163 : ú : a(. : a breve dot-below     |
+===================================+========================================+
                                                                        31
        Appendix B: Vietnamese Characters under VISCII and VIQR by
                    Encoding Order (continued)
+========================================+===================================+
|VISCII:Chr:VIQR : Descriptive Name      |VISCII:Chr:VIQR : Descriptive Name |
+------:---:-----:-----------------------+------:---:-----:------------------+
|  164 : ñ : a^' : a circumflex acute    |  210 : ╥ : O`  : O grave          |
|  165 : Ñ : a^` : a circumflex grave    |  211 : ╙ : O'  : O acute          |
|  166 : ª : a^? :a circumflex hook-above|  212 : ╘ : O^  : O circumflex     |
|  167 : º : a^. :a circumflex dot-below |  213 : ╒ : O~  : O tilde          |
|  168 : ¿ : e~  :e tilde                |  214 : ╓ : y?  : y hook-above     |
|  169 : ⌐ : e.  :e dot-below            |  215 : ╫ : u+` : u horn grave     |
|  170 : ¬ : e^' :e circumflex acute     |  216 : ╪ : u+? : u horn hook-above|
|  171 : ½ : e^` :e circumflex grave     |  217 : ┘ : U`  : U grave          |
|  172 : ¼ : e^? :e circumflex hook-above|  218 : ┌ : U'  : U acute          |
|  173 : ¡ : e^~ :e circumflex tilde     |  219 : █ : y~  : y tilde          |
|  174 : « : e^. :e circumflex dot-below |  220 : ▄ : y.  : y dot-below      |
|  175 : » : o^' :o circumflex acute     |  221 : ▌ : Y'  : Y acute          |
|  176 : ░ : o^` :o circumflex grave     |  222 : ▐ : o+~ : o horn tilde     |
|  177 : ▒ : o^? :o circumflex hook-above|  223 : ▀ : u+  : u horn           |
|  178 : ▓ : o^~ :o circumflex tilde     |  224 : α : a`  : a grave          |
|  179 : │ : O+~ :O horn tilde           |  225 : ß : a'  : a acute          |
|  180 : ┤ : O+  :O horn                 |  226 : Γ : a^  : a circumflex     |
|  181 : ╡ : o^. :o circumflex dot-below |  227 : π : a~  : a tilde          |
|  182 : ╢ : o+` : o horn grave          |  228 : Σ : a?  : a hook-above     |
|  183 : ╖ : o+? : o horn hook-above     |  229 : σ : a(  : a breve          |
|  184 : ╕ : i.  : i dot-below           |  230 : µ : u+~ : u horn tilde     |
|  185 : ╣ : U+. : U horn dot-below      |  231 : τ : a^~ :a circumflex tilde|
|  186 : ║ : U+' : U horn acute          |  232 : Φ : e`  : e grave          |
|  187 : ╗ : U+` : U horn grave          |  233 : Θ : e'  : e acute          |
|  188 : ╝ : U+? : U horn hook-above     |  234 : Ω : e^  : e circumflex     |
|  189 : ╜ : o+  : o horn                |  235 : δ : e?  : e hook-above     |
|  190 : ╛ : o+' : o horn acute          |  236 : ∞ : i`  : i grave          |
|  191 : ┐ : U+  : U horn                |  237 : φ : i'  : i acute          |
|  192 : └ : A`  : A grave               |  238 : ε : i~  : i tilde          |
|  193 : ┴ : A'  : A acute               |  239 : ∩ : i?  : i hook-above     |
|  194 : ┬ : A^  : A circumflex          |  240 : ≡ : dd  : d bar            |
|  195 : ├ : A~  : A tilde               |  241 : ± : u+. : u horn dot-below |
|  196 : ─ : A?  : A hook-above          |  242 : ≥ : o`  : o grave          |
|  197 : ┼ : A(  : A breve               |  243 : ≤ : o'  : o acute          |
|  198 : ╞ : a(? : a breve hook-above    |  244 : ⌠ : o^  : o circumflex     |
|  199 : ╟ : a(~ : a breve tilde         |  245 : ⌡ : o~  : o tilde          |
|  200 : ╚ : E`  : E grave               |  246 : ÷ : o?  : o hook-above     |
|  201 : ╔ : E'  : E acute               |  247 : ≈ : o.  : o dot-below      |
|  202 : ╩ : E^  : E circumflex          |  248 : ° : u.  : u dot-below      |
|  203 : ╦ : E?  : E hook-above          |  249 : ∙ : u`  : u grave          |
|  204 : ╠ : I`  : I grave               |  250 : · : u'  : u acute          |
|  205 : ═ : I'  : I acute               |  251 : √ : u~  : u tilde          |
|  206 : ╬ : I~  : I tilde               |  252 : ⁿ : u?  : u hook-above     |
|  207 : ╧ : y`  : y grave               |  253 : ² : y'  : y acute          |
| *208 : ╨ : DD  : D bar                 |  254 : ■ : o+. : o horn dot-below |
|  209 : ╤ : u+' : u horn acute          |  255 :   : U+~ : U horn tilde     |
+========================================+===================================+
         * VIQR also allows "╨" to be represented by "Dd" or "dD".

           MôT KHU╘N KHæ THÅNG NHäT CHO VIÄC X╝ L▌ D  KIÄN VIÄT NG


                  Nh≤m NghiΩn C╤u TiΩu Chuªn Ti¬ng Vi«t {1}
                             Thßng Chφn 1992 {2}



                                   Toßt Y¬u

Nhi½u lo╒i nhu li«u ╤ng d°ng c≤ th¼ d∙ng Vi«t ngµ  ≡π xuñt hi«n nhóm  ≡ßp  ╤ng
nhu cÑu x╪ l² dµ ki«n Vi«t ngµ bóng ≡i«n toßn ngαy cαng gia tσng.  Nhu cÑu tñt
y¬u cⁿa vi«c tφch h■p ti¬ng Vi«t vαo m⌠i tr▀╢ng ≡i«n toßn hi«n th╢i,  c√ng nh▀
vi«c trao ≡▒i dµ ki«n giµa cßc m⌠i tr▀╢ng nαy ≡½u cho thñy s± cÑn  thi¬t  phΣi
c≤ m╡t tiΩu chuªn chung.   Vσn ki«n nαy tr∞nh bαy nhµng cΓn nhíc k█  thuºt  c≤
tφnh cßch th±c ti¡n vα quan tr≈ng mα m╡t tiΩu chuªn nh▀ trΩn cÑn phΣi c≤, ≡░ng
th╢i c√ng duy«t l╒i m╡t s» quy ▀╛c vα ≡½ ßn hi«n hµu trong nhµng lπnh v±c quan
tr≈ng nαy.   Vσn ki«n c√ng tr∞nh bαy tr≈n v⌐n ≡½ ßn  cⁿa  nh≤m  Viet-Std,  g░m
nhµng ≡i¼m sau: 1) BΣng mπ s» 8-bit cho tñt cΣ mτu t± Vi«t nguyΩn v⌐n (tΩn Anh
ngµ lα Vietnamese Standard  Code  for  Information  Interchange,  g≈i  tít  lα
VISCII),   2) M╡t tiΩu chuªn 7-bit ≡≈c-≡▀■c-trong-ngoúc (c≤  tΩn  Anh  ngµ  lα
Vietnamese Quoted-Readable, g≈i tít lα VIQR), d∙ng ≡¼ trao ≡▒i dµ ki«n qua cßc
m╒ch 7-bit, c≤ giao di«n su⌠ng sδ v╛i h« mπ t± 8-bit nΩu trΩn, 3) M╡t quy ≡╕nh
giao di«n ≡ßnh chµ cho ng▀╢i d∙ng c≤ th¼ vºn hαnh d¡ dαng v╛i cΣ 1  vα  2. Tñt
cΣ nhµng ≡i¼m trΩn t╒o thαnh m╡t khu⌠n kh▒ th»ng nhñt cho  m⌠i  tr▀╢ng  x╪  l²
Vi«t ngµ,  v╫a ≡╜n giΣn,  v╫a c≤ hi«u nσng vα tφch h■p d¡ dαng. Vi«c xΓy  d±ng
khu⌠n kh▒ nαy trΩn th±c t¬ ≡π thαnh c⌠ng xuyΩn qua nhµng ╤ng d°ng h■p th╤c sΣn
xuñt b╖i m╡t s» tºp th¼ vα cß nhΓn trΩn m╡t s» h« th»ng mßy khßc nhau,  g░m cΣ
khi¼n h« Unix vα nhµng bi¬n th¼ t▀╜ng t±, h« th»ng khung X (X-window), MS-DOS,
Windows, vα xuyΩn qua cßc c⌠ng tr∞nh ≡ang ≡▀■c th±c hi«n ╖ cßc n╜i khßc.


1. LûI GIòI THIÄU
~~~~~~~~~~~~~~~~~

V╛i s» l▀■ng ng▀╢i Vi«t t╒i hΣi ngo╒i ngαy cαng gia tσng vα vi«c s╪  d°ng  mßy
vi tφnh ngαy cαng lan r╡ng t╒i Vi«t Nam,  vi«c s╪ d°ng chµ Vi«t trong lπnh v±c
x╪ l² tin t╤c ≡π tσng tr▀╖ng nhanh ch≤ng.  ╨░ng th╢i nhu cÑu v½ nhu li«u ti¬ng
Vi«t c√ng gia tσng khi¬n cho nhi½u c⌠ng ty ≡π ≡▀■c thαnh lºp vα thαnh c⌠ng t╒i
Hoa K╧ vα cßc n╜i khßc, phÑn l╛n chuyΩn v½ nhu li«u x╪ l² chµ Vi«t (Vietnamese
word processing).  Ngoαi ra, nhi½u t▒ ch╤c c√ng nh▀ cß nhΓn ≡π n▓ l±c cung cñp
nhi½u ╤ng d°ng c⌠ng c╡ng mi¡n phφ v╛i phªm chñt cao cho c╡ng  ≡░ng  Vi«t  Nam.
T╒i Vi«t Nam, cßc trung tΓm nh▀ Vi«n Tin H≈c [1] ch╞ng h╒n ≡π ghi nhºn s± ti¬n
b╡ khΣ quan v½ nhi½u mút,  trong ≡≤ c≤ vi«c Vi«t Nam h≤a nhµng b╡ nhu li«u ph▒

------------------------------------------------------------------------------
    {1} ╨╕a ch∩: Viet-Std, 1212 Somerset Dr., San Jose, California 95132, USA.
╨╕a ch∩ ≡i«n th▀: [e-mail]

    {2} Tßi bΣn ngαy 3 thßng 12 nσm 1992. än bΣn 1.1 nαy thay th¬ ñn  bΣn  1.0
xuñt bΣn vαo thßng giΩng 1992. S± khßc bi«t chⁿ y¬u giµa hai ñn bΣn lα s± hoßn
≡▒i v╕ trφ cⁿa hai mτu t± ╒ vα á trong bΣng mπ 8-bit.

                                         32
                                                                        33
th⌠ng.

    Tñt cΣ nhµng ≡i½u trΩn cho thñy r⌡ hai ≡i¼m quan tr≈ng:  1) Nhu cÑu v½ nhu
li«u d∙ng ≡▀■c v╛i chµ Vi«t cαng ngαy cαng tσng, vα 2) Kh⌠ng thi¬u tαi nσng ≡¼
ph°c v° nhµng nhu cÑu trΩn.   Ti¬c thay,  ch·ng ta vñp phΣi m╡t tr╖  ng╒i  rñt
l╛n:  hÑu h¬t cßc ╤ng d°ng d∙ng chµ Vi«t hi«n th╢i ch∩ ho╒t  ≡╡ng  ≡▀■c  trong
m╡t khu⌠n kh▒ hay m╡t m⌠i tr▀╢ng duy nhñt cⁿa ng▀╢i sΣn xuñt vα cßc  ╤ng  d°ng
do cßc nhα sΣn xuñt khßc nhau kh⌠ng t▀╜ng h■p v╛i nhau.   Kh»i ╤ng  d°ng  d∙ng
chµ Vi«t s¿ kh⌠ng bao gi╢ theo k╕p ≡α ≡≥i h÷i cⁿa th╕ tr▀╢ng  m╡t  khi  khuynh
h▀╛ng nΩu trΩn vτn c≥n t░n t╒i.   Ng▀╢i d∙ng mu»n s╪  d°ng  ti¬ng  Vi«t  trong
nhi½u lπnh v±c khßc nµa ch╤ kh⌠ng ch∩ gi╛i h╒n trong lπnh v±c  x╪  l²  chµ  mα
th⌠i,  vα vi«c mong ≡■i m╡t c⌠ng ty cung cñp m≈i ╤ng d°ng cho m≈i lπnh v±c cho
m≈i giαn mßy khßc nhau lα m╡t vi«c v⌠ t▀╖ng.  Ngoαi ra, nhµng chuyΩn viΩn vi¬t
cßc ╤ng d°ng nαy l╒i b╕ gi╛i h╒n vαo cßc nhu li«u  d°ng  c°  (software  tools)
ti¬ng Vi«t mα chφnh h≈ phΣi h≈c vα t± phßt tri¼n lñy.   Do ≡≤, vi«c ≡╕nh chuªn
lα m╡t ≡i½u bít bu╡c.   Bñt c╤ ai ≡π gúp phΣi t∞nh tr╒ng bñt  t▀╜ng  h■p  giµa
ASCII vα EBCDIC ≡½u c≤ th¼ m▀╢ng t▀■ng ra m╡t m⌠i tr▀╢ng mα m▓i mßy  d∙ng  m╡t
b╡ mπ t± khßc nhau,  l·c ≡≤ m╛i nhºn ra róng s» l▀■ng ╤ng d°ng cho m⌠i  tr▀╢ng
≡≤ rñt lα gi╛i h╒n vα vi«c trao ≡▒i dµ ki«n rñt phi½n  toßi.   M╡t  khu⌠n  kh▒
th»ng nhñt s¿ t╒o rñt nhi½u thuºn l■i cho cΣ ng▀╢i d∙ng  lτn  ng▀╢i  vi¬t  nhu
li«u.

    Bñt c╤ d± thΣo tiΩu chuªn ti¬ng Vi«t nαo c√ng phΣi c╤u  xΘt  m╡t  s»  ≡i¼m
tr≈ng y¬u trong ≡·ng ph╒m vi cⁿa n≤.   ╨i¼m ≡Ñu tiΩn vα quan tr≈ng nhñt lα vñn
≡½ tφch h■p.   V∞ vσn ki«n nαy ch· tr≈ng v½ m⌠i tr▀╢ng  7-bit  vα  8-bit  hi«n
th╢i,  m°c ≡φch chφnh y¬u phΣi lα s± tφch h■p (h╡i nhºp) ti¬ng Vi«t tr±c  ti¬p
vα d¡ dαng vαo cßc h« th»ng mßy hi«n th╢i.   TiΩu chuªn phΣi s╪ d°ng ≡▀■c ngay
lºp t╤c.   ╨i½u nαy ng° ² vi«c s╪ d°ng cßc mτu t± Vi«t nguyΩn v⌐n (precomposed
character), thay v∞ d∙ng cßc dñu r╢i (diacritic) ≡i kΦm v╛i cßc nguyΩn Γm,  v∞
ngoαi cßc nhu li«u ≡úc bi«t kh⌠ng c≤ ╤ng d°ng t▒ng qußt nαo c≤ th¼  d∙ng  ≡▀■c
cßc dñu r╢i.   TiΩu chuªn ≡½ ra phΣi ≡▀■c thi¬t k¬ khΘo lΘo ≡¼ tºn d°ng t»i ≡a
cßc nhu li«u hi«n th╢i.   Quy luºt quen thu╡c "╨╫ng phßt minh bßnh xe lÑn nµa"
kh⌠ng nhµng ch∩ lα ▀u th¬ mα c≥n lα bít bu╡c n¬u ch·ng ta mu»n xΓy d±ng m╡t s»
╤ng d°ng c╜ bΣn cÑn thi¬t trong m╡t th╢i gian v╫a phΣi.   Ngoαi ra,  n≤i chung
v½ th╢i-gian x╪-l² (time) c√ng nh▀ ch▓ ch╤a (space),  m≈i ng▀╢i ≡½u r⌡ lα vi«c
x╪ l² cßc mπ t± nguyΩn v⌐n c≤ hi«u nσng cao h╜n lα x╪ l² cßc mπ  t±  d∙ng  dñu
r╢i [2].   Do ≡≤ vi«c d∙ng dñu r╢i phΣi ≡▀■c gi╛i h╒n vαo nhµng tr▀╢ng h■p cÑn
thi¬t bñt khΣ khßng, nh▀ khi ≡ßnh bαn chµ hay khi truy½n dµ ki«n 7-bit.  Ngoαi
ra kh⌠ng c≤ l² do g∞ ≡¼ bít bu╡c m≈i ╤ng d°ng phΣi ≡▀╜ng ≡Ñu v╛i s±  ph╤c  t╒p
vα kΘm hi«u nσng cⁿa dñu r╢i trong vi«c x╪ l², l▀u trµ,  truy½n tin,  t╒o h∞nh
trΩn mαn Σnh vα in dµ ki«n 8-bit.

    ╨i¼m quan tr≈ng th╤ nh∞ lα phΣi c╤u xΘt nhµng ti½n l« ≡π c≤ s╟n trong kh»i
nhu li«u ti¬ng Vi«t. Bñt c╤ vi«c tiΩu chuªn h≤a nαo c√ng ≡≥i h÷i th╢i gian  ≡¼
thφch nghi,  n¬u tiΩu chuªn ≡≥i h÷i quß nhi½u thay ≡▒i th∞ s¿ gúp  nhi½u  phΣn
khßng cⁿa ng▀╢i d∙ng,  vα ch∩ lαm chºm tr¡ vi«c ßp d°ng tiΩu  chuªn  mα  th⌠i.
Cßc tiΩu chuªn dµ ki«n kh▒ 16-bit hoúc r╡ng h╜n nµa ≡π dÑn dÑn xuñt  hi«n  nh▀
Unicode [3] vα ISO 10646 [4].   Trong khi ch╢ ≡■i cßc tiΩu chuªn kh▒ r╡ng  nαy
tr╖ thαnh ph▒ th⌠ng,  ch·ng ta cÑn phΣi c≤ m╡t tiΩu chuªn ti¬ng Vi«t 8-bit  vα
tiΩu chuªn nαy phΣi ≡▀■c chñp nhºn mau ch≤ng ≡¼ kh÷i tr╖ thαnh l▓i th╢i.   M╡t
tiΩu chuªn 8-bit mu»n ≡▀■c chñp nhºn mau ch≤ng kh⌠ng th¼ b÷ qua nhµng ti½n  l«
trong kh»i nhu li«u hi«n th╢i.

                                                                        34
    ╨i¼m quan tr≈ng th╤ ba lα phΣi giΣi quy¬t vñn ≡½ giao di«n v╛i ng▀╢i d∙ng;
n¬u kh⌠ng ≡út ra th∞ t»i thi¼u phΣi suy xΘt Σnh h▀╖ng cⁿa  n≤  ≡»i  v╛i  ng▀╢i
d∙ng.   ╨i¼m nαy phÑn l╛n liΩn quan t╛i bαn ≡ßnh chµ 7-bit vα cßch t▀■ng tr▀ng
ti¬ng Vi«t bóng cßc k² t± 7-bit --- trong cΣ hai tr▀╢ng h■p nαy,  cßc dñu  ph°
phΣi lα dñu r╢i vα ≡▀■c t▀■ng tr▀ng b╖i cßc k² t± 7-bit v╛i h∞nh th∙ gÑn gi»ng
nh▀ dñu thºt.   ╨»i v╛i cßch ≡ßnh chµ Vi«t,  ch·ng ta phΣi duy tr∞,  n¬u ≡▀■c,
nhµng th≤i quen ≡ßnh chµ ≡π ≡▀■c quy ≡╕nh trΩn di¡n  ≡αn  Viet-Net  (m╡t  m╒ng
l▀╛i ≡i«n th▀ cⁿa ng▀╢i Vi«t) vα di¡n ≡αn Soc.Culture.Vietnamese  (nh≤m  th⌠ng
tin Vi«t Nam trΩn m╒ng l▀╛i Usenet) v╛i thαnh viΩn trΩn toαn  th¬  gi╛i.   ╨»i
v╛i cßch t▀■ng tr▀ng chµ Vi«t bóng 7-bit,  ≡i½u quan tr≈ng lα phΣi "≡≈c ≡▀■c."
M°c ≡φch ╖ ≡Γy lα r·t ngín th╢i gian h≈c vα c▒ v⌡ m╡t giao di«n thuÑn nhñt  ≡¼
ng▀╢i d∙ng kh⌠ng phΣi t»n th╢i gian h≈c cßch d∙ng cho m▓i  b╡  nhu  li«u  khßc
nhau.

    Cu»i c∙ng,  tiΩu chuªn phΣi c» gíng bóng m≈i cßch tuΓn theo khu⌠n kh▒  cⁿa
cßc tiΩu chuªn qu»c t¬, nh▀ ISO-8859/x [5],  ≡¼ ≡Σm bΣo s± t▀╜ng h■p v╛i nhµng
m⌠i tr▀╢ng hi«n hµu.   Ch╞ng h╒n,  ≡i½u nαy ≡≥i h÷i tiΩu chuªn ti¬ng Vi«t phΣi
duy tr∞ bΣng mπ s» ASCII cⁿa Hoa K╧,  c√ng nh▀ phΣi duy tr∞ v╕ trφ cⁿa tñt  cΣ
nhµng mτu t± Vi«t nαo ≡π c≤ s╟n trong bΣng 8859/Latin-1 ≡¼  bΣo  ≡Σm  cho  bαn
≡ßnh chµ 8859/Latin-1 c≤ th¼ ho╒t ≡╡ng b∞nh th▀╢ng cho cßc  mτu  t±  ≡≤.   Tuy
nhiΩn, v½ mút th±c t¬ c≤ m╡t s» yΩu cÑu ≡π l▓i th╢i.   Thφ d° nh▀ gÑn ≡Γy,  ⁿy
ban Unicode/ISO-10646 ≡π quy¬t ≡╕nh bπi b÷ vi«c cñm d∙ng v∙ng ki¼m t± (control
characters) --- mπ s» t╫ xx00h cho ≡¬n xx1Fh,  ngo╒i tr╫ C0 --- v╛i l²  do  lα
≡i½u nαy ch∩ lαm phφ ph╒m mπ s» v⌠ φch.  Nh▀ ta s¿ thñy d▀╛i ≡Γy, vi«c ñn ≡╕nh
cßc mτu t± Vi«t vαo nhµng khoΣng tr»ng nαy c≤ nhµng ≡i¼m l■i h╒i cⁿa n≤.  Vi«c
ch≈n l±a l■i h╒i phΣi ≡▀■c bi«n minh bóng nhµng l²  do  chφnh  ≡ßng,  vα  phΣi
nghiΩng v½ th±c ti¡n h╜n lα l² thuy¬t.

    Nhµng yΩu cÑu chⁿ y¬u trΩn ≡Γy cⁿa m╡t tiΩu chuªn ≡▀■c t≤m tít nh▀ sau:

    R1. Tφch h■p d¡ dαng vα tr±c ti¬p vαo cßc h« th»ng mßy hi«n th╢i.

    R2. Lαm cho cßc nhu li«u hi«n th╢i thφch nghi d¡ dαng v╛i tiΩu chuªn m╛i.

    R3. Ph▀╜ng phßp mπ h≤a vα giao di«n phΣi d¡ nh╛ vα d¡ s╪ d°ng.

    R4. TuΓn theo cßc tiΩu chuªn qu»c t¬.

    R5. Vi«c ch≈n l■i h╒i phΣi ≡▀■c cΓn nhíc d±a trΩn th±c ti¡n vα  c≤  l²  do
        chφnh ≡ßng.

    Trong cßc phÑn sau ≡Γy,  ch·ng t⌠i s¿ duy«t l╒i ▀u khuy¬t ≡i¼m  cⁿa  nhµng
cßch mπ h≤a ti¬ng Vi«t.   PhÑn 3 s¿ m⌠ tΣ chi ti¬t bΣng mπ s» Vi«t  ngµ  8-bit
cⁿa Viet-Std.   PhÑn 4 s¿ tr∞nh bαy m╡t ph▀╜ng phßp  mπ  h≤a  "≡≈c-≡▀■c-trong-
ngoúc" ßp d°ng cho d≥ng dµ ki«n 7-bit trong ≡≤ c≤ ≡i«n th▀ vα cßch  ≡ßnh  chµ.
Cu»i c∙ng, PhÑn 5 phßc h≈a m╡t s» luºt l« vα quy ▀╛c riΩng bi«t thφch nghi cho
m╡t vαi ╤ng d°ng c° th¼.


2. DUYÄT LÇI NH NG QUY ┐òC HIÄN THûI
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Khi duy«t qua nhµng quy ▀╛c d∙ng b╖i cßc nh≤m phßt tri¼n nhu li«u  hi«n  th╢i,
                                                                        35
ta thñy m╡t ≡úc ≡i¼m n▒i bºt: hÑu h¬t m≈i ng▀╢i ≡½u  c⌠ng  nhºn  ▀u  ≡i¼m  cⁿa
ph▀╜ng phßp mπ h≤a mτu t± nguyΩn v⌐n vα ch≈n ≡≤ lαm ≡i½u ki«n tiΩn  quy¬t. Tuy
nhiΩn,  khi ch≈n ph▀╜ng phßp nαy ta gúp phΣi nhµng kh≤ khσn quen thu╡c:  ngoαi
nhµng mτu t± ≡π c≤ s╟n trong bΣng ASCII,  ti¬ng Vi«t Nam c≥n cÑn thΩm 134  mτu
t± nµa. Trong s» nαy, 128 mτu t± c≤ th¼ ≡▀■c ≡út trong v∙ng C1 vα G1.  Sßu mτu
t± Vi«t c≥n l╒i c≤ th¼ ≡▀■c ≡út trong v∙ng C0 vα G0  bóng  nhµng  ph▀╜ng  phßp
khßc nhau:

    A1. X¬p vαo ch▓ cⁿa 6 mτu t± t▀╜ng ≡»i "φt ≡▀■c d∙ng nhñt" trong  v∙ng  G0
        khi x╪ l² ti¬ng Vi«t.

    A2. X¬p vαo ch▓ cⁿa 6 k² t± trong tºp k² t± khΣ  hoßn  NRC  {3}  (National
        Replacement Character set).

    A3. B÷ h╞n 6 mτu t± t▀╜ng ≡»i "φt d∙ng nhñt" {4} trong ti¬ng Vi«t,  nh▀ ,
        , , , , vα .

    A4. Thay th¬ cßc mτu t± v╛i g»c "y" bóng "i," thφ d° nh▀ "k█  s▀"  s¿  tr╖
        thαnh "kε s▀."

    A5. ╨út 6 mτu t± nαy vαo v∙ng ki¼m t± C0 cⁿa ASCII.

    GiΣi phßp A1 vα A2 th÷a mπn cßc nhu cÑu tiΩu bi¼u cⁿa nhµng m⌠i-tr▀╢ng x╪-
l²-chµ mα trong ≡≤ ta c≤ th¼ trßnh s╪ d°ng nhµng k² t± ASCII φt d∙ng  hoúc  s╪
d°ng ch·ng bóng cßch chuy¼n ph⌠ng (font shifting).  Tuy nhiΩn cΣ hai giΣi phßp
nαy ≡½u phß tan vi¡n t▀■ng tφch h■p ti¬ng Vi«t vαo cßc m⌠i tr▀╢ng  ASCII  hi«n
th╢i mα trong ≡≤ tñt cΣ cßc k² t± trong v∙ng G0 ≡½u phΣi ≡▀■c duy tr∞.  M▓i k²
t± G0 ch∩ c≤ m╡t nhi«m v° riΩng vα kh⌠ng th¼ ≡▀■c d∙ng l╒i vαo  m╡t  nhi«m  v°
khßc ≡▀■c.   L² do th╤ nhñt lα vi«c t╒o h∞nh cⁿa k² t± G0 ≡≤ s¿ b╕ sai v∞ h∞nh
t╒o ra s¿ lα m╡t mτu t± Vi«t.   V∞ cßc k² t± G0 ≡▀■c s╪ d°ng rñt th▀╢ng xuyΩn,
vi«c xung ≡╡t giµa mτu t± Vi«t vα k² t± G0 trong m╡t m⌠i tr▀╢ng  tφch  h■p  lα
vi«c kh⌠ng th¼ chñp nhºn ≡▀■c.   L² do th╤ hai lα trong khi ph▀╜ng phßp chuy¼n
ph⌠ng c≤ th¼ cΣi thi«n t∞nh tr╒ng xung ≡╡t nαy trong vαi tr▀╢ng h■p,  ch·ng ta
s¿ gúp kh≤ khσn trÑm tr≈ng h╜n.  M⌠i tr▀╢ng nhu li«u th▀╢ng th▀╢ng ñn ≡╕nh cho
m▓i k² t± trong G0 m╡t ² nghεa riΩng,  ≡úc bi«t lα cßc k²  t±  NRC.   Hπy  xΘt
tr▀╢ng h■p ch·ng ta thay th¬ k² t± g╒ch-chΘo-ng▀■c "\" bóng m╡t  mτu  t±  Vi«t
(nh▀ mτu t± ▒ ch╞ng h╒n) trong m⌠i tr▀╢ng Unix.   K² t±  g╒ch-chΘo-ng▀■c  ≡▀■c
d∙ng trong nhi½u c╜ ch¬ thoßt (escape mechanism) cⁿa Unix thαnh th╪ ra mτu  t±
"▒" kh⌠ng th¼ ≡▀■c d∙ng m╡t cßch b∞nh th▀╢ng mα phΣi  ≡▀■c  "thoßt"  ≡úc  bi«t
bóng cßch nαy hay cßch khßc.   ╨Γy kh⌠ng phΣi ch∩ lα m╡t s± bñt ti«n nh÷; vi«c
trao ≡▒i dµ ki«n s¿ gúp ríc r»i nhi½u v∞ cßc h« th»ng mßy khßc s¿  kh⌠ng  hi¼u
c╜ ch¬ thoßt ≡úc bi«t nαy, do ≡≤ dµ ki«n s¿ kh⌠ng ≡▀■c bΣo toαn.   Bñt c╤ tiΩu
chuªn nαo ßp d°ng giΣi phßp nαy s¿ kh⌠ng lαm tr≥n ch╤c nσng cσn bΣn cⁿa n≤  lα
cung cñp s± thuÑn nhñt trΩn cßc h« th»ng mßy khßc nhau.  TrΩn ≡Γy lα n≤i v½ k²
t± g╒ch-chΘo-ng▀■c nh▀ng cßc k² t± G0 khßc c√ng c≤ nhµng kh≤ khσn t▀╜ng t±.

------------------------------------------------------------------------------
    {3} Tºp NRC g░m c≤ 12 k² t± ASCII ╖ v∙ng G0 lα #, $, @, [, \, ], ^, `,  {,
|, },  vα ~.   Nhµng k² t± nαy c≤ th¼ ≡▀■c thay th¬ bóng cßc chµ khßc t∙y theo
nhu cÑu cⁿa m▓i qu»c gia.

    {4} ═t d∙ng b╖i v∞ nhµng mτu t± nαy (a) φt khi bít ≡Ñu m╡t chµ vα  v∞  th¬
φt khi cÑn mút hoa, vα (b) φt xuñt hi«n trong cßc chµ Vi«t.
                                                                        36
    Cßc ≡½ ngh╕ A3 vα A4 h╒n ch¬ dµ ki«n Vi«t ngµ  bóng  cßch  nαy  hoúc  cßch
khßc.   HÑu h¬t m≈i ng▀╢i ≡░ng ² róng vi«c lo╒i b÷ m╡t s» mτu t± Vi«t lα  vi«c
kh⌠ng th¼ chñp nhºn ≡▀■c; thºt vºy,  trong phÑn thΣo luºn ╖ trΩn,  ch·ng ta ≡π
coi vi«c bΣo toαn tñt cΣ mτu t± Vi«t lα m╡t yΩu cÑu ≡▀╜ng nhiΩn.  Tuy th¬, cÑn
phΣi n≤i thΩm n╜i ≡Γy lα ≡½ ngh╕ A4 kh⌠ng phΣi lα kh⌠ng c≤ l² do  chφnh  ≡ßng.
╨π c≤ m╡t tr▀╢ng phßi t▀ t▀╖ng nghε róng trong cßc chµ ch∩ ch╤a  "y"  (vα  dñu
gi≈ng n¬u c≤) lα nguyΩn Γm duy nhñt,  chµ "y" c≤ th¼ ≡▀■c thay th¬ bóng mτu t±
"i" v∞ cßch phßt Γm cⁿa cΣ hai chµ t▀╜ng ≡▀╜ng v╛i nhau.   Khßi ni«m nαy ≡π c≤
t╫ nσm 1948 [6,7]. Tuy nhiΩn, nhi«m v° cⁿa m╡t tiΩu chuªn mπ h≤a kh⌠ng phΣi lα
giΣi quy¬t cßc vñn ≡½ liΩn h« t╛i ng⌠n ngµ.  Do ≡≤ vi«c ch≈n ≡½ ngh╕ A4 lα m╡t
≡i½u kh⌠ng t»t.

    L² do ≡Ñu tiΩn ≡¼ bßc b÷ ≡½ ngh╕ A5 chⁿ y¬u phßt xuñt t╫ lπnh  v±c  truy½n
tin v∞ cßc m╒ch chuy¼n tin (data communication channel) d∙ng nhi½u  mπ  t±  C0
trong vi«c ki¼m soßt dµ ki«n.   Ngoαi ra, ≡½ ngh╕ nαy t╒o thΩm m╡t s» kh≤ khσn
khi tφch h■p ti¬ng Vi«t vαo nhµng m⌠i tr▀╢ng mα trong ≡≤ m╡t s» mπ t± C0  ≡▀■c
d∙ng trong giao di«n bαn ≡ßnh chµ (keyboard  interface)  vα  trong  vi«c  ≡i½u
khi¼n khu⌠n th╤c dµ ki«n (data format control),  t▀╜ng t± nh▀ nhµng  kh≤  khσn
gúp phΣi trong ≡½ ngh╕ A1 vα A2.   Tuy th¬,  nh▀ s¿ ≡▀■c tr∞nh bαy  trong  cßc
phÑn k¬ ti¬p, vi«c ch≈n l±a thºn tr≈ng 6 mπ t± C0 trong th±c t¬ ≡π cho thñy c≤
k¬t quΣ t»t ≡⌐p mα vτn trßnh ≡▀■c cßc mπ t± quan tr≈ng d∙ng trong  vi«c  th⌠ng
tin dµ ki«n.  H╜n nµa, hÑu h¬t cßc m╒ch chuy¼n tin cho phΘp chuy¼n ≡╒t dµ ki«n
8-bit m╡t cßch minh b╒ch,  trung th±c,  vα kh⌠ng c≤ l² do g∞ ≡¼  e  ng╒i  róng
ch·ng ta kh⌠ng th¼ chuy¼n bñt c╤ mπ s» nαo qua cßc m╒ch chuy¼n tin nαy.

    Trong nhµng tr▀╢ng h■p cß bi«t mα C0 ≡▀■c d∙ng trong giao  di«n  bαn  ≡ßnh
chµ,  vi«c ch≈n l±a kh⌠n khΘo c√ng nh▀ vi«c ñn ≡╕nh l╒i nhi«m v° cⁿa cßc  phφm
chµ s¿ lαm giΣm thi¼u nhµng xung khíc.   Cßc mπ t± d∙ng trong vi«c ≡i½u  khi¼n
khu⌠n th╤c dµ ki«n th▀╢ng thay ≡▒i theo t╫ng nhu li«u  ╤ng  d°ng  nh▀ng  th⌠ng
th▀╢ng ch·ng nóm rΣi rßc trong v∙ng C0 vα C1.   Do ≡≤ ≡Γy lα m╡t kh≤ khσn  ph▒
th⌠ng cho vi«c tφch h■p b╖i v∞ ch·ng ta cÑn phΣi d∙ng tñt cΣ cßc mπ  s»  trong
C1 cho chµ Vi«t.   Tuy nhiΩn,  m╡t lÑn nµa,  vi«c xung khíc c≤ th¼  ≡▀■c  giΣm
thi¼u bóng cßch nghiΩn c╤u cßc ╤ng d°ng quan tr≈ng.   Cu»i c∙ng,  ch·ng ta  c≤
th¼ ch≈n 6 mτu t± Vi«t φt d∙ng nhñt ≡¼ lαm giΣm t»i ≡a xßc suñt xung khíc.

    Xin ch· ² lα phÑn tr∞nh bαy trΩn ≡Γy ≡π phΓn tφch t╫ng ≡½ ngh╕ m╡t ≡¼  xem
ch·ng c≤ th÷a mπn cßc ≡i½u ki«n cho vi«c tφch h■p chµ Vi«t  Nam  vαo  cßc  ╤ng
d°ng vα cßc h« th»ng mßy hi«n th╢i,  nh▀ ≡π nΩu ra trong  PhÑn  1.   Ch·ng  ta
kh⌠ng th¼ kh⌠ng nhñn m╒nh tÑm m╤c quan tr≈ng l╛n lao cⁿa  m°c  tiΩu  tφch  h■p
nαy.   M°c tiΩu nαy ≡π nΣy sinh nhi½u kh≤ khσn vα khi¬n cho ch·ng ta phΣi chñp
nhºn NguyΩn Tíc Th±c D°ng sau ≡Γy: Kh⌠ng c≤ cßch nαo t╒o ra m╡t tiΩu chuªn vºn
hαnh hoαn hΣo v╛i tñt cΣ m≈i ╤ng d°ng hi«n th╢i, do ≡≤,  phΣi cΓn nhíc th±c t¬
≡¼ ≡½ ra m╡t tiΩu chuªn c≤ th¼ s╪ d°ng ≡▀■c trong cαng  nhi½u  ╤ng  d°ng  quan
tr≈ng vα cαng nhi½u h« th»ng mßy cαng t»t.  Ch·ng t⌠i xin nhñn m╒nh ╖ ≡i¼m "c≤
th¼ s╪ d°ng ≡▀■c."


                                                                        37
3. VISCII: QUY ╨ÿNH M├ 8-BIT CHO VIÄT NG
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

3.1 ╨ôNG L╣C
~~~~~~~~~~~~

Nhµng bóng ch╤ng c° th¼ cho thñy h▀╛ng giΣi quy¬t A5 m⌠ tΣ ╖ phÑn trΩn,  ≡út 6
mτu t± Vi«t vαo v∙ng C0,  lα h▀╛ng c≤ nhi½u khΣ nσng nhñt ≡¼  th÷a  mπn  nhµng
≡i½u ki«n ≡út ra trong PhÑn 1.   Vi«c l±a ch≈n 6 mπ t± C0 vα 6 mτu t± Vi«t  φt
d∙ng nhñt,  n¬u ≡▀■c cΓn nhíc k█ cαng,  trΩn th±c t¬ s¿ lαm giΣm rñt nhi½u xßc
suñt xung khíc.   Nhµng ▀u t▀ trong lπnh v±c th⌠ng tin dµ ki«n ≡▀■c giΣi quy¬t
bóng cßch trßnh cßc mπ t± C0 th▀╢ng d∙ng ≡¼ ≡i½u khi¼n  khu⌠n  th╤c  dµ  ki«n.
Thºt ra, m»i quan tΓm trong lπnh v±c th⌠ng tin dµ ki«n nΩn h▀╛ng vαo cßc k² t±
trong v∙ng C1 vα G1;  m╡t thφ d° n▒i bºt lα vi«c truy½n ≡i«n th▀ qua cßc  c▒ng
7-bit vα cßc b╡ chuy¼n ≡i«n th▀.   Nhµng thñt b╒i trong vi«c chuy¼n tin ╖  ≡Γy
th▀╢ng lα do vi«c d∙ng bit th╤ tßm ch╤ kh⌠ng phΣi v∞ cßc mπ t±  C0.   DÑu  th¬
nαo ≡i nµa,  ta vτn c≤ nhµng ph▀╜ng th╤c ≡¼ truy½n dµ ki«n  trong  tr╒ng  thßi
"8-bit" nαo ≡≤,  hoúc d∙ng d╒ng "ti¬ng Vi«t ≡≈c-≡▀■c-trong-ngoúc" nh▀ ≡▀■c  m⌠
tΣ trong PhÑn 4.

    ┐u ≡i¼m chφnh cⁿa ph▀╜ng phßp nαy lα s± s╟n sαng vα d¡ dαng tφch  h■p  vαo
m⌠i tr▀╢ng hi«n th╢i mα kh⌠ng gúp phΣi nhµng tríc tr╖ nh▀ cßc h▀╛ng giΣi quy¬t
khßc, giΣ s╪ róng cßc h▀╛ng giΣi quy¬t kia c≤ th¼ tφch h■p ≡▀■c.   S± ki«n bΣn
vσn ki«n nαy ≡▀■c so╒n bóng h« th»ng TeX ch╒y trΩn Unix lα m╡t bóng ch╤ng h∙ng
h░n cho thñy s± thαnh c⌠ng cⁿa h▀╛ng giΣi quy¬t nαy.   BΣn vσn ≡▀■c so╒n trong
m╡t khung-X 8-bit, d∙ng m╡t ñn bΣn Elvis ≡▀■c bi¬n ≡▒i {5} ch·t φt.  (Elvis lα
m╡t ñn bΣn 8-bit c⌠ng c╡ng d∙ng gi»ng nh▀ b╡ so╒n vσn Vi cⁿa  Unix.)   CΣ  TeX
(m╡t h« th»ng so╒n tαi li«u) vα Dvi2ps (m╡t ╤ng d°ng t╒o ra  d╒ng  PostScript)
≡½u nhºn vα x╪ l² dµ ki«n ti¬ng Vi«t (8-bit) m╡t cßch d¡ dαng vα  th⌠ng  su»t.
Cßc ╤ng d°ng khßc g░m c≤ bΣng tφnh (spreadsheet),  ╤ng  d°ng  nh∞n  chµ  (text
viewer),  in PostScript vα ma trºn ≡i¼m (dot-matrix),  WordPerfect,  Word,  PC
Tools cⁿa DOS,  v.v... ≡π ≡▀■c th╪ nghi«m qua vα ch╒y t»t  ≡⌐p  v╛i  vσn  ki«n
ti¬ng Vi«t.   Tñt cΣ cßc bi¬n ≡▒i, n¬u c≤, chⁿ y¬u lα ≡¼ cßc ╤ng d°ng nαy chñp
nhºn dµ ki«n 8-bit.   M╡t ╤ng d°ng gißo khoa ti¬ng Vi«t ≡π ≡▀■c vi¬t bóng ng⌠n
ngµ thΣo ch▀╜ng C v╛i cßc cΓu vσn ti¬ng Vi«t 8-bit ≡φnh  kΦm  trong  bΣn  thΣo
ch▀╜ng. V╛i ≡α gia tσng qu»c t¬ h≤a nhu li«u hi«n nay,  cßc ╤ng  d°ng  vα  nhu
li«u d°ng c° ≡ang ≡▀■c s╪a ≡▒i ≡¼ nhºn dµ ki«n 8-bit,  vα do d≤ vi«c tφch  h■p
cⁿa b╡ mπ t± Vi«t nαy l╒i cαng d¡ dαng h╜n.

3.2 C┴C L▌ DO BIÄN MINH VIÄC M├ H╙A
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

M╡t ≡i½u ki«n cσn bΣn lα phΣi bΣo toαn bΣng k²  t±  h∞nh  (graphic  character)
ASCII 7-bit (G0) v∞ m°c tiΩu lα tφch h■p v╛i cßc m⌠i tr▀╢ng hi«n th╢i.  Do ≡≤,
v∙ng G0 ≡▀■c giµ nguyΩn v⌐n.   ╨»i v╛i 6 mπ t± C0, ≡Ñu tiΩn ch·ng t⌠i phßc h≈a
ra v∙ng C0 vα xem xΘt cßc cßch d∙ng tiΩu bi¼u vα li«t kΩ trong BΣng 1. Cßc  mπ
t± ≡▀■c ch≈n, STX (2),  ENQ (5),  ACK (6),  DC4 (20),  EM (25),  vα RS (30) lα
nhµng mπ t± φt gΓy tr╖ ng╒i nhñt cho vi«c truy½n tin c√ng nh▀ cho cßc ╤ng d°ng
quan tr≈ng ≡π ≡▀■c c╤u xΘt.  Ch╞ng h╒n nh▀ cßch d∙ng cⁿa mπ t± ACK thºt s± t∙y
thu╡c vαo ngµ cΣnh cⁿa n≤.   Trong nhµng biΩn bΣn (protocol) mα ch·ng  t⌠i  ≡π

------------------------------------------------------------------------------
    {5} S╪a ≡¼ c≤ th¼ ≡ßnh chµ Vi«t nh▀ s¿ ≡▀■c m⌠ tΣ trong cßc phÑn sau.
                                                                        38
   BΣng 1: Nhµng xung ≡╡t c≤ th¼ xΣy ra khi d∙ng C0.  Cßc mπ s» d∙ng
   trong tiΩu chuªn 8-bit VISCII ≡▀■c nΩu ra v╛i dñu "+" bΩn c╒nh.

   =====================================================================
   M├ SÅ  T╩N  CTRL-  CHUNG   M┴Y IN(PC)    PC        Unix     vi (Unix)
   ---------------------------------------------------------------------
     0    NUL   @    C string                       strings
     1    SOH   A
    +2    STX   B                                            back screen
     3    ETX   C    INTR                  INTR      INTR      INTR
     4    EOT   D    EOF                              EOF    back tab
    +5    ENQ   E
    +6    ACK   F                                            forw.screen
     7    BEL   G    BEL         BEL        BEL       BEL
     8    BS    H    BS          BS         BS        BS        BS
     9    HT    I    HT          HT         HT        HT        HT
    10    LF    J    LF          LF         LF        LF        LF
    11    VT    K                VT
    12    FF    L    FF          FF         FF                redraw
    13    CR    M    CR          CR         CR        CR        CR
    14    SO    N            wide on(IBM)
    15    SI    O            comp.on(IBM)
    16    DLE   P                          Prt.on/off
    17    DC1   Q    XOFF      XOFF         XOFF      XOFF
    18    DC2   R            comp.off(IBM)            retype
    19    DC3   S    XON       XON          XON       XON
   +20    DC4   T            wide off(IBM)                     forw.tab
    21    NAK   U            clr buf(IBM)             kill     kill
    22    SYN   V                                     literal  literal
    23    ETB   W                                     werase   werase
    24    CAN   X                                     kill
   +25    EM    Y                                     suspend
    26    SUB   Z                           EOF       suspend
    27    ESC   [    ESC     chu▓i ESC      ESC       ESC       ESC
    28    FS    \                                     quit
    29    GS    ] Telnet ESC
   +30    RS    ^
    31    US    _                          Windows
  =====================================================================




   BΣng 2:  Nhµng mτu t± Vi«t Nam ≡π c≤ s╟n trong tiΩu chuªn  8859/Latin-1.

        +===============================================================+
        | 0 : 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : A : B : C : D : E : F |
   +----+---------------------------------------------------------------+
   | Cx | └ : ┴ : ┬ : ├ :   :   :   :   : ╚ : ╔ : ╩ :   : ╠ : ═ :  :  : |
   | Dx | ╨ :   : ╥ : ╙ : ╘ : á :   :   :   : ┘ : ┌ :   :   : ▌ :  :  : |
   | Ex | α : ß : Γ : π :   :   :   :   : Φ : Θ : Ω :   : ∞ : φ :  :  : |
   | Fx | ≡ :   : ≥ : ≤ : ⌠ : ⌡ :   :   :   : ∙ : · :   :   : ² :  :  : |
   +====================================================================+
                                                                        39


duy«t qua,  ACK ch∩ ≡▀■c coi lα m╡t mπ t± "≡i½u khi¼n" khi nóm bΩn ngoαi khung
dµ ki«n;  bΩn trong khung dµ ki«n n≤ ≡▀■c coi nh▀ m╡t mπ t± b∞nh  th▀╢ng.   ╨¼
lαm giΣm xßc suñt xung khíc h╜n nµa,  6 mτu t± chµ hoa ti¬ng Vi«t t▀╜ng ≡»i φt
d∙ng nhñt, , , , , , vα , ≡▀■c ≡út vαo ch▓ cⁿa 6 mπ t± C0 nαy.

    Vñn ≡½ ti¬p theo lα mπ h≤a 128 chµ Vi«t c≥n l╒i trong v∙ng ASCII-r╡ng  (C1
vα G1).   V∞ kh⌠ng c≤ m╡t tiΩu chuªn qu»c t¬ duy nhñt trong v∙ng  nαy  ≡¼  d±a
theo,  ph▀╜ng chΓm t»t nhñt lα cªn thºn t»i ≡a ≡¼ khi gúp tr▀╢ng h■p xñu  nhñt
th∞ ng▀╢i d∙ng vτn c≥n c≤ th¼ s╪ d°ng tñt cΣ mτu t± con.

    Vi«c mπ h≤a cßc k² t± trong v∙ng C1 gúp φt ríc r»i h╜n, múc dÑu m╡t vαi k²
t± C1 ≡▀■c s╪ d°ng v╛i ² nghεa ≡úc bi«t trong m╡t s» ╤ng d°ng.   Duy«t qua cßc
c⌠ng tr∞nh hi«n th╢i ≡¼ tiΩu chuªn h≤a vi«c chuy¼n ≡i«n th▀  8-bit,  ch·ng  ta
thñy cßc k² t± C1 ≡▀■c coi lα cßc k² t± h∞nh vα kh⌠ng b╕ gßn cho m╡t  ²  nghεa
≡úc bi«t g∞ cΣ.  Tuy nhiΩn, ≡▀╢ng l»i thºn tr≈ng lα ch∩ nΩn ≡út cßc mτu t± chµ
hoa vαo trong v∙ng C1.

    ╨»i v╛i v∙ng G1, ch·ng t⌠i nhím vαo vi«c ≡ßp ╤ng nhu cÑu cⁿa b╡ mπ t± d∙ng
trΩn mßy vi tφnh PC (code page 850) vα tuΓn theo,  n¬u c≤ th¼ ≡▀■c,  tºp k² t±
8859/Latin-1 mα trong ≡≤ m╡t s» mτu t± Vi«t ≡π c≤ s╟n.

    Kinh nghi«m trong vi«c thi¬t k¬ b╡ mπ t± cho h« th»ng MS-DOS  ≡π  ≡▀a  ≡¬n
vi«c c╤u xΘt cßc k² t± v¿ ≡▀╢ng th╞ng trong b╡ mπ t± PC.   N¬u mu»n  cho  phΘp
m╡t s» ╤ng d°ng v╫a d∙ng chµ Vi«t v╫a d∙ng k² t± v¿ ≡▀╢ng th╞ng mα kh⌠ng  phΣi
chuy¼n ph⌠ng,  ch·ng ta ch∩ c≤ th¼ bΣo toαn t»i ≡a lα cßc mτu t± con vα cßc k²
t± v¿ ≡▀╢ng ≡╜n vα ≡⌠i (single- and double-line drawing characters)  mα  th⌠i.
╨i½u nαy c≤ nghεa lα phΣi ≡út cßc mτu t± chµ hoa vαo v╕ trφ cⁿa cßc k²  t±  v¿
≡▀╢ng ≡╜n vα ≡⌠i.  V╛i cßch nαy, ng▀╢i d∙ng MS-DOS c≤ th¼ ≡▀■c cung cñp tºp mπ
t± c≤ tñt cΣ cßc mτu t± Vi«t hoúc tºp mπ t± trong ≡≤ m╡t s» chµ  hoa  b╕  thay
th¬ bóng cßc k² t± v¿ ≡▀╢ng ≡╜n vα ≡⌠i.   ╨»i v╛i cßc ╤ng d°ng c≤  s╟n,  ng▀╢i
d∙ng c≤ th¼ ch≈n tºp mπ t± nαo thφch h■p nhñt cho m°c ≡φch cⁿa  h≈.   Khi  bít
bu╡c phΣi s╪ d°ng tºp mπ t± v¿ ≡▀╢ng th╞ng,  nhµng thi«t h╒i v∞ thi¬u cßc  mτu
t± Vi«t c√ng giΣm thi¼u v∞ cßc mτu t± thi¬u lα mτu t± chµ  hoa  t▀╜ng  ≡»i  φt
≡▀■c d∙ng.  ╨»i v╛i cßc ╤ng d°ng m╛i, ph▀╜ng th╤c "thay tºp mπ t±" c≤ th¼ ≡▀■c
s╪ d°ng, n¬u mu»n.

    Nhu cÑu t▀╜ng h■p v╛i tiΩu chuªn 8859/Latin-1 ch∩ lα m╡t y¬u t»  ph°  nhóm
thuºn ti«n cho ng▀╢i d∙ng ch╤ kh⌠ng phΣi lα ≡i½u bít bu╡c.   M╡t thφ d° c° th¼
lα ng▀╢i d∙ng ╖ Phßp nghε róng h≈ ch∩ cÑn nhñn c∙ng nhµng phφm chµ nh▀ nhau ≡¼
t╒o ra nhµng mτu t± Vi«t vα mτu t± Phßp gi»ng nhau,  nh▀ chµ Θ ch╟ng h╒n.   ╨≤
c√ng lα ≡i½u t± nhiΩn vα h■p l².   Vi«c ch≈n t▀╜ng h■p v╛i  8859/Latin-1  xuñt
phßt t╫ s± ph▒ th⌠ng vα th╕nh hαnh cⁿa bαn ≡ßnh chµ vα ph⌠ng chµ 8859/Latin-1,
nh▀ lo╒t thi¬t b╕ ≡Ñu cu»i VT (VT-terminal series) cⁿa hπng Digital, bΣng phφm
chµ Xterm,  vα ╤ng d°ng khung cⁿa Microsoft (MS Windows).   BΣng 2 li«t kΩ  ra
cßc mτu t± 8859/Latin-1 trong v∙ng G1 tr∙ng h■p v╛i mτu t±  Vi«t  {6}. C≤  th¼
k¬t luºn róng tñt cΣ vσn bΣn ch╤a chµ 8859/Latin-1 mα  phÑn  l╛n  cßc  chµ  lα
ASCII vα nhµng mτu t± thu╡c BΣng 2,  ch╞ng h╒n nh▀ vσn bΣn ti¬ng Phßp,  ≡½u c≤

------------------------------------------------------------------------------
    {6} L▀u ² róng chµ ≡ trong BΣng 2 thºt ra lα chµ "edh" cⁿa ti¬ng Bσng  ╨Σo
theo 8859/Latin-1; d╒ng ≡ cⁿa Vi«t Nam ≡·ng theo bΣng 8859/Latin-2 h╜n.
                                                                        40
            BΣng 3: VISCII -- D± thΣo tiΩu chuªn 8-bit cho chµ Vi«t.
    +======================================================================+
    |    ||  0 : 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : A : B : C : D : E : F |
    +======================================================================+
    | 0x || NUL:SOH:  :ETX:EOT:  :  :BEL:BS :HT :LF :VT :FF :CR :SO :SI |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 1x || DLE:DC1:DC2:DC3:  :NAK:SYN:ETB:CAN:  :SUB:ESC:FS :GS :  :US |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 2x || SP : ! : " : # : $ : % : & : ' : ( : ) : * : + : , : - : . : / |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 3x ||  0 : 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : : : ; : < : = : > : ? |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 4x ||  @ : A : B : C : D : E : F : G : H : I : J : K : L : M : N : O |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 5x ||  P : Q : R : S : T : U : V : W : X : Y : Z : [ : \ : ] : ^ : _ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 6x ||  ` : a : b : c : d : e : f : g : h : i : j : k : l : m : n : o |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 7x ||  p : q : r : s : t : u : v : w : x : y : z : { : | : } : ~ :DEL|
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 8x ||  Ç : ü : é : â : ä : à : å : ç : ê : ë : è : ï : î : ì : Ä : Å |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | 9x ||  É : æ : Æ : ô : ö : ò : û : ù : ÿ : Ö : Ü : ¢ : £ : ¥ : ₧ : ƒ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Ax ||  á : í : ó : ú : ñ : Ñ : ª : º : ¿ : ⌐ : ¬ : ½ : ¼ : ¡ : « : » |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Bx ||  ░ : ▒ : ▓ : │ : ┤ : ╡ : ╢ : ╖ : ╕ : ╣ : ║ : ╗ : ╝ : ╜ : ╛ : ┐ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Cx ||  └ : ┴ : ┬ : ├ : ─ : ┼ : ╞ : ╟ : ╚ : ╔ : ╩ : ╦ : ╠ : ═ : ╬ : ╧ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Dx ||  ╨ : ╤ : ╥ : ╙ : ╘ : ╒ : ╓ : ╫ : ╪ : ┘ : ┌ : █ : ▄ : ▌ : ▐ : ▀ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Ex ||  α : ß : Γ : π : Σ : σ : µ : τ : Φ : Θ : Ω : δ : ∞ : φ : ε : ∩ |
    |----||----:---:---:---:---:---:---:---:---:---:---:---:---:---:---:---|
    | Fx ||  ≡ : ± : ≥ : ≤ : ⌠ : ⌡ : ÷ : ≈ : ° : ∙ : · : √ : ⁿ : ² : ■ :   |
    +======================================================================+


th¼ ≡≈c ≡▀■c, v╛i t∩ l« cao, trong m⌠i tr▀╢ng Vi«t ngµ.

    Cu»i c∙ng, m╡t s» ╤ng d°ng kh⌠ng hi¼n th╕ (render) ≡▀■c m╡t s» mπ t± trong
v∙ng G1 nh▀ nhµng chµ c≤ mπ  s»  160  (mπ  t±  cßch-dφnh  (non-breaking  space
character) trong 8859/Latin-1), 202 (mπ t± cßch-dφnh d∙ng trΩn mßy Macintosh),
hoúc 255.   Danh sßch nhµng mπ t± c≤ th¼ kh⌠ng hi¼n th╕ ≡▀■c c≤ th¼  rñt  dαi:
gÑn 30 mπ t± trong MS Windows 3.0 vα khoΣng 25 mπ t±  trong  MS  Windows  3.1.
TuΓn theo ph▀╜ng chΓm cªn thºn ≡π nΩu trΩn,  ch·ng t⌠i phΣi ≡út nhµng  mτu  t±
chµ hoa vαo nhµng v╕ trφ nαy.   Trong nhµng ╤ng d°ng cho  phΘp  chuy¼n  ph⌠ng,
vi«c hi¼n th╕ cßc chµ hoa c≤ th¼ ≡▀■c giΣi quy¬t bóng cßch cung  cñp  m╡t  cúp
ph⌠ng cho m▓i ki¼u chµ: ph⌠ng b∞nh th▀╢ng vα ph⌠ng chµ hoa.   Trong ph⌠ng  chµ
hoa,  tñt cΣ nhµng mτu t± ≡ßng l¿ lα chµ con ≡½u ≡▀■c bi¬n ≡▒i thαnh  chµ  hoa
t▀╜ng ╤ng.   Trong th±c t¬ khi gúp m╡t chµ hoa (thφ d° chµ á) kh⌠ng  th¼  hi¼n
th╕ ≡▀■c,  ng▀╢i d∙ng ch∩ cÑn chuy¼n qua ph⌠ng chµ hoa t▀╜ng ╤ng vα  ≡ßnh  vαo
chµ th▀╢ng (chµ ⌡).
                                                                        41
    Sau khi ≡π ≡╕nh ra nhµng nguyΩn tíc ch∩ ≡╒o nΩu trΩn, c⌠ng vi«c ch∩ c≥n lα
x¬p ≡út cßc mτu t± Vi«t c≥n l╒i theo m╡t l»i nαo ≡≤,  hay c≤ th¼ t∙y ti«n c√ng
≡▀■c.   Vi«c nαy ≡π ≡▀■c th±c hi«n sao cho bΣng mτu t± c≤ m╡t d╒ng  t▀╜ng  ≡»i
cΓn ≡»i, thªm m█.  K¬t quΣ lα tñt cΣ nguyΩn tíc ch∩ ≡╒o nΩu trΩn ≡½u ≡▀■c th÷a
mπn,  ngo╒i tr╫ m╡t ≡i½u lα v╕ trφ chµ á trong 8859/Latin-1 kh⌠ng th¼ duy  tr∞
≡▀■c.   CÑn ghi nhºn lα kh⌠ng c≤ cßch nαo c≤ th¼ bΣo toαn ≡▀■c th╤ t± cßc  mτu
t± Vi«t,  nh▀ng ≡Γy kh⌠ng phΣi lα m╡t vñn ≡½ l╛n v∞ th╤ t± cßc  mτu  t±  kh⌠ng
phΣi ASCII c≤ th¼ ≡▀■c giΣi quy¬t bóng cßch "tra bΣng."

    BΣn d± thΣo tºp mπ t± ti¬ng Vi«t VISCII 8-bit (BΣng 3) ≡π ≡▀■c h∞nh  thαnh
d±a trΩn cßc nguyΩn tíc nΩu trΩn.   Ch·ng t⌠i c≤ ² ≡╕nh xem ≡Γy lα m╡t bΣng mπ
t± duy nhñt ßp d°ng vαo m≈i vi«c s╪ d°ng dµ ki«n ti¬ng Vi«t nh▀  l▀u  trµ,  x╪
l², truy½n, vα mπ h≤a ph⌠ng chµ.  ╨i½u nαy s¿ ≡╜n giΣn h≤a rñt nhi½u quß tr∞nh
tφch h■p,  th±c hi«n,  vα s╪ d°ng,  vα thºt s± lα m╡t trong nhµng ≡i¼m son cⁿa
bΣn d± thΣo nαy.


4. VIQR: QUY ╨ÿNH VIÄT NG  ╨ÜC-╨┐öC-TRONG-NGOâC
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

4.1 ╨ôNG L╣C
~~~~~~~~~~~~

Trong khi quy ≡╕nh 8-bit ≡ang c» gíng tiΩu chuªn h≤a chµ Vi«t trong m⌠i tr▀╢ng
8-bit th∞ vτn c≥n rñt nhi½u vñn ≡½ phΣi  ≡▀■c  giΣi  quy¬t  trong  m⌠i  tr▀╢ng
7-bit,  ch╞ng h╒n nh▀ vi«c chuy¼n ≡i«n th▀ vα cßc ≡▀╢ng dΓy chuy¼n  tin  7-bit
khßc,  c√ng nh▀ cßc giao di«n ≡¼ t╒o chµ Vi«t c√ng cÑn phΣi  ≡▀■c  tiΩu  chuªn
h≤a.

    S± kh≤ khσn khi phΣi truy½n nhi½u  h╜n  128  mπ  s»  khßc  nhau  qua  kΩnh
(channel) 7-bit kh⌠ng phΣi lα m╡t vñn ≡½ riΩng cⁿa ti¬ng Vi«t.   Ngay  t╫  sau
khi Quy Luºt Chuy¼n ╨i«n Th▀ LiΩn L▀╛i ╨╜n GiΣn ("SMTP",  [8] ≡▀■c ≡½ ngh╕ nσm
1982,  ≡π c≤ nhi½u n▓ l±c khai tri¼n quy luºt nαy nhóm ≡ßp ╤ng nhu cÑu  chuy¼n
dµ ki«n 8-bit hoúc nhi½u h╜n cho nhµng chµ Latin ╖ ┬u ChΓu vα nhµng chµ  t▀■ng
h∞nh ╖ ╨⌠ng ph▀╜ng (xem [9] ch╞ng h╒n).   Múc d∙ nhu cÑu chuy¼n vºn 8-bit thºt
cÑn thi¬t,  cßc c▒ng chuy¼n ≡i«n th▀ kh⌠ng d¡ g∞ thay ≡▒i trong  m╡t  s╛m  m╡t
chi½u.   Trong t▀╜ng lai tr▀╛c mít,  ch·ng ta vτn c≤ nhu cÑu chuy¼n  ≡i«n  th▀
ti¬ng Vi«t minh b╒ch qua m╒ch 7-bit.

    QuΣ thºt ≡π c≤ m╡t tiΩu chuªn ≡úc bi«t d∙ng trΩn Viet-Net  vα  trong  nh≤m
th⌠ng tin Soc.Culture.Vietnamese trΩn m╒ng l▀╛i Usenet. ╨≤ lα cßch  d∙ng nhµng
k² t± thφch h■p d¡ nh╛ ≡i theo sau m╡t nguyΩn Γm ≡¼ t▀■ng tr▀ng  cho  dñu  ph°
(thφ d° nh▀ dñu ^ t▀■ng tr▀ng dñu m√);  ch╞ng h╒n,  "Vi«t Nam" ≡▀■c vi¬t thαnh
"Vie^.t Nam."  Tuy nhiΩn, quy tíc nαy kh⌠ng ≡▀■c r⌡ rαng b╖i v∞ m╡t s»  k²  t±
t▀■ng tr▀ng dñu ph° c≤ th¼ d∙ng lαm chñm cΓu;  thφ d° "tha?" c≤  th¼  hi¼u  lα
tha? hoúc thΣ.

    Quy ▀╛c cⁿa Viet-Net c√ng t▀╜ng t± nh▀ khßi ni«m "≡≈c ≡▀■c trong ngoúc" do
K. Simonsen [10,11] ≡½ ngh╕,  ≡π lαm sßng t÷ nhµng tr▀╢ng h■p m╜ h░  nh▀  trΩn
bóng cßch quy ≡╕nh thΩm tr╒ng thßi cho bΣn vσn ╖ cΣ cñp bºc mτu t± lτn cñp bºc
h« mτu t±.   Kh⌠ng may, trong n▓ l±c cung cñp m╡t giΣi phßp cho toαn th¬ gi╛i,
≡½ ngh╕ nαy kh⌠ng giΣi quy¬t thoΣ ≡ßng ti¬ng Vi«t.  ╨Ñu tiΩn, quy tíc nαy gi╛i
h╒n nhµng k² t± d¡ nh╛ trong tºp h■p 83 k² t± h∞nh c» ≡╕nh cⁿa  ISO-646  [12].
                                                                        42
╨i½u nαy t÷ ra t»t ≡⌐p trΩn nguyΩn tíc,  nh▀ng l╒i lαm cho chµ Vi«t  kh≤  ≡≈c.
Thφ d° dñu "h÷i" vα "ngπ" ≡▀■c quy ≡╕nh lÑn l▀■t lα 2 vα ?,  ≡¼ trßnh d∙ng dñu
~,  v∞ dñu nαy kh⌠ng phΣi lα m╡t k² t± c» ≡╕nh.   S± ph▒ bi¬n r╡ng rπi cⁿa cßc
bαn ≡ßnh chµ ASCII trong ≡a s» cßc ng▀╢i d∙ng chµ Vi«t cho thñy vi«c gi╛i  h╒n
nαy kh⌠ng h■p l².   C√ng xin nhñn m╒nh lα ch·ng t⌠i ≡ang bi«n h╡ quan ≡i¼m "d¡
≡≈c cho ≡a s» ng▀╢i d∙ng" h╜n lα "kh≤ ≡≈c cho tñt cΣ ng▀╢i  d∙ng."   H╜n  nµa,
v╛i ≡α ti¬n cⁿa vi«c qu»c t¬ h≤a cßc bαn ≡ßnh chµ vα mαn Σnh, thφ d° trong m⌠i
tr▀╢ng bi¼u h≈a khung (graphical window environment),  vi«c tßi ≡╕nh nghεa cßc
phφm chµ vα thay ≡▒i ph⌠ng c≤ th¼ ≡▀■c th±c hi«n d¡ dαng khi¬n  cho  gi╛i  h╒n
trΩn cαng ngαy cαng l▓i th╢i.

    ╨i¼m kh≤ khσn l╛n h╜n cⁿa quy tíc trΩn lα ph▀╜ng phßp mπ h≤a bóng  hai  k²
t± (chi½u dαi c» ≡╕nh) {7} lαm cho chµ Vi«t kh≤ ≡≈c, nhñt lα nhµng mτu t± Vi«t
c≤ hai dñu ph° (thφ d° ñ).   Ph▀╜ng phßp mπ h≤a d∙ng nhi½u k²  t±  (chi½u  dαi
thay ≡▒i) {8} c√ng kh≤ ≡≈c vα kh⌠ng c≤ hi«u nσng v∞ ch╤a ≡Ñy ngh⌐t nhµng k² t±
d∙ng ≡¼ m╖ vα ≡≤ng trong khi Vi«t ngµ l╒i d∙ng nhi½u dñu.   Múc d∙ mßy vi tφnh
c≤ th¼ ≡≈c d¡ dαng bñt c╤ ph▀╜ng phßp "d¡ nh╛" nαo,  ph▀╜ng phßp ßp  d°ng  cho
ng▀╢i d∙ng phΣi lα ph▀╜ng phßp khi¬n h≈ c≤ th¼ ≡≈c vα vi¬t m╡t  cßch  d¡  dαng
khi d∙ng nhµng nhu li«u vi¬t bαi (editor) 7-bit.   Ng▀╢i d∙ng Vi«t  ngµ  kh⌠ng
mu»n phΣi h≈c hoúc nh╛ nhµng chu▓i k² t± nh▀ "a5" t▀■ng tr▀ng cho í, hoúc phΣi
≡ßnh nhµng chu▓i dαi nh▀ "&_a('_" ≡¼ t▀■ng tr▀ng cho m╡t mτu t±  Vi«t  nαo  ≡≤
trong cΣ m╡t bαi dαi.

    ╨¼ th÷a mπn nhu cÑu d¡ ≡≈c vα uy¼n chuy¼n,  ch·ng ta cÑn ñn ≡╕nh  m╡t  quy
tíc khßc.   Cßch t»t h╜n  lα  d∙ng  ph▀╜ng  phßp  chuy¼n  mπ-t±-h«  (code-page
switching) nh▀ quy ≡╕nh ISO-2022 [13] ≡¼ chuy¼n vσn bΣn vαo  tr╒ng  thßi  Vi«t
ngµ vα t»i ▀u h≤a vi«c mπ h≤a tu╧ theo tr╒ng thßi ng⌠n ngµ.   GÑn ≡Γy, van der
Poel ≡½ x▀╛ng m╡t ph▀╜ng-phßp d¡ nh╛ [14],  nhñn m╒nh v½ nhµng quy  ▀╛c  riΩng
cⁿa t╫ng ng⌠n ngµ.   ╨½ ngh╕ nαy cung cñp m╡t ph▀╜ng ti«n ≡¼  quy  ≡╕nh  tr╒ng
thßi ng⌠n ngµ,  v╛i m▓i ng⌠n ngµ t± quy ≡╕nh lñy cßch mπ h≤a sao cho hµu  hi«u
nhñt.   L■i ≡i¼m chφnh cⁿa n≤ lα nhµng ╤ng d°ng d±a theo ph▀╜ng phßp nαy kh⌠ng
cÑn phΣi t╒o h∞nh cho tñt cΣ cßc tºp mπ t± ≡▀■c ch∩ ≡╕nh trong  d≥ng  tin,  mα
ch∩ cÑn t∙y nghi bßo tin v½ nhµng ng⌠n ngµ kh⌠ng ≡▀■c h▓ tr■ nh▀ "╖ ≡Γy c≤ chµ
Hy L╒p kh⌠ng th¼ t╒o h∞nh ≡▀■c" (xin xem [14] ≡¼ bi¬t thΩm chi ti¬t chφnh  xßc
cⁿa quy ≡╕nh).   Ph▀╜ng phßp nαy cho phΘp m▓i c╡ng ≡░ng d∙ng cßch  th╤c  riΩng
t»t nhñt ≡¼ mπ h≤a ng⌠n ngµ cⁿa m∞nh.  Quy ▀╛c VIQR ph∙ h■p v╛i ≡▀╢ng l»i nαy,
vα c≤ th¼ ≡▀■c sßt nhºp d¡ dαng vαo khu⌠n kh▒ nαy.

    Nhµng quy ≡╕nh ≡út ra ╖ ≡Γy s¿ ßp d°ng vαo m≈i d≥ng dµ ki«n, g░m c≤ chuy¼n
chµ (text transfer), h░ s╜ xuñt nhºp (file I/O), vα cßch ≡ßnh chµ.  NguyΩn tíc
nαy lα nguyΩn nhΓn chφnh ≡▀a ≡¬n s± thαnh c⌠ng cⁿa khi¼n  h«  Unix,  trong  ≡≤
ng▀╢i vi¬t nhu li«u kh⌠ng phΣi quan tΓm ≡¬n cßc chi ti¬t ≡úc th∙  cⁿa  cßc  b╡
phºn ph°,  mα ch∩ quan tΓm ≡¬n m╡t giao di«n ≡░ng nhñt (uniform interface)  ≡¼
t╫ ≡≤ c≤ th¼ chia xδ cßc nhu li«u th▀ vi«n cⁿa khi¼n h«.  Do ≡≤ ≡i½u cÑn thi¬t
lα cung cñp m╡t cσn bΣn chung  ≡¼  t╫  ≡≤  xΓy  d±ng  nhµng  phÑn  d╕ch  (data
interpreter) cho m≈i d≥ng dµ ki«n bñt k¼ xuñt x╤.   TrΩn th±c t¬,  ≡i½u nαy ≡π
gi·p cho vi«c phßt tri¼n nhµng nhu li«u d∙ng chµ Vi«t ≡▀■c d¡ dαng rñt nhi½u.


------------------------------------------------------------------------------
    {7} Cßch d∙ng lα "&xy", mα x lα chµ chφnh vα y lα chµ ph° ≡¼ ghΘp v╛i x.

    {8} Cßch d∙ng lα "&_xxxx_", mα xxxx c≤ th¼ lα bñt c╤ chu▓i k² t± nαo.
                                                                        43
    Ngoαi ra,  ng▀╢i d∙ng ≡▀■c h▀╖ng l■i φch l╛n t╫ vi«c tiΩu chuªn  h≤a  cßch
≡ßnh chµ.   H≈ kh⌠ng cÑn phΣi h≈c nhi½u cßch ≡ßnh chµ khßc nhau  khi  x╪  d°ng
nhµng nhu li«u khßc nhau.   N¬u tñt cΣ cßc nhu li«u c∙ng h▓ tr■ m╡t tiΩu chuªn
chung,  ng▀╢i d∙ng ≡π quen v╛i tiΩu chuªn nαy c≤ th¼ ≡ßnh ti¬ng Vi«t  ngay  mα
kh⌠ng cÑn phΣi h≈c l╒i.   TiΩu chuªn trong bαi nαy quy ≡╕nh nhµng ≡úc tφnh t»i
thi¼u mα cßc nhu li«u h▓ tr■ n≤ cÑn phΣi c≤;  dε nhiΩn cßc k█ thuºt  ≡ßnh  chµ
khßc c≤ th¼ ≡▀■c sßt nhºp c∙ng v╛i tiΩu chuªn nαy thαnh m╡t quy ≡╕nh t▒ng qußt
h╜n.   ╨i½u nαy s¿ ≡▀■c bαn luºn thΩm nµa trong PhÑn 5.2 n≤i v½ cßch ≡ßnh  chµ
Vi«t.

4.2 QUY ╨ÿNH "╨ÜC-╨┐öC-TRONG-NGOâC" (VIQR)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Quy ≡╕nh nαy hoαn toαn s╪ d°ng ki¼u mτu "d¡ ≡≈c" cⁿa Viet-Net. Quy  ≡╕nh  Vi«t
ngµ "≡≈c ≡▀■c trong ngoúc," VIQR,  g░m c≤ ba tr╒ng thßi: nguyΩn d╒ng,  Anh Ngµ
vα Vi«t Ngµ.   Tr╒ng thßi nguyΩn d╒ng chⁿ ² ≡¼ chuy¼n tin y nguyΩn, kh⌠ng thay
≡▒i (ngo╒i tr╫ nhµng chu▓i thoßt (escape sequence) ≡¼ m╖ ≡Ñu vα k¬t th·c tr╒ng
thßi nguyΩn d╒ng). Tr╒ng thßi Anh Ngµ vα Vi«t Ngµ ≡▀■c d∙ng chⁿ y¬u cho  nhµng
d≥ng dµ ki«n c≤ pha tr╡n Anh ngµ vα Vi«t ngµ,  v╛i m▓i tr╒ng thßi ≡▀■c t»i  ▀u
h≤a v½ h∞nh th╤c c√ng nh▀ kh»i l▀■ng t∙y theo vσn bΣn ch╤a ≡a s»  lα  Anh  ngµ
hoúc Vi«t ngµ.   M▓i tr╒ng thßi ≡½u c≤ c╜ ch¬ riΩng ≡¼ vi¬t mτu t± Vi«t,  bóng
cßch d∙ng m╡t hoúc hai dñu ph° theo sau nguyΩn Γm cσn bΣn.

    Tr▀╛c h¬t xin gi╛i thi«u khßi ni«m t╒o chµ ngÑm (implicit composition)  vα
t╒o chµ ch∩ ≡╕nh (explicit composition).

4.2.1 PhΘp T╒o Chµ NgÑm
~~~~~~~~~~~~~~~~~~~~~~~

PhΘp t╒o chµ ngÑm th▀╢ng ≡▀■c d∙ng m╡t cßch hµu hi«u cho nhµng  dµ  ki«n  phÑn
l╛n lα chµ Vi«t.  Trong phΘp t╒o chµ ngÑm, m▓i khi m╡t hay hai dñu ph° ≡i theo
sau m╡t nguyΩn Γm c╜ bΣn th∞ ch·ng s¿ k¬t h■p v╛i nguyΩn Γm ≡≤ thαnh  m╡t  mτu
t± Vi«t duy nhñt sao cho ph∙ h■p v╛i quy tíc vσn ph╒m.  Thφ d°:

                        a^      --->      Γ
                        o+?     --->      ╖
                        ╜?      --->      ╖
                        Vie^.t  --->      Vi«t
                        ViΩ.t   --->      Vi«t
                        la'^n   --->      lß^n (kh⌠ng phΣi lñn)
                        lß^n    --->      lß^n (kh⌠ng phΣi lñn)

    Trong hai thφ d° ch≤t,  chu▓i a^' kh⌠ng t▀╜ng ≡▀╜ng v╛i a'^ hay ß^ v½  mút
vσn ph╒m.   Th⌠ng th▀╢ng, ba k² t± ph° (, ^, vα + phΣi theo sßt sau cßc nguyΩn
Γm thφch h■p th∞ m╛i k¬t h■p ≡▀■c.

    Chu▓i ≡úc bi«t dd k¬t h■p thαnh ≡; DD, dD, vα Dd ≡½u t▀■ng tr▀ng cho ╨.

    Nhµng nguyΩn Γm c╜ bΣn g░m c≤ a, σ, Γ, e, Ω, i, o, ⌠, ╜, u, ▀, y, vα nhµng
mτu t± hoa t▀╜ng ╤ng. Mπ s» cⁿa nhµng mτu t± nαy ≡▀■c li«t kΩ trong BΣng 3, D±
ThΣo TiΩu Chuªn 8-bit VISCII.

    Nhµng dñu ti¬ng Vi«t ≡▀■c t▀■ng tr▀ng bóng nhµng k² t± ASCII c≤ h∞nh  d╒ng
                                                                        44
      BΣng 4.  K² t± ASCII d¡ nh╛ d∙ng thay th¬ dñu ti¬ng Vi«t

    ==============================================================
    Dñu        K² t±    Mπ s» ASCII                Thφ d°
    --------------------------------------------------------------
    trσng (ß)    (      0x28, dñu m╖ ngoúc         ba(n khoa(n
    m√           ^      0x5E, dñu m√               ho^m nay
    m≤c          +      0x2B, dñu c╡ng             Qui Nho+n

    síc          '      0x27, ngoúc ≡╜n            La'i Thie^u
    huy½n        `      0x60, ngoúc ≡╜n ng▀■c      Bi`nh Du+o+ng
    h÷i          ?      0x3F, dñu h÷i              Thu? DDu+'c
    ngπ          ~      0x7E, dñu ngπ              di~ va~ng
    núng         .      0x2E, dñu chñm             ho.c ta^.p
    ==============================================================


t▀╜ng t±.  BΣng 4 li«t kΩ 7 k² t± ASCII d¡ nh╛ ≡▀■c d∙ng ≡¼ thay th¬ nhµng dñu
ti¬ng Vi«t.   Ph° l°c A vα B li«t kΩ, theo th╤ t± síp chµ vα th╤ t± mπ s», tñt
cΣ mτu t± Vi«t vα chu▓i VIQR t▀╜ng ╤ng.

4.2.2 PhΘp T╒o Chµ Ch∩ ╨╕nh
~~~~~~~~~~~~~~~~~~~~~~~~~~~

PhΘp t╒o chµ ch∩ ≡╕nh d±a trΩn khßi ni«m d∙ng m╡t k² t± ≡i tr▀╛c  ≡¼  bßo  tin
vi«c t╒o chµ m╡t cßch r⌡ rαng.   K² t± bßo tin lα g╒ch-chΘo-ng▀■c ("\",  mπ s»
ASCII 0x5C), t╫ nay s¿ g≈i lα k² t± <COM> {*}.   Nhµng k² t± ≡i theo sau n≤ s¿
≡▀■c k¬t h■p theo c∙ng quy tíc vσn ph╒m nh▀ phΘp t╒o chµ gißn  ti¬p.   Do  ≡≤,
nhµng thφ d° ╖ phÑn trΩn s¿ hi«n ra nh▀ sau khi d∙ng phΘp t╒o chµ ch∩ ≡╕nh:

             a^         --->    Γ
             o+?        --->    ╖
             Vie^.t     --->    Vi«t

    PhΘp t╒o chµ ch∩ ≡╕nh t÷ ra ti«n l■i trong d≥ng dµ ki«n mα phÑn l╛n lα Anh
ngµ,  ≡░ng th╢i c√ng thφch h■p v╛i cßch ≡ßnh chµ th±c th╢i ≡▀■c ≡½ cºp ╖  PhÑn
5.2.

    Sau ≡Γy, ch·ng ta s¿ phΓn tφch cßch x╪ d°ng hai phΘp t╒o chµ trΩn trong ba
tr╒ng thßi. Tr╒ng thßi cⁿa d≥ng dµ ki«n ≡▀■c ch∩ ≡╕nh bóng chu▓i g░m hai k² t±
<COM>x, v╛i x ≡▀■c quy ≡╕nh nh▀ sau ≡Γy.

4.2.3 Tr╒ng Thßi NguyΩn D╒ng
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

S± xuñt hi«n cⁿa <COM>L or <COM>l trong d≥ng dµ ki«n m╖ ≡Ñu tr╒ng thßi  nguyΩn
d╒ng (hay nguyΩn tr╒ng).   M°c ≡φch lα ≡¼ chuy¼n vºn dµ ki«n trong tr╒ng  thßi
hÑu nh▀ nguyΩn v⌐n kh⌠ng bi¬n ≡▒i.   CΣ hai phΘp t╒o chµ ngÑm vα ch∩ ≡╕nh  ≡½u

------------------------------------------------------------------------------
    {*} K² hi«u <...> ≡▀■c d∙ng ≡¼ nhñn m╒nh róng cΣ chu▓i <...> ch∩ th¼  hi«n
m╡t byte trong b╡ nh╛ hoúc kho ch╤a; nghεa lα ch∩ t▀╜ng ╤ng v╛i m╡t vα ch∩ m╡t
k² t± mα th⌠i.  Thφ d°, chµ H┴T c≥n c≤ th¼ ≡▀■c vi¬t lα <H><┴><T>.
                                                                        45
kh⌠ng ≡▀■c ßp d°ng ╖ ≡Γy, k¼ cΣ k² t± <COM> c√ng kh⌠ng c≤ ² nghεa ≡úc bi«t tr╫
khi k² t± nαy ≡▀■c theo sau b╖i m╡t trong sßu chµ l, L, v, V,  m,  hoúc M,  v∞
l·c ≡≤ n≤ s¿ m╖ ≡Ñu m╡t trong ba tr╒ng thßi {9}.

4.2.4 Tr╒ng Thßi Anh Ngµ
~~~~~~~~~~~~~~~~~~~~~~~~

Tr╒ng thßi Anh ngµ ≡▀■c bít ≡Ñu bóng chu▓i <COM>M  hay  <COM>m.   Trong  tr╒ng
thßi Anh ngµ, ch∩ c≤ phΘp t╒o chµ ch∩ ≡╕nh ≡▀■c h▓ tr■.   ╨i½u nαy c≤ nghεa lα
≡¼ t╒o ra m╡t chµ Vi«t, ta cÑn phΣi d∙ng k² t± bßo tin <COM>.  N¬u chu▓i k² t±
kh⌠ng ≡▀■c m╖ ≡Ñu bóng <COM>, chu▓i nαy s¿ kh⌠ng ≡▀■c phΘp k¬t h■p.  Thφ d°:

        \mD\u~ng, how are you?  --->    D√ng, how are you?
        \mKho\e? kh\o^ng?       --->    Khoδ kh⌠ng?

    Nh▀ ≡π n≤i, chu▓i "you?" kh⌠ng ≡▀■c phΘp ≡▒i thαnh "yoⁿ" v∞ kh⌠ng c≤ k² t±
bßo tin <COM> ≡i tr▀╛c mτu t± u.

4.2.4 Tr╒ng Thßi Vi«t Ngµ
~~~~~~~~~~~~~~~~~~~~~~~~~

Chu▓i <COM>V hay <COM>v chuy¼n tr╒ng thßi cⁿa d≥ng dµ  ki«n  sang  tr╒ng  thßi
Vi«t ngµ.   ù tr╒ng thßi nαy,  ta c≤ th¼ d∙ng cΣ hai phΘp t╒o chµ ngÑm vα  ch∩
≡╕nh. Nhµng thφ d° sau ≡Γy d±a trΩn giΣ thi¬t tr╒ng thßi ban ≡Ñu cⁿa  d≥ng  dµ
ki«n lα tr╒ng thßi Anh ngµ:

        \vChu+~ Vie^.t          --->    Chµ Vi«t
        \vCh\u+~ Vi\e^.t        --->    Chµ Vi«t
        Chu+~ \vVie^.t          --->    Chu+~ Vi«t

    Tr╒ng thßi Vi«t ngµ d∙ng phΘp t╒o chµ ngÑm hÑu gi·p cho vσn  bΣn  gπy  g≈n
h╜n v∞ kh⌠ng phΣi ch╤a nhi½u k² t± bßo tin <COM> m╡t cßch r▀╢m rα v⌠  φch  nh▀
trong tr▀╢ng h■p t╒o chµ ch∩ ≡╕nh.   Ngoαi ra,  tr╒ng thßi Vi«t ngµ  c√ng  cho
phΘp t╒o chµ ch∩ ≡╕nh ≡¼ duy tr∞ s± t▀╜ng h■p (compatibility) v╛i  tr╒ng  thßi
Anh ngµ nhóm trßnh vi«c quy ≡╕nh thΩm ² nghεa cⁿa nhµng chu▓i bít ≡Ñu bóng  k²
t± <COM>. Ngoαi ra, ph▀╜ng phßp t╒o chµ ch∩ ≡╕nh c√ng duy tr∞ s± t▀╜ng h■p v╛i
cßch ≡ßnh chµ th±c th╢i (real-time keyboarding).

4.2.6 NguyΩn T± trong Tr╒ng Thßi Anh Ngµ vα Vi«t Ngµ
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Hπy xΘt thφ d° sau:

      \vDu~ng, how are you?    --->    D√ng, how are yoⁿ

    Trong thφ d° nαy,  chu▓i "you?" tr╖ thαnh "yoⁿ" v∞ d≥ng  dµ  ki«n  ≡ang  ╖
tr╒ng thßi Vi«t ngµ.  V∞ th¬ ta thñy ≡⌠i khi cÑn t╒m ng▀ng vi«c k¬t h■p chµ mα
kh⌠ng phΣi chuy¼n tr╒ng thßi.   Tφnh chñt "nguyΩn d╒ng" cⁿa mτu t±  <COM>  tr╖

------------------------------------------------------------------------------
    {9} ╨¼ c≤ ≡▀■c chφnh chµ <COM>L,  <COM>M,  hoúc <COM>V ta cÑn phΣi  chuy¼n
sang tr╒ng thßi Vi«t hoúc Anh vα d∙ng ≡úc ≡i¼m "nguyΩn d╒ng"  quy  ≡╕nh  trong
nhµng tr╒ng thßi ≡≤. Xem 4.2.6
                                                                        46
thαnh ti«n d°ng ╖ ≡Γy.   Trong cΣ hai tr╒ng thßi Vi«t ngµ vα Anh ngµ,  m▓i khi
k² t± <COM> ≡▀■c theo sau b╖i m╡t k² t± kh⌠ng th¼ k¬t h■p c,  k¬t  quΣ  s¿  lα
chφnh k² t± c c≥n k² t± gi╛i thi«u <COM> s¿ b╕ lo╒i b÷ kh÷i d≥ng dµ ki«n.   ╨¼
c≤ k² t± <COM>, d∙ng <COM><COM>. Hπy xem nhµng thφ d° sau:

                \vddi dda^u?       --->        ≡i ≡Γⁿ
                \vddi dda^u\?      --->        ≡i ≡Γu?
                \m\ddi v\o^?       --->        ≡i v▒
                \m\ddi v\o^\?      --->        ≡i v⌠?
                \\                 --->        \
                \\V                --->        \V
                \\M                --->        \M
                \\L                --->        \L

4.2.7 K² T± Hoαn Cñu
~~~~~~~~~~~~~~~~~~~~

D≥ng dµ ki«n c≤ th¼ ch╤a m╡t k² t± ≡úc bi«t g≈i lα k² t± hoαn cñu (k² t±  hoαn
tñt vi«c cñu t╒o chµ,  closure character) d∙ng ≡¼ k¬t th·c vi«c  k¬t  h■p  chµ
≡ang di¡n ti¬n.   K² t± nαy lα CTRL-A (ASCII 0x1),  t╫ nay g≈i lα <CLS>.   Khi
gúp <CLS> trong d≥ng dµ ki«n, tñt cΣ nhµng vi«c k¬t h■p chµ ≡ang ti¬n hαnh ≡½u
≡▀■c k¬t th·c.   <CLS> lu⌠n lu⌠n b╕ lo╒i b÷,  tr╫ khi n≤ xuñt hi«n trong chu▓i
nguyΩn d╒ng <COM><CLS>.

    K² t± hoαn cñu c≤ φch trong nhµng ╤ng d°ng th±c th╢i nh▀ ≡ßnh chµ, khi cÑn
phΣi cho bi¬t lα chu▓i k¬t h■p ≡π k¬t th·c, vα c╜ phºn nhºn tin kh⌠ng cÑn phΣi
ch╢ thΩm dµ ki«n nµa.


5. C┴C ║NG D₧NG ╨âC BIÄT
~~~~~~~~~~~~~~~~~~~~~~~~

PhÑn nαy phßc h≈a nhµng nguyΩn tíc ch∩ ≡╒o vα quy ▀╛c ≡úc th∙  cho  nhµng  ╤ng
d°ng ≡π ≡▀■c d∙ng trong gi╛i phßt tri¼n nhu li«u.  M°c ≡φch cⁿa n≤ lα cung cñp
m╡t tαi li«u s»ng bao g░m nhµng kinh nghi«m tφch l√y trong th╢i gian qua  c√ng
nh▀ nhµng kinh nghi«m síp t╛i.   Ch·ng t⌠i hoan nghΩnh ≡╡c giΣ  tham  gia  vαo
nhµng cu╡c thΣo luºn nαy vα c»ng hi¬n vαo vi«c phßt tri¼n nhµng nguyΩn tíc ch∩
≡╒o n≤i riΩng, vα vi«c tiΩu chuªn h≤a n≤i chung.

5.1 ╨IÄN TH┐ CHUYîN QUA MÇCH 7-BIT
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

╨a s» nhµng m╒ch hi«n hµu d∙ng ≡¼ chuy¼n ≡i«n th▀ vτn c≥n  ╖  trong  gi╛i  h╒n
7-bit.   Tºp mπ t± 8-bit quy ≡╕nh ╖ PhÑn 3 kh⌠ng th¼ ≡▀■c truy½n  nguyΩn  d╒ng
qua nhµng m╒ch nαy.   Do ≡≤ VIQR ≡≤ng m╡t vai quan tr≈ng v∞ n≤ c≤ th¼ d∙ng  ≡¼
chuy¼n vσn bΣn ti¬ng Vi«t 7-bit m╡t cßch minh b╒ch,  kh⌠ng b╕ m╜  h░  v∞  tφnh
cßch l▀▐ng d°ng cⁿa nhµng k² t± v╫a t▀■ng tr▀ng dñu ph° v╫a  t▀■ng  tr▀ng  dñu
chñm cΓu nh▀ dñu "?".   Do tφnh chñt 7-bit cⁿa cßc m╒ch nαy,  b╡ chuy¼n th▀ s¿
kh⌠ng gúp phΣi nhµng mτu t± Vi«t nóm trong G1 nh▀ σ, ┼, Γ, ┬, Ω, ╩, ⌠,  ╘,  ╜,
┤, ▀ vα ┐.  Tuy nhiΩn, b╡ chuy¼n th▀ ch¬ t╒o cho m╒ch 8-bit s¿ phΣi giΣi quy¬t
nhµng mτu t± nαy ≡·ng theo quy tíc k¬t h■p VIQR,  nghεa lα phΣi k¬t h■p nguyΩn
Γm c╜ bΣn vα dñu ph° n¬u ≡▀■c.  Vφ d°:

                                                                        47
                    σ'     --->      í

    ╨¼ ≡▀■c hi¼u ≡·ng,  ≡i«n th▀ phΣi ch∩ ≡╕nh r⌡ rαng tr╒ng  thßi  ng⌠n  ngµ,
hoúc ╖ trong phÑn dτn ≡Ñu (header), hoúc ╖ trong phÑn n╡i dung (text body) cⁿa
th▀.   Ch·ng ta kh⌠ng th¼ ph÷ng ≡oßn tr╒ng thßi cⁿa b╡ phºn nhºn tin ╖ ≡Ñu m▓i
lß th▀,  v∞ th▀ c≤ th¼ ≡▀■c ≡≈c t╫ m╡t h░ s╜ (file) g░m nhi½u lß (message) ch╤
kh⌠ng phΣi ch∩ c≤ m╡t lß, do ≡≤ kh≤ bi¬t ≡Γu lα ch▓ bít ≡Ñu cⁿa lß th▀ khßc.

    H╜n nµa,  n¬u trong th▀ c≤ ch╤a m╡t chu▓i ch∩ ≡╕nh  tr╒ng  thßi  ng⌠n  ngµ
(<COM>L,  <COM>V hoúc <COM>M),  th∞ lß th▀ nΩn ≡▀■c k¬t th·c trong tr╒ng  thßi
nguyΩn d╒ng,  nghεa lα k¬t th·c bóng <COM>L.   Vi«c nαy gi·p cho ╤ng d°ng  ≡≈c
th▀ ≡≈c ≡▀■c nhµng lß th▀ sau nóm c∙ng trong m╡t h░ s╜, ch╞ng h╒n ╤ng d°ng ≡≈c
th▀ trΩn mαn Σnh.   ╨i½u nαy t÷ ra φch l■i v∞ phÑn dτn ≡Ñu cⁿa  ≡i«n  th▀  n≤i
chung kh⌠ng tuΓn theo quy tíc VIQR,  vα do ≡≤ c≤ th¼ b╕ hi¼u sai khi  kh⌠ng  ╖
trong tr╒ng thßi nguyΩn d╒ng.

5.2 ╨┴NH CH  VIÄT
~~~~~~~~~~~~~~~~~

Bαn ≡ßnh chµ cαng ngαy cαng ≡▀■c qu»c t¬ h≤a thΩm.  Nh▀ ≡π n≤i ╖ phÑn quy ≡╕nh
8-bit,  ≡Γy lα m╡t l² do chφnh ≡¼ d∙ng c∙ng m╡t mπ s» cho nhµng mτu t± Vi«t ≡π
c≤ s╟n trong bΣng ISO 8859/Latin-1.  B╡ ≡i½u khi¼n bαn chµ Vi«t, ≡▀■c thi¬t k¬
≡¼ d∙ng trong m⌠i tr▀╢ng 7-bit mα th⌠i,  c≤ th¼ giΣ s╪ lα s¿ kh⌠ng bao gi╢ gúp
nhµng nguyΩn Γm c╜ bΣn cⁿa ti¬ng Vi«t nóm trong v∙ng G1.   Nh▀ng b╡ ≡i½u khi¼n
bαn chµ d∙ng trong m⌠i tr▀╢ng 8-bit,  c√ng nh▀ b╡ nhºn tin 8-bit  (PhÑn  5.1),
phΣi s╟n sαng ti¬p nhºn bñt c╤ nguyΩn Γm c╜ bΣn nαo, k¼ cΣ nhµng nguyΩn Γm nóm
trong G1.

    Vi«c t╒o h∞nh mτu t± trΩn mαn Σnh (echoing behavior) khi ≡ang k¬t h■p  chµ
t╫ bαn ≡ßnh chµ cÑn ≡▀■c quy ≡╕nh thΩm.   Ch·ng ta c≤ th¼ t╒o h∞nh cho mτu  t±
ch∩ sau khi vi«c k¬t h■p ≡π hoαn tñt.   Ch·ng ta c√ng c≤ th¼ t╒o h∞nh cho  tñt
cΣ nhµng d╒ng trung gian,  h∞nh cⁿa d╒ng sau ≡▀■c t╒o ra  bóng  cßch  tr╖  lui
(backspace) ≡¼ x≤a r░i in ch░ng lΩn d╒ng tr▀╛c.   M▓i cßch  ≡½u  c≤  hµu  d°ng
riΩng nh▀ s¿ m⌠ tΣ sau ≡Γy.

5.2.1 Cßch T╒o H∞nh Lºp T╤c Trong PhΘp T╒o Chµ NgÑm
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

PhΘp t╒o chµ ngÑm ≡▀■c ≡út ra ≡¼ ti«n cho vi«c x╪ l² nhµng dµ ki«n mα phÑn l╛n
lα Vi«t ngµ.   V╛i m°c tiΩu ≡≤,  ng▀╢i ≡ßnh chµ cÑn thñy ngay  nhµng  chµ  v╫a
≡ßnh.   Trong phΘp t╒o chµ ngÑm,  bαn ≡ßnh chµ ho╒t ≡╡ng trong tr╒ng thßi  t╒o
h∞nh lºp t╤c.  M▓i phφm chµ ≡▀■c nhñn (keypress) s¿ lºp t╤c t╒o ra m╡t bi¬n c»
phφm chµ (key event).   N¬u m╡t mτu t± (thφ d° a) k¬t h■p v╛i m╡t dñu ph° (thφ
d° ^) theo sau n≤, m╡t thoßi t± (backspace) (th▀╢ng lα BS,  ASCII 0x8) s¿ ≡▀■c
g╪i ≡i kΦm theo sau lα mτu t± m╛i v╫a thαnh lºp (Γ).   Chu k╧ nαy tßi di¡n cho
≡¬n khi vi«c k¬t h■p chµ hoαn tñt.  Trong cßch t╒o h∞nh lºp t╤c, nhµng bi¬n c»
t╒o ra do vi«c nhñn 4 phφm chµ a^'n lα:

       1. Ng▀╢i d∙ng nhñn phφm a, a ≡▀■c g╪i cho ╤ng d°ng
       2. Ng▀╢i d∙ng nhñn phφm ^, BS vα Γ ≡▀■c g╪i ≡i
       3. Ng▀╢i d∙ng nhñn phφm ', BS vα ñ ≡▀■c g╪i ≡i
       4. Ng▀╢i d∙ng nhñn phφm n, n ≡▀■c g╪i ≡i

                                                                        48
    Thoßi t± c≤ th¼ ≡▀■c thay ≡▒i t∙y theo ╤ng d°ng,  khi¼n h«,  vα m⌠i tr▀╢ng
cⁿa ng▀╢i d∙ng.  B╡ ≡i½u khi¼n cⁿa bαn ≡ßnh chµ nΩn d∙ng ≡·ng thoßi t±, vα/hay
cho phΘp ng▀╢i d∙ng ch∩ ≡╕nh thoßi t± theo ² thφch.

5.2.2 Cßch T╒o H∞nh Chºm Trong PhΘp T╒o Chµ Ch∩ ╨╕nh
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Khi vi«c k¬t h■p chµ m╛i bít ≡Ñu, b╡ ≡i½u khi¼n cⁿa bαn ≡ßnh chµ kh⌠ng g╪i cho
╤ng d°ng t╫ng bi¬n c» phφm chµ mα phΣi ch╢ cho ≡¬n khi vi«c k¬t h■p chñm  d╤t.
Vi«c k¬t h■p c≤ th¼ chñm d╤t m╡t cßch t± nhiΩn khi chu▓i k² t± k¬t h■p ≡π  ≡Ñy
≡ⁿ hoúc khi b╡ ≡i½u khi¼n nhºn ≡▀■c k² t± kh⌠ng th¼ k¬t h■p  ≡▀■c,  hoúc  chñm
d╤t khi nhºn ≡▀■c mπ t± hoαn cñu <CLS>.   Sau ≡≤,  ch∩ c≤ m╡t mτu t± duy  nhñt
≡▀■c t╒o thαnh vα ≡▀■c g╪i cho ╤ng d°ng ≡ang ch╢.   Vi«c x╪ l² sau ≡≤ s¿  ti¬n
hαnh t± nhiΩn tr╖ l╒i.   Hπy xem di¡n ti¬n khi  ng▀╢i  d∙ng  nhñn  chu▓i  phφm
"\a^'n":

       1. Ng▀╢i d∙ng ≡ßnh \, kh⌠ng c≤ chµ nαo ≡▀■c g╪i ≡i
       2. Ng▀╢i d∙ng ≡ßnh a, kh⌠ng c≤ chµ nαo ≡▀■c g╪i ≡i
       3. Ng▀╢i d∙ng ≡ßnh ^, kh⌠ng c≤ chµ nαo ≡▀■c g╪i ≡i
       4. Ng▀╢i d∙ng ≡ßnh ', chµ ñ ≡▀■c g╪i ≡i
       5. Ng▀╢i d∙ng ≡ßnh n, chµ n ≡▀■c g╪i ≡i

    Vφ d° sau ≡Γy d∙ng mπ t± hoαn cñu <CLS>, chu▓i "t\o+<CLS>":

       1. Ng▀╢i d∙ng ≡ßnh t, chµ t ≡▀■c g╪i ≡i
       2. Ng▀╢i d∙ng ≡ßnh \, kh⌠ng c≤ chµ nαo ≡▀■c g╪i ≡i
       3. Ng▀╢i d∙ng ≡ßnh o, kh⌠ng c≤ chµ nαo ≡▀■c g╪i ≡i
       4. Ng▀╢i d∙ng ≡ßnh +, kh⌠ng c≤ chµ nαo ≡▀■c g╪i ≡i
       5. Ng▀╢i d∙ng ≡ßnh CTRL-A, m╡t chµ ╜ ≡▀■c g╪i ≡i

    ╨¼ ² lα n¬u kh⌠ng c≤ phφm hoαn cñu <CLS>,  b╡ ≡i½u khi¼n bαn ≡ßnh  chµ  s¿
vτn ti¬p t°c ch╢ sau khi phφm + ≡▀■c bñm,  v∞ ng▀╢i d∙ng vτn c≥n c≤  th¼  ≡ßnh
m╡t dñu gi≈ng nh▀ lα m╡t phÑn t╪ cⁿa chu▓i k¬t h■p.

    Ph▀╜ng phßp t╒o h∞nh chºm cⁿa phΘp t╒o chµ ch∩ ≡╕nh ≡▀■c ≡½ ra ≡¼ bΣo  ≡Σm
vi«c tφch h■p v╛i nhµng ╤ng d°ng ≡≥i h÷i m▓i mπ t± phΣi liΩn k¬t v╛i m╡t  bi¬n
c» phφm chµ,  nhñt lα trong tr╒ng thßi Anh ngµ v∞ tr╒ng thßi nαy ch∩ cho  phΘp
t╒o chµ ch∩ ≡╕nh mα th⌠i.

    Múc d∙ c≤ th¼ t╒o h∞nh lºp t╤c trong phΘp t╒o chµ ch∩ ≡╕nh  hay  t╒o  h∞nh
chºm trong phΘp t╒o chµ ngÑm,  nh▀ng nhµng cßch th╤c nαy kh⌠ng hµu φch vα  ch∩
lαm cho ng▀╢i d∙ng lÑm lτn.   Do ≡≤,  vi«c ≡╜n giΣn nhñt lα ch∩ liΩn k¬t  cßch
t╒o h∞nh lºp t╤c v╛i phΘp t╒o chµ ngÑm, vα cßch t╒o h∞nh chºm v╛i phΘp t╒o chµ
ch∩ ≡╕nh.  Nhµng cßch th╤c nαy c≤ vδ t± nhiΩn h╜n.

    TiΩu chuªn trong vσn ki«n nαy quy ≡╕nh nhµng ≡úc tφnh  t»i  thi¼u  v½  mút
"h∞nh th╤c vα cΣm gißc" mα ng▀╢i d∙ng c≤ th¼ k╧ v≈ng ╖ m╡t nhu li«u  Vi«t  Ngµ
h■p th╤c.   M╡t giao di«n ≡▀■c tiΩu chuªn h≤a s¿ giΣm thi¼u th╢i gian phΣi h≈c
cho m▓i ╤ng d°ng m╛i.   TiΩu chuªn trong bαi nαy kh⌠ng lo╒i b÷ nhµng h«  th»ng
≡ßnh chµ khßc mα m°c ≡φch cⁿa ch·ng lα gi·p ng▀╢i d∙ng d¡ s╪ d°ng  h╜n,  ch╞ng
h╒n,  ≡ßnh dñu bóng bΣng-≡i½u-khi¼n kh⌠n ngoan,  hay gi·p ≡ßnh chµ nhanh  bóng
cßch d∙ng cßc phφm CONTROL hay  FUNCTION  ch╞ng  h╒n.   Bñt  c╤  s±  gia  c⌠ng
(enhancement) nαo trong nhµng ╤ng d°ng h■p th╤c (compliant application) ≡½u lα
                                                                        49
m╡t ≡i½u t»t cho ng▀╢i d∙ng,  mi¡n lα nhµng s± gia c⌠ng nαy kh⌠ng xung ≡╡t v╛i
nhµng ≡úc tφnh t»i thi¼u m⌠ tΣ trong bαi nαy.

5.3 HöP TH║C H╙A ║NG D₧NG VIÄT NG  HIÄN H└NH
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Bñt c╤ ph▀╜ng phßp th±c t¬ nαo ≡¼ ≡╕nh chuªn c√ng cÑn phΣi  d±  tr∙  nhµng  s±
ch»ng ≡▒i thay cⁿa cßc ╤ng d°ng hi«n hαnh. Trong khi mong mu»n róng tiΩu chuªn
8-bit trong bαi nαy ≡▀■c ⁿng h╡ hoαn toαn,  ch·ng t⌠i c√ng ≡½  ngh╕  m╡t  giΣi
phßp khßc d¡ dαng ≡▀■c chñp nhºn nhanh ch≤ng h╜n.   Tñt cΣ nhµng ╤ng d°ng  cÑn
phΣi cung cñp ph▀╜ng ti«n ≡¼ nhºn vαo vα xuñt ra nhµng dµ  ki«n  mπ  h≤a  theo
tiΩu chuªn VISCII 8-bit.  ╨░ng th╢i, nhµng ╤ng d°ng ≡≤ phΣi th±c hi«n m╡t giao
di«n ≡ßnh chµ tuΓn theo VIQR,  n¬u kh⌠ng phΣi lα ph▀╜ng phßp ≡ßnh chµ chⁿ  y¬u
th∞ φt nhñt c√ng lα m╡t ph▀╜ng phßp ph° thΩm cho ng▀╢i d∙ng.   Nhµng vi«c  nαy
rñt cÑn thi¬t cho cΣ ng▀╢i d∙ng lτn ng▀╢i bßn.   Ng▀╢i d∙ng c≤  th¼  d∙ng  nhu
li«u ngay v∞ giao di«n ≡ßnh chµ ≡░ng nhñt,  c√ng nh▀ c≤ th¼ x╪ l² dµ  ki«n  t╫
nhµng ╤ng d°ng khßc nhau vα trΩn nhµng h« th»ng mßy khßc nhau.  ╨i½u ≡≤ s¿ lαm
gia tσng nσng suñt vα s± trao ≡▒i giµa cßc ng▀╢i d∙ng.   Vi«c d¡  s╪  d°ng  s¿
khi¬n cho ╤ng d°ng ≡▀■c chñp nhºn r╡ng rπi h╜n, vα do ≡≤ ng▀╢i bßn s¿ c≤ nhi½u
khßch hαng h╜n.


6. T╙M TüT & KèT LUçN
~~~~~~~~~~~~~~~~~~~~~

Vσn ki«n nαy v╫a tr∞nh bαy m╡t d± thΣo tiΩu chuªn h≤a vi«c x╪ l² dµ ki«n  Vi«t
Ngµ.  Nhu cÑu tiΩu chuªn h≤a c√ng ≡π ≡▀■c lαm sßng t÷.  Ch·ng t⌠i mong róng ≡π
khuy¬n khφch gi╛i ch¬ t╒o nhu li«u vα ng▀╢i d∙ng nhu li«u Vi«t  ngµ  c╡ng  tßc
v╛i nhau ≡¼ ≡╒t m°c ≡φch nαy hÑu ≡em l■i φch ≡¬n cho tñt cΣ m≈i ng▀╢i liΩn h«.
Vi«c bαn luºn nhµng ph▀╜ng phßp mπ h≤a khßc nhau ≡π ≡▀a ≡¬n  s±  ch≈n  l±a  d±
thΣo VISCII 8-bit.  Ch·ng t⌠i ≡π ≡½ ngh╕ m╡t bΣng mπ t± duy nhñt, vα quß tr∞nh
th╪ nghi«m th±c ti¡n cho thñy bΣng nαy vºn hαnh t»t ≡⌐p cho Vi«t Ngµ  qua  cßc
c⌠ng vi«c nh▀ vi¬t bαi, x╪ l², l▀u trµ,  chuy¼n tin,  mπ h≤a ph⌠ng chµ,  vα ñn
loßt.   Trong nhµng lπnh v±c mα vi«c d∙ng 8-bit ch▀a cho phΘp hoúc kh⌠ng  ≡ßng
tin cºy, ch╞ng h╒n nh▀ vi«c chuy¼n ≡i«n th▀,  ch·ng t⌠i ≡π ≡½ ra quy ≡╕nh Vi«t
ngµ ≡≈c-≡▀■c-trong-ngoúc (VIQR) ≡¼ cung cñp m╡t c▒ng l≈c su⌠ng  sδ.   VIQR  ≡π
≡▀■c quy ≡╕nh ≡╡c lºp v╛i ngu░n xuñt phßt dµ ki«n,  do ≡≤ ≡π ≡▀■c thi¬t k¬  ≡¼
c≤ th¼ ßp d°ng ≡▀■c cho cΣ bαn ≡ßnh chµ ti¬ng Vi«t lτn cßc mßy  l≈c  dµ  ki«n.
Tñt cΣ nhµng ≡i½u nαy ≡π ≡▀■c ch╤ng t÷ lα c≤ th¼ tφch h■p vαo nhµng m⌠i tr▀╢ng
hi«n hµu,  gi·p cho vi«c s╪ d°ng nhµng h▀ c° vα ╤ng d°ng  hi«n  hµu  ≡▀■c  tr╖
thαnh d¡ dαng h╜n --- m╡t ▀u ≡i¼m l╛n cⁿa ph▀╜ng phßp mπ h≤a nαy.   Cu»i c∙ng,
nhµng quy ≡╕nh nαy ≡π ≡▀■c liΩn k¬t v╛i nhau m╡t cßch su⌠ng sδ trong m≈i  giai
≡o╒n cⁿa chu k╧ x╪ l² dµ ki«n (g░m c≤ nhºn dµ ki«n,  x╪ l²/truy½n dµ ki«n,  vα
xuñt ra dµ ki«n,  g≈i tít lα chu k╧ nhºp-bi¬n-xuñt).   Nhµng quy ≡╕nh  nαy  ≡π
cung cñp m╡t khu⌠n kh▒ th»ng nhñt th±c s± cho vi«c x╪ l² dµ ki«n Vi«t Ngµ.


T└I LIÄU THAM KH─O
~~~~~~~~~~~~~~~~~~

[1]  B╒ch H▀ng Khang.  "Institute of Informatics."  Hα N╡i,
     Vi«t Nam, thßng hai 1991.

                                                                        50
[2]  B. Jerman-Blazic, "Will the Multy-octet Standard Character Set
     Code Solve the World Coding Problems for Information Interchange?"
     Computer Standards & Interfaces, vol. 8, trang 127-136, 1988.

[3]  The Unicode Consortium.  The Unicode Standard: Worldwide
     Character Encoding Version 1.0.  Addison-Wesley, Reading, MA, ñn
     bΣn th╤ nhñt, thßng m▀╢i 1991.

[4]  ISO Technical Committee, "Universal Multiple-Octet Coded
     Character Set (UCS), ISO/IEC DIS 10646-1.2," Draft standard,
     International Organization for Standardization, 1992.

[5]  International Organization for Standardization, ISO 8859/x:
     8-bit Internatonal Code Sets.  ISO, 1977.

[6]  Famjxuaen Thais.  Vi«t Ngµ CΣi Cßch.  T╤ HΣi, Hα N╡i,
     Vi«t Nam, thßng ba 1948.

[7]  Ph╒m XuΓn Thßi.  Chµ Vi«t H■p Lφ, Tφn-╨╤c Th▀ Xπ,
     Sαi G≥n, Vi«t Nam, thßng t▀ 1958.

[8]  J. Postel, "Simple Mail Transfer Protocol," RFC 822, USC
     Information Sciences Institute, thßng tßm 1982.

[9]  J. C. Klensin et al., "SMTP Extensions for Transport of
     Text-Based Messages Containing 8-bit Characters," Internet draft,
     Massachusetts Institute of Technology, thßng bΣy 1991.

[10] K. Simonsen, "Character Mnemonics & Character Sets," Internet
     Draft, Danish Unix Users Group, thßng giΩng 1992.

[11] K. Simonsen, "Mnemonic Text Format," Internet draft, Danish
     Unix Users Group, thßng tßm 1991.

[12] International Organization for Standardization.  ISO 646:
     7-bit Coded Character Set for Information Interchange.  ISO,
     ñn bΣn th╤ ba, 1991.

[13] International Organization for Standardization.  ISO 2022:
     7-bit and 8-bit Coded Character Sets--Code Extension Techniques.
     ISO, ñn bΣn th╤ ba, 1986.

[14] E. M. van der Poel, "Multilingual Character Encoding for
     Internet Messages," Internet draft, Software Research Associates,
     Japan, thßng giΩng 1992.

[15] D.E. Knuth.  The TeXbook.  Addison-Wesley, Reading, MA, 1984.


                                                                        51
THUçT-NG  ANH VIÄT
~~~~~~~~~~~~~~~~~~

Announcer: mπ t± (hay chu▓i mπ t±) bßo tin.   Khi mπ t± nαy  xuñt  hi«n  trong
    d≥ng dµ ki«n th∞ n≤ bßo cho bi¬t nhµng mπ t± ≡i sau c≤  m╡t  ²  nghεa  ≡úc
    bi«t.   Trong vσn ki«n nαy,  n≤ cho bi¬t s± m╖ ≡Ñu cⁿa vi«c  k¬t  h■p  chµ
    Vi«t.

ASCII: American Standard Code for Information Interchange, b╡ mπ t± tiΩu-chuªn
    Hoa-k╧ dαnh cho vi«c trao ≡▒i tin-t╤c.   B╡ mπ nαy c≤ 128 mπ s»  ≡▀■c  hÑu
    h¬t cßc mßy vi-tφnh d∙ng ≡¼ ≡úc tr▀ng vα truy½n ≡i cßc dµ ki«n  chµ.   M▓i
    chµ trong b╡ mπ nαy c≤ mπ s» trong khoΣng t╫ 0 ≡¬n 127.  Nhµng b╡ mπ 8-bit
    hoúc 9-bit trong ≡≤ 128 mπ t± ≡Ñu tiΩn t▀╜ng ╤ng v╛i ASCII ≡▀■c g≈i lα  b╡
    mπ ASCII-r╡ng (extended ASCII).  Nhµng mπ t± thΩm vαo lα mτu t± La-tinh c≤
    dñu r╢i, mτu t± phi-La-tinh, ki¼m t± ≡i½u khi¼n mαn Σnh, vΓn vΓn.

Backslash: g╒ch-chΘo-ng▀■c (\).

Base Vowel: nguyΩn Γm c╜ bΣn.  ╨╤ng trΩn ph▀╜ng di«n mπ h≤a chµ Vi«t, vσn ki«n
    nαy coi nhµng nguyΩn Γm sau ≡Γy lα c╜ bΣn: a, σ, Γ, e, Ω, i, o, ⌠,  ╜,  u,
    ▀, y vα nhµng chµ hoa t▀╜ng ╤ng.

Binary Data: dµ ki«n nh╕ phΓn; t∙y theo ngµ cΣnh c≥n mang nghεa dµ ki«n 8-bit,
    nhñt lα trong lπnh v±c chuy¼n tin.

C0 Space: V∙ng (mi½n) C0.   ╨Γy lα tºp h■p g░m nhµng mπ t± c≤ s» thºp-l°c phΓn
    t╫ 00 ≡¬n 1F (v∙ng "ki¼m t±" cⁿa b╡ mπ ASCII).

C1 Space: V∙ng (mi½n) C1.   ╨Γy lα tºp h■p g░m nhµng mπ t± c≤ s» thºp-l°c phΓn
    t╫ 80 ≡¬n 8F (v∙ng "ki¼m t±" cⁿa b╡ mπ ASCII-r╡ng).

Character: mτu t±, k² t±, mπ t±. Trong tin-h≈c,  character th▀╢ng ≡▀■c d∙ng ≡¼
    ch∩ bñt c╤ cßi g∞ ≡▀■c liΩn k¬t v╛i m╡t mπ s» (code) nΩn nghεa  ≡·ng  nhñt
    lα mπ t±.   Mπ t± c≤ th¼ lα mτu t± (nh▀ a, b, c, ...),  hoúc dñu hi«u,  k²
    hi«u (nh▀ +, -, =, ...), hoúc m╡t tφn hi«u ≡i½u khi¼n.   K² t± ch∩ k² hi«u
    theo nghεa r╡ng, bao g░m cßc dñu hi«u, mτu t±, hoúc chµ t▀■ng h∞nh nh▀ chµ
    Hßn.

Character Set: tºp mπ t±, b╡ mπ t±.   C≤ th¼ d╕ch thoßt lα h« mπ t± v∞ m▓i tºp
    mπ t± lα m╡t h« th»ng k² t± cho m╡t hoúc nhi½u ng⌠n ngµ.  C√ng c≤ th¼ d╕ch
    thoßt lα bΣng mπ t± v∞ th▀╢ng th▀╢ng tºp mπ t± ≡▀■c tr∞nh  bαy  d▀╛i  d╒ng
    bΣng.   S» l▀■ng mπ t± trong m▓i tºp lα 2^n (2 l√y th╫a n) v╛i n lα s» bit
    d∙ng ≡¼ mπ h≤a m╡t mπ t±.   Cßc h« mπ t± quen thu╡c lα b╡ ASCII 7-bit  cⁿa
    Hoa-k╧, h« 8-bit nh▀ cßc tºp mπ t± ISO-8859/X,  h« 16-bit nh▀ Unicode,  h«
    32-bit nh▀ d± thΣo ISO DIS 10646.

Code: mπ s» (trong th⌠ng tin dµ ki«n),  con s» t▀■ng tr▀ng cho m╡t mτu t±,  k²
    t±, hoúc tφn hi«u ≡i½u khi¼n.  Thφ d° s» thºp phΓn 65 trong b╡ mπ t± ASCII
    Hoa k╧ t▀■ng tr▀ng cho chµ A.

Code Page: thuºt ngµ th▀╢ng d∙ng ≡¼ ch∩ nhµng tºp mπ t± d∙ng trΩn mßy IBM  PC,
    vi¬t tít lα CP.  CP850 lα tºp mπ t± ≡a ngµ, CP860 lα tºp mπ t± B░-≡αo-nha,
    CP863 cho ti¬ng Phßp ╖ Gia-nπ-≡╒i, CP865 cho Na-uy.
                                                                        52
Code Page Switching: ≡▒i bΣng (tºp, b╡, h«) mπ t±.

Compatible: t▀╜ng h■p, t▀╜ng dung

Compliant: h■p th╤c, tuΓn theo ≡·ng cßch.

Composed Character: chµ ghΘp,  chµ  r╢i  (theo  quan  ≡i¼m  mπ  h≤a). Xem  chµ
    floating diacritic.

Context-Dependent: t∙y thu╡c vαo ngµ cΣnh (² nghεa chung quanh).

Control Character: mπ t± ≡i½u khi¼n,  ki¼m t±.   ╨≤ lα mπ t± ASCII  nóm  trong
    khoΣng t╫ 0 ≡¬n 31, vα mπ t± 127,  t▀╜ng phΣn v╛i nhµng mπ t± c≤ th¼ in ra
    ≡▀■c (g≈i lα k² t± h∞nh) nóm trong khoΣng t╫ 32 ≡¬n 126.  TrΩn cßc bαn chµ
    ASCII,  ki¼m t± (thφ d° CTRL-A,  mπ s» 1) ≡▀■c t╒o ra bóng cßch chºn  phφm
    CTRL xu»ng r░i ≡ßnh chµ liΩn h« (A).

Cross-Platform: xuyΩn-giαn, xuyΩn qua nhi½u h« th»ng mßy khßc nhau.

Data: dµ ki«n, dµ li«u.

Data Channel: m╒ch dµ ki«n

Data Communication: th⌠ng tin dµ ki«n, lπnh v±c chuy¼n tin.

Data Frame: khung dµ ki«n.

Data Integrity: s± toαn v⌐n dµ ki«n, s± bΣo toαn dµ ki«n.

Data Stream: lu░ng (d≥ng) dµ ki«n, d≥ng tin.

Diacritic: dñu ph°.   Dñu ph° lα nhµng nΘt thΩm vαo m╡t mτu t± "g»c" ≡¼ t╒o ra
    mτu t± khßc.  Ch╞ng h╒n mτu t± ┬ ≡▀■c cñu t╒o t╫ mτu t± g»c A vα dñu ph° ^
    (dñu m√).

Display: Hi¼n th╕, t╒o h∞nh, in h∞nh (trΩn mαn Σnh).  Xem chµ rendering.

EBCDIC: Extended Binary Coded Decimal Interchange Code, b╡ mπ 8-bit g░m 256 mπ
    t± d∙ng trΩn cßc mßy IBM mainframes.

Editor: ╤ng d°ng vi¬t bαi, vi¬t-c° (d°ng c° vi¬t bαi).

Electronic Mail: ≡i«n th▀.

to Encode: mπ h≤a.

Escape Mechanism: c╜ ch¬ thoßt.

Fax: ≡i«n h∞nh th▀.   Khßc v╛i ≡i«n th▀ v∞ ch∩  chuy¼n  ≡i  nhµng  chñm  h∞nh.
    Ng▀╢i nhºn kh⌠ng th¼ d∙ng nhµng ╤ng d°ng vi¬t bαi ≡¼ s╪a ≡▒i ≡▀■c.

File: h░ s╜.  C≤ tr▀╢ng phßi d╕ch lα t«p.

                                                                        53
Floating Diacritic: dñu r╢i (theo quan ≡i¼m mπ h≤a).  M╡t mτu t± c≤ dñu ph° c≤
    th¼ ≡▀■c mπ h≤a bóng m╡t mπ s» duy nhñt hoúc nhi½u mπ s».  Thφ d° ╘ c≤ th¼
    ≡▀■c mπ h≤a bóng m╡t mπ s» duy nhñt vα ≡▀■c g≈i lα chµ nguyΩn v⌐n hoúc chµ
    d±ng s╟n (precomposed character),  hoúc mπ h≤a bóng hai  mπ  s»,  m╡t  cho
    nguyΩn Γm g»c O vα m╡t cho dñu m√ (^).   Trong tr▀╢ng h■p sau,  ╘ ≡▀■c g≈i
    lα chµ ghΘp (composed character) vα dñu m√ ≡▀■c g≈i lα dñu r╢i.

Font: "ph⌠ng," b╡ ki¼u chµ,  m╡t tºp h■p cßc h∞nh chµ c≤ chung m╡t s» ≡úc tφnh
    nαo ≡≤ vα c≤ th¼ in ra ≡▀■c trΩn mαn Σnh hoúc trΩn giñy.   Italic font: b╡
    chµ nghiΩng.   Bold face font: b╡ chµ in ≡ºm.   M▓i h∞nh chµ (glyph) trong
    ph⌠ng ≡▀■c ñn ≡╕nh m╡t mπ s».   Ph⌠ng 8-bit c≤ th¼ c≤ tñt cΣ 256 h∞nh chµ.
    Mπ s» cⁿa h∞nh chµ (glyph code) kh⌠ng nhñt thi¬t phΣi gi»ng v╛i mπ s»  cⁿa
    chµ (character code) t▀╜ng ╤ng trong tºp mπ t±.   Thφ d°, dµ ki«n chµ A c≤
    mπ s» 65 trong bΣng ASCII nh▀ng h∞nh chµ A c≤ th¼ ≡▀■c quy ≡╕nh ╖  v╕  trφ
    th╤ 35 trong m╡t bΣng ph⌠ng nαo ≡≤,  n¬u mu»n.   Nh▀ng m╡t h∞nh chµ  trong
    ph⌠ng l╒i c≤ th¼ t▀╜ng ╤ng v╛i nhi½u chµ trong tºp mπ t±.  ╨i½u nαy th▀╢ng
    xΣy ra trong ph▀╜ng phßp mπ h≤a d∙ng dñu  r╢i  (xem  floating  diacritic).
    Trong ph▀╜ng phßp nαy,  chµ └ th±c ra lα s± k¬t h■p cⁿa dµ ki«n chµ A  (mπ
    s» 65) vα dµ ki«n dñu huy½n ` (thφ d° mπ s» 196),  nh▀ng khi t╒o h∞nh  th∞
    d∙ng h∞nh d╒ng cⁿa chµ └,  thφ d° nóm ╖ v╕ trφ 135  cⁿa  bΣng  ph⌠ng  chµ.
    ╨i½u nαy lαm cho vi«c x╪ l² ph╤c t╒p vα kΘm hi«u nσng nΩn hÑu h¬t cßc  nhu
    li«u vα c▀╜ng li«u ┬u M█ trßnh kh⌠ng d∙ng.   Do ≡≤, ≡¼ c≤ th¼ tφch h■p vαo
    cßc m⌠i tr▀╢ng nhu li«u hi«n hµu,  mπ s» h∞nh chµ vα mπ s» chµ phΣi  gi»ng
    nhau.

Font Shifting: chuy¼n ph⌠ng, chuy¼n mτu chµ.

Format: khu⌠n th╤c (khu⌠n kh▒ vα h∞nh th╤c).

Format Data Control Character: mπ t± ≡i½u khi¼n khu⌠n th╤c.

Frame: khung, s▀╢n.

Framework: khu⌠n kh▒.

G0 Space: v∙ng k² t± h∞nh (graphic character) c≤ mπ s» thºp-l°c phΓn t╫ 20 ≡¬n
    7F.

G1 Space: v∙ng k² t± h∞nh c≤ mπ s» thºp-l°c phΓn t╫ A0 ≡¬n FF.

Gateway: c▒ng, ≡Ñu cÑu.

Glyph: h∞nh chµ, m╡t phÑn t╪ cⁿa ph⌠ng chµ (font).

Graphic Character: k² t± h∞nh (k² t± c≤ th¼ in ra ≡▀■c).

Interface: giao-di«n, m╒ch n»i.

Interface Between 2 Computers: giao di«n (m╒ch n»i) giµa 2 mßy.

ISO: International Organization for Standardization.   M╡t ⁿy ban qu»c  t¬  t±
    nguy«n g░m cßc ⁿy ban ≡╕nh chuªn cßc n▀╛c h■p tßc v╛i nhau ≡¼ ≡╕nh ra  cßc
    tiΩu chuªn trong tñt cΣ m≈i lπnh v±c, trong ≡≤ bao g░m mßy vi tφnh,  x╪ l²
                                                                        54
    tin t╤c, tºp mπ t±.

ISO 646: TiΩu chuªn mπ t± 7-bit, t▀╜ng t± v╛i ASCII.

ISO Standard 8859: TiΩu chuªn ISO 8859.   TiΩu chuªn nαy quy ≡╕nh m╡t lo╒t cßc
    tºp mπ t± 8-bit bao g░m chµ cⁿa nhi½u ng⌠n ngµ.   Lo╒t nαy bao  g░m  nhµng
    tºp mτu t± La-tinh 1--9, ßp d°ng cho tñt cΣ nhµng ng⌠n ngµ c≤ chµ vi¬t d±a
    trΩn mτu t± La-mπ,  c╡ng thΩm m╡t  s»  nhµng  tºp  mτu  t±  ≡úc  bi«t  nh▀
    Cyrillic, Hy-l╒p, vα Do-thßi.

ISO 8859/1: TiΩu chuªn ISO 8859 Mτu t± La-tinh s» 1.  T»i thi¼u nhµng ng⌠n ngµ
    sau ≡Γy s╪ d°ng ≡▀■c: La-tinh, ╨an-m╒ch, ╨╤c,  Dutch,  Anh-ngµ,  Faeroese,
    PhÑn-lan, Phßp, Bσng-≡Σo, ┴i-nhε-lan, ▌, Na-uy,  B░-≡αo-nha,  TΓy-ban-nha,
    vα Th°y-≡i¼n.

ISO 2022 and ISO 4873: TiΩu chuªn ISO d∙ng ≡¼ chuy¼n h« mπ t±.

ISO DIS 10646: ISO Draft International Standard,  d± thΣo tiΩu chuªn  qu»c  t¬
    ISO 16-bit vα 32-bit cho tñt cΣ cßc ng⌠n ngµ trΩn th¬ gi╛i.

Latin: ch∩ b╡ mτu t± La-tinh hoúc La-mπ, g░m c≤ cßc chµ t╫ A ≡¬n Z, hay tñt cΣ
    cßc b╡ mτu t± d±a vαo La-tinh.

Keyboard Interface: giao di«n bαn ≡ßnh chµ.

Integrated Environment: m⌠i tr▀╢ng tφch h■p.

Integration: s± tφch h■p, sßt nhºp.

Keyboard: bαn ≡ßnh chµ.

Line-Drawing: v¿ ≡▀╢ng th╞ng.

Literal: nguyΩn d╒ng, theo sßt nghεa ≡en.

Literal character: nguyΩn t±, chµ vi¬t sao hi¼u vºy.

Lower-Case Character: chµ in th▀╢ng.

Look-and-Feel: h∞nh th╤c vα cΣm gißc.

Mail Agent: b╡ chuy¼n th▀, ≡╒i l² th▀ t╫.

Mnemonic: d¡ nh╛.

Modifier: Dñu ph° ≡¼ thay ≡▒i Γm.   Trong ti¬ng Vi«t, ≡≤ lα cßc dñu trσng (dñu
    ß), dñu m√ (dñu ^), vα dñu m≤c (trong chµ ┤ hoúc ┐).

On the Fly: theo l·c ≡≤, ngay l·c ≡≤.

PC: Personal Computer,  mßy vi tφnh cß nhΓn.   Trong vσn ki«n nαy,  chµ PC ch∩
    toαn b╡ cßc mßy IBM PC vα PS/2 c∙ng nhµng mßy t▀╜ng h■p,  k¼  cΣ  mßy  AT,
    286, 386 vα 486.
                                                                        55
PostScript: M╡t ng⌠n ngµ m⌠-tΣ t╫ng trang (giñy, sßch, bßo, v.v.) m╡t,  c≤ khΣ
    nσng x╪ l² ≡░-h∞nh (graphic capabilities),  d∙ng trong vi«c ñn  loßt  bóng
    mßy vi-tφnh.   ╨Γy lα m╡t ng⌠n ngµ thΣo ch▀╜ng cñp cao vα ≡╡c lºp v╛i  m≈i
    thi¬t b╕.   PostScript lα nhπn hi«u cÑu ch╤ng cⁿa c⌠ng  ty  Adobe  Systems
    Incorporated.

Precomposed Character: chµ nguyΩn v⌐n,  chµ d±ng s╟n (theo quan ≡i¼m mπ  h≤a).
    Xem chµ floating diacritic.

Processing: x╪ l².

Protocol: biΩn bΣn, nghi th╤c

Public-Domain: thu╡c lπnh v±c c⌠ng c╡ng.

Quoted-Readable: ≡≈c ≡▀■c trong ngoúc.   T╫ x▀a,  ² nghεa cⁿa nhµng mπ t±  ≡úc
    bi«t c≤ th¼ thay ≡▒i bóng cßch b÷ ch·ng trong ngoúc ≡╜n  hoúc  ngoúc  kΘp,
    hoúc d∙ng m╡t k² t± bßo tin ≡i tr▀╛c nh▀ dñu  g╒ch  chΘo  ng▀■c  (\)  d∙ng
    trong khi¼n h« Unix hay ng⌠n ngµ thΣo  ch▀╜ng  C.   Do  ≡≤  xuñt  hi«n  t╫
    "trong-ngoúc" (quoted).  C╜ ch¬ ngoúc ≡▀■c g≈i lα quoting mechanism.   Thφ
    d° trong khi¼n h« Unix,  k² t± * ≡▀■c khai tri¼n thαnh nhµng k²  t±  khßc,
    nΩn m«nh l«nh
                          rm *
    s¿ x≤a h¬t m≈i h░ s╜, trong khi ≡≤ m«nh l«nh
                          rm "*"
    ch∩ x≤a m╡t h░ s╜ c≤ tΩn lα k² t± *.   T▀╜ng t±, n¬u c╜ ch¬ ngoúc ≡▀■c quy
    ▀╛c lα [cd] ≡¼ t▀■ng tr▀ng cho m╡t nguyΩn Γm c≤ dñu,  trong ≡≤ c lα nguyΩn
    Γm, d lα dñu ph°, th∞ ta c≤ th¼ di¡n tΣ chµ cΣ bóng c[a?],  v∞ chu▓i a? c≤
    nghεa lα Σ khi nóm trong ngoúc.   Quy ≡╕nh VIQR trong vσn  ki«n  nαy  d∙ng
    nhµng k² t± g■i h∞nh nh▀ ' ` ? ~ . ≡¼ t▀■ng tr▀ng cho dñu gi≈ng nΩn d¡ ≡≈c
    vα d¡ nh╛.

Real Time: th±c th╢i.  Trong lπnh v±c ≡i«n toßn, ch∩ vi«c x╪ l² xΣy ra ≡░ng b╡
    v╛i bi¬n c» th±c s±.

Rendering: s± hi¼n th╕, s± t╒o h∞nh, in h∞nh (lΩn mαn Σnh hoúc giñy).  Xem chµ
    display.

Sequence: chu▓i, lo╒t.

Software: nhu li«u.  C≤ tr▀╢ng phßi d╕ch thoßt lα h▀ li«u (h▀ lα h▀ th¼, kh⌠ng
    c≤ th±c).

Software Application: nhu li«u ╤ng d°ng, hoúc d╕ch tít lα ╤ng d°ng.

Software Library: th▀ vi«n nhu li«u.  C≤ th¼ d╕ch thoßt lα h▀-vi«n.

Software Tool: nhu li«u d°ng c°, h▀ c°.

Source Code: ch▀╜ng tr∞nh g»c (bΣn  g»c  cⁿa  ch▀╜ng  tr∞nh  ≡¼  cho  b╡  d╕ch
    (compiler) ≡≈c vα d╕ch ra ng⌠n ngµ mßy).

Specification: quy ≡╕nh.
                                                                        56
Table-Lookup: tra bΣng.

TeX: M╡t h« th»ng ñn loßt ≡▀■c ≡i«n toßn h≤a do Donald Knuth [15] phßt  tri¼n,
    c≤ khΣ nσng ≡ßp ╤ng m≈i nhu cÑu ñn loßt cßc k² hi«u toßn h≈c  vα  cßc  vσn
    bΣn v╛i phªm chñt cao.   TeX lα nhπn hi«u cÑu ch╤ng cⁿa H╡i Toßn-h≈c  Hoa-
    k╧.

Text: vσn bΣn.

Text Viewer: ╤ng d°ng nh∞n chµ.   Th▀╢ng d∙ng ≡¼ duy«t qua trΩn mαn Σnh mßy vi
    tφnh xem vσn bΣn hi«n ra nh▀ th¬ nαo tr▀╛c khi in ra giñy.

Trade-off: vi«c ch≈n l■i h╒i.

Transmission: chuy¼n, truy½n.

Transparent: th⌠ng su»t, v⌠ h∞nh, kh⌠ng dñu v¬t.  ù ≡Γy d∙ng ≡¼ ch∩ vi«c x╪ l²
    ╖ nhµng cñp bºc d▀╛i c≤ th¼ khßc nhau nh▀ng vτn lαm cho giao  di«n  ╖  cñp
    bºc cao h╜n kh⌠ng thñy c≤ g∞ thay ≡▒i hoúc phΓn bi«t ≡▀■c.

Unicode: M╡t b╡ mπ t± 16-bit do liΩn-≡oαn c⌠ng  ty  Unicode  Consortium  thi¬t
    lºp.  än bΣn (version) 1.0 [3] ≡▀■c phßt hαnh nσm 1991 bao g░m hÑu h¬t cßc
    ng⌠n ngµ trΩn th¬ gi╛i,  trong ≡≤ c≤ cßc chµ  La-tinh,  TΓy-┬u,  ╨⌠ng  ┬u,
    Vi«t-nam, Hy-l╒p, Nga,  Do-Thßi,  ─-rºp,  chµ t▀■ng-h∞nh Hßn Nhºt ╨╒i-hαn,
    v.v.  Trong ñn-bΣn 1.0, chµ Vi«t-nam ≡▀■c mπ h≤a theo ph▀╜ng phßp d∙ng dñu
    r╢i (floating diacritic).   Tin bßn chφnh th╤c cho  bi¬t  khi  tiΩu  chuªn
    Unicode sßt nhºp v╛i ISO DIS 10646,  chµ Vi«t-nam  s¿  ≡▀■c  mπ  h≤a  theo
    ph▀╜ng phßp chµ nguyΩn.

Unix: M╡t h« th»ng ≡i½u khi¼n (operating system) mßy vi tφnh  rñt  ph▒  th⌠ng,
    ≡▀■c phßt tri¼n ╖ trung tΓm nghiΩn c╤u AT&T Bell  Laboratories.   Rñt  n▒i
    ti¬ng v∞ c≤ th¼ vºn hαnh trΩn nhi½u h« th»ng mßy khßc nhau.

Upper-Case Character: chµ hoa.

Usenet: m╡t m╒ng l▀╛i th⌠ng tin ≡i«n toßn qu»c t¬ cho phΘp ng▀╢i d∙ng mßy  vi-
    tφnh g╪i tin t╛i nhµng ng▀╢i khßc vα nhµng ng▀╢i  nαy  c≤  th¼  ph·c  ≡ßp.
    Vi«c tham gia vαo m╒ng l▀╛i Usenet c√ng gi»ng nh▀ vi«c ≡σng k²  xem  nhµng
    t╒p chφ ≡i«n toßn (electronic magazine,  t╒p chφ hi«n di«n trong m╒ng l▀╛i
    ≡i«n toßn, kh⌠ng phΣi lα t╒p chφ in trΩn giñy).  Nhµng t╒p chφ nαy, g≈i lα
    nh≤m-tin (newsgroup),  chuyΩn v½ rñt nhi½u chⁿ  ≡½  khßc  nhau.   Nh≤m-tin
    "Soc.Culture.Vietnamese" rñt ph▒ th⌠ng trong gi╛i ng▀╢i Vi«t vα ngo╒i qu»c
    ╖ khíp n╜i trΩn th¬ gi╛i.

User: ng▀╢i d∙ng, ng▀╢i s╪ d°ng.

User-Interface: giao-di«n cho ng▀╢i d∙ng.

Utility Software: nhu li«u d°ng c°,  d°ng li«u (nhu li«u hµu d°ng). Nhµng  ╤ng
    d°ng ≡¼ tr■ gi·p vi«c phßt tri¼n nhu li«u nh▀ nhu li«u vi¬t bαi  (editor),
    nhu li«u t∞m chµ (grep,  awk trong khi¼n h« Unix),  nhu li«u ≡¼ ≡i½u khi¼n
    in nhi½u bαi.

                                                                        57
Vietnamese Character Code: b╡ mπ chµ Vi«t, b╡ Vi«t-t±-mπ.

Vietnamese Character Encoding Standard: b╡ Vi«t-t±-mπ tiΩu chuªn.

Viet-Std: M╡t nh≤m v⌠ v° l■i g░m cßc chuyΩn viΩn Vi«t-nam ╖ hΣi ngo╒i h■p  tßc
    ≡¼ ≡╕nh ra cßc tiΩu chuªn v½ c▀╜ng-li«u vα nhu li«u  cho  chµ  Vi«t.   Cßc
    thαnh viΩn trao ≡▒i ² ki¬n qua ≡i«n th▀ vα cßc cu╡c h≈p.

Variant: d╕-bΣn

Word Processing: x╪ l² chµ.

Word Processor: mßy x╪ l² chµ.

Word Processing Software: nhu li«u x╪ l² chµ.

                                                                        58
           Ph° L°c A: Mτu T± Vi«t Li«t KΩ Theo Th╤ T± Síp Chµ
+============================================================================+
|Chµ: VIQR:VISCII || Chµ: VIQR:VISCII || Chµ: VIQR:VISCII || Chµ: VIQR:VISCII|
+---:-----:-------++----:-----:-------++----:-----:-------++----:-----:------+
| A : A   :  065  ||  N : N   :  078  ||  a : a   :  097  ||  n : n   :  110 |
| ┴ : A'  :  193  ||  O : O   :  079  ||  ß : a'  :  225  ||  o : o   :  111 |
| └ : A`  :  192  ||  ╙ : O'  :  211  ||  α : a`  :  224  ||  ≤ : o'  :  243 |
| ─ : A?  :  196  ||  ╥ : O`  :  210  ||  Σ : a?  :  228  ||  ≥ : o`  :  242 |
| ├ : A~  :  195  ||  Ö : O?  :  153  ||  π : a~  :  227  ||  ÷ : o?  :  246 |
| Ç : A.  :  128  ||  ╒ : O~  :  213  ||  á : a.  :  160  ||  ⌡ : o~  :  245 |
| ┼ : A(  :  197  ||  Ü : O.  :  154  ||  σ : a(  :  229  ||  ≈ : o.  :  247 |
| ü : A(' :  129  ||  ╘ : O^  :  212  ||  í : a(' :  161  ||  ⌠ : o^  :  244 |
| é : A(` :  130  ||  Å : O^' :  143  ||  ó : a(` :  162  ||  » : o^' :  175 |
|  : A(? :  002  ||  É : O^` :  144  ||  ╞ : a(? :  198  ||  ░ : o^` :  176 |
|  : A(~ :  005  ||  æ : O^? :  145  ||  ╟ : a(~ :  199  ||  ▒ : o^? :  177 |
| â : A(. :  131  ||  Æ : O^~ :  146  ||  ú : a(. :  163  ||  ▓ : o^~ :  178 |
| ┬ : A^  :  194  ||  ô : O^. :  147  ||  Γ : a^  :  226  ||  ╡ : o^. :  181 |
| ä : A^' :  132  ||  ┤ : O+  :  180  ||  ñ : a^' :  164  ||  ╜ : o+  :  189 |
| à : A^` :  133  ||  ò : O+' :  149  ||  Ñ : a^` :  165  ||  ╛ : o+' :  190 |
| å : A^? :  134  ||  û : O+` :  150  ||  ª : a^? :  166  ||  ╢ : o+` :  182 |
|  : A^~ :  006  ||  ù : O+? :  151  ||  τ : a^~ :  231  ||  ╖ : o+? :  183 |
| ç : A^. :  135  ||  │ : O+~ :  179  ||  º : a^. :  167  ||  ▐ : o+~ :  222 |
| B : B   :  066  ||  ö : O+. :  148  ||  b : b   :  098  ||  ■ : o+. :  254 |
| C : C   :  067  ||  P : P   :  080  ||  c : c   :  099  ||  p : p   :  112 |
| D : D   :  068  ||  Q : Q   :  081  ||  d : d   :  100  ||  q : q   :  113 |
| ╨ : DD  :  208* ||  R : R   :  082  ||  ≡ : dd  :  240  ||  r : r   :  114 |
| E : E   :  069  ||  S : S   :  083  ||  e : e   :  101  ||  s : s   :  115 |
| ╔ : E'  :  201  ||  T : T   :  084  ||  Θ : e'  :  233  ||  t : t   :  116 |
| ╚ : E`  :  200  ||  U : U   :  085  ||  Φ : e`  :  232  ||  u : u   :  117 |
| ╦ : E?  :  203  ||  ┌ : U'  :  218  ||  δ : e?  :  235  ||  · : u'  :  250 |
| ê : E~  :  136  ||  ┘ : U`  :  217  ||  ¿ : e~  :  168  ||  ∙ : u`  :  249 |
| ë : E.  :  137  ||  £ : U?  :  156  ||  ⌐ : e.  :  169  ||  ⁿ : u?  :  252 |
| ╩ : E^  :  202  ||  ¥ : U~  :  157  ||  Ω : e^  :  234  ||  √ : u~  :  251 |
| è : E^' :  138  ||  ₧ : U.  :  158  ||  ¬ : e^' :  170  ||  ° : u.  :  248 |
| ï : E^` :  139  ||  ┐ : U+  :  191  ||  ½ : e^` :  171  ||  ▀ : u+  :  223 |
| î : E^? :  140  ||  ║ : U+' :  186  ||  ¼ : e^? :  172  ||  ╤ : u+' :  209 |
| ì : E^~ :  141  ||  ╗ : U+` :  187  ||  ¡ : e^~ :  173  ||  ╫ : u+` :  215 |
| Ä : E^. :  142  ||  ╝ : U+? :  188  ||  « : e^. :  174  ||  ╪ : u+? :  216 |
| F : F   :  070  ||    : U+~ :  255  ||  f : f   :  102  ||  µ : u+~ :  230 |
| G : G   :  071  ||  ╣ : U+. :  185  ||  g : g   :  103  ||  ± : u+. :  241 |
| H : H   :  072  ||  V : V   :  086  ||  h : h   :  104  ||  v : v   :  118 |
| I : I   :  073  ||  W : W   :  087  ||  i : i   :  105  ||  w : w   :  119 |
| ═ : I'  :  205  ||  X : X   :  088  ||  φ : i'  :  237  ||  x : x   :  120 |
| ╠ : I`  :  204  ||  Y : Y   :  089  ||  ∞ : i`  :  236  ||  y : y   :  121 |
| ¢ : I?  :  155  ||  ▌ : Y'  :  221  ||  ∩ : i?  :  239  ||  ² : y'  :  253 |
| ╬ : I~  :  206  ||  ƒ : Y`  :  159  ||  ε : i~  :  238  ||  ╧ : y`  :  207 |
| ÿ : I.  :  152  ||   : Y?  :  020  ||  ╕ : i.  :  184  ||  ╓ : y?  :  214 |
| J : J   :  074  ||   : Y~  :  025  ||  j : j   :  106  ||  █ : y~  :  219 |
| K : K   :  075  ||   : Y.  :  030  ||  k : k   :  107  ||  ▄ : y.  :  220 |
| L : L   :  076  ||  Z : Z   :  090  ||  l : l   :  108  ||  z : z   :  122 |
| M : M   :  077  ||    :     :       ||  m : m   :  109  ||    :     :      |
+============================================================================+

        * Quy ≡╕nh VIQR c≥n cho phΘp t▀■ng tr▀ng "╨" bóng "Dd" hoúc "dD".
                                                                        59
     Ph° L°c B: Mτu T± Vi«t Li«t KΩ Theo Th╤ T± Mπ S» VISCII Thºp PhΓn
+===================================+========================================+
|VISCII:Chµ:VIQR : TΩn Anh Ngµ      |VISCII:Chµ:VIQR : TΩn Anh Ngµ           |
+------:---:-----:------------------+------:---:-----:-----------------------+
|  002 :  : A(? :A breve hook-above|  112 : p : p   : p                     |
|  005 :  : A(~ :A breve tilde     |  113 : q : q   : q                     |
|  006 :  : A^~ :A circumflex tilde|  114 : r : r   : r                     |
|  020 :  : Y?  :Y hook-above      |  115 : s : s   : s                     |
|  025 :  : Y~  :Y tilde           |  116 : t : t   : t                     |
|  030 :  : Y.  :Y dot-below       |  117 : u : u   : u                     |
|  065 : A : A   : A                |  118 : v : v   : v                     |
|  066 : B : B   : B                |  119 : w : w   : w                     |
|  067 : C : C   : C                |  120 : x : x   : x                     |
|  068 : D : D   : D                |  121 : y : y   : y                     |
|  069 : E : E   : E                |  122 : z : z   : z                     |
|  070 : F : F   : F                |  128 : Ç : A.  : A dot-below           |
|  071 : G : G   : G                |  129 : ü : A(' : A breve acute         |
|  072 : H : H   : H                |  130 : é : A(` : A breve grave         |
|  073 : I : I   : I                |  131 : â : A(. : A breve dot-below     |
|  074 : J : J   : J                |  132 : ä : A^' : A circumflex acute    |
|  075 : K : K   : K                |  133 : à : A^` : A circumflex grave    |
|  076 : L : L   : L                |  134 : å : A^? :A circumflex hook-above|
|  077 : M : M   : M                |  135 : ç : A^. :A circumflex dot-below |
|  078 : N : N   : N                |  136 : ê : E~  :E tilde                |
|  079 : O : O   : O                |  137 : ë : E.  :E dot-below            |
|  080 : P : P   : P                |  138 : è : E^' :E circumflex acute     |
|  081 : Q : Q   : Q                |  139 : ï : E^` :E circumflex grave     |
|  082 : R : R   : R                |  140 : î : E^? :E circumflex hook-above|
|  083 : S : S   : S                |  141 : ì : E^~ :E circumflex tilde     |
|  084 : T : T   : T                |  142 : Ä : E^. :E circumflex dot-below |
|  085 : U : U   : U                |  143 : Å : O^' :O circumflex acute     |
|  086 : V : V   : V                |  144 : É : O^` :O circumflex grave     |
|  087 : W : W   : W                |  145 : æ : O^? :O circumflex hook-above|
|  088 : X : X   : X                |  146 : Æ : O^~ :O circumflex tilde     |
|  089 : Y : Y   : Y                |  147 : ô : O^. :O circumflex dot-below |
|  090 : Z : Z   : Z                |  148 : ö : O+. : O horn dot-below      |
|  097 : a : a   : a                |  149 : ò : O+' : O horn acute          |
|  098 : b : b   : b                |  150 : û : O+` : O horn grave          |
|  099 : c : c   : c                |  151 : ù : O+? : O horn hook-above     |
|  100 : d : d   : d                |  152 : ÿ : I.  : I dot-below           |
|  101 : e : e   : e                |  153 : Ö : O?  : O hook-above          |
|  102 : f : f   : f                |  154 : Ü : O.  : O dot-below           |
|  103 : g : g   : g                |  155 : ¢ : I?  : I hook-above          |
|  104 : h : h   : h                |  156 : £ : U?  : U hook-above          |
|  105 : i : i   : i                |  157 : ¥ : U~  : U tilde               |
|  106 : j : j   : j                |  158 : ₧ : U.  : U dot-below           |
|  107 : k : k   : k                |  159 : ƒ : Y`  : Y grave               |
|  108 : l : l   : l                |  160 : á : a.  : a dot-below           |
|  109 : m : m   : m                |  161 : í : a(' : a breve acute         |
|  110 : n : n   : n                |  162 : ó : a(` : a breve grave         |
|  111 : o : o   : o                |  163 : ú : a(. : a breve dot-below     |
+===================================+========================================+
                                                                        60
 Ph° L°c B: Mτu T± Vi«t Li«t KΩ Theo Th╤ T± Mπ S» VISCII Thºp PhΓn (ti¬p theo)
+========================================+===================================+
|VISCII:Chµ:VIQR : TΩn Anh Ngµ           |VISCII:Chµ:VIQR : TΩn Anh Ngµ      |
+------:---:-----:-----------------------+------:---:-----:------------------+
|  164 : ñ : a^' : a circumflex acute    |  210 : ╥ : O`  : O grave          |
|  165 : Ñ : a^` : a circumflex grave    |  211 : ╙ : O'  : O acute          |
|  166 : ª : a^? :a circumflex hook-above|  212 : ╘ : O^  : O circumflex     |
|  167 : º : a^. :a circumflex dot-below |  213 : ╒ : O~  : O tilde          |
|  168 : ¿ : e~  :e tilde                |  214 : ╓ : y?  : y hook-above     |
|  169 : ⌐ : e.  :e dot-below            |  215 : ╫ : u+` : u horn grave     |
|  170 : ¬ : e^' :e circumflex acute     |  216 : ╪ : u+? : u horn hook-above|
|  171 : ½ : e^` :e circumflex grave     |  217 : ┘ : U`  : U grave          |
|  172 : ¼ : e^? :e circumflex hook-above|  218 : ┌ : U'  : U acute          |
|  173 : ¡ : e^~ :e circumflex tilde     |  219 : █ : y~  : y tilde          |
|  174 : « : e^. :e circumflex dot-below |  220 : ▄ : y.  : y dot-below      |
|  175 : » : o^' :o circumflex acute     |  221 : ▌ : Y'  : Y acute          |
|  176 : ░ : o^` :o circumflex grave     |  222 : ▐ : o+~ : o horn tilde     |
|  177 : ▒ : o^? :o circumflex hook-above|  223 : ▀ : u+  : u horn           |
|  178 : ▓ : o^~ :o circumflex tilde     |  224 : α : a`  : a grave          |
|  179 : │ : O+~ :O horn tilde           |  225 : ß : a'  : a acute          |
|  180 : ┤ : O+  :O horn                 |  226 : Γ : a^  : a circumflex     |
|  181 : ╡ : o^. :o circumflex dot-below |  227 : π : a~  : a tilde          |
|  182 : ╢ : o+` : o horn grave          |  228 : Σ : a?  : a hook-above     |
|  183 : ╖ : o+? : o horn hook-above     |  229 : σ : a(  : a breve          |
|  184 : ╕ : i.  : i dot-below           |  230 : µ : u+~ : u horn tilde     |
|  185 : ╣ : U+. : U horn dot-below      |  231 : τ : a^~ :a circumflex tilde|
|  186 : ║ : U+' : U horn acute          |  232 : Φ : e`  : e grave          |
|  187 : ╗ : U+` : U horn grave          |  233 : Θ : e'  : e acute          |
|  188 : ╝ : U+? : U horn hook-above     |  234 : Ω : e^  : e circumflex     |
|  189 : ╜ : o+  : o horn                |  235 : δ : e?  : e hook-above     |
|  190 : ╛ : o+' : o horn acute          |  236 : ∞ : i`  : i grave          |
|  191 : ┐ : U+  : U horn                |  237 : φ : i'  : i acute          |
|  192 : └ : A`  : A grave               |  238 : ε : i~  : i tilde          |
|  193 : ┴ : A'  : A acute               |  239 : ∩ : i?  : i hook-above     |
|  194 : ┬ : A^  : A circumflex          |  240 : ≡ : dd  : d bar            |
|  195 : ├ : A~  : A tilde               |  241 : ± : u+. : u horn dot-below |
|  196 : ─ : A?  : A hook-above          |  242 : ≥ : o`  : o grave          |
|  197 : ┼ : A(  : A breve               |  243 : ≤ : o'  : o acute          |
|  198 : ╞ : a(? : a breve hook-above    |  244 : ⌠ : o^  : o circumflex     |
|  199 : ╟ : a(~ : a breve tilde         |  245 : ⌡ : o~  : o tilde          |
|  200 : ╚ : E`  : E grave               |  246 : ÷ : o?  : o hook-above     |
|  201 : ╔ : E'  : E acute               |  247 : ≈ : o.  : o dot-below      |
|  202 : ╩ : E^  : E circumflex          |  248 : ° : u.  : u dot-below      |
|  203 : ╦ : E?  : E hook-above          |  249 : ∙ : u`  : u grave          |
|  204 : ╠ : I`  : I grave               |  250 : · : u'  : u acute          |
|  205 : ═ : I'  : I acute               |  251 : √ : u~  : u tilde          |
|  206 : ╬ : I~  : I tilde               |  252 : ⁿ : u?  : u hook-above     |
|  207 : ╧ : y`  : y grave               |  253 : ² : y'  : y acute          |
| *208 : ╨ : DD  : D bar                 |  254 : ■ : o+. : o horn dot-below |
|  209 : ╤ : u+' : u horn acute          |  255 :   : U+~ : U horn tilde     |
+========================================+===================================+

       * Quy ≡╕nh VIQR c≥n cho phΘp t▀■ng tr▀ng "╨" bóng "Dd" hoúc "dD".

      ANNOUNCEMENT OF VISCII-COMPLIANT SOFTWARE APPLICATIONS - RELEASE 3

                     San Jose, California, September 1992

                                      by

                              The TriChlor Group
                             3388 Burgundy Drive
                          San Jose, CA 95132, U.S.A.
                      Email: [e-mail]



This release includes major support for Microsoft Windows 3.1,  and an upgrade
to VISCII 1.1.   It has been an unusually productive period for  TriChlor  and
TriChlor-Talk members.  Thanks everyone for a great effort. New software added
in this release:

         winvnkey, vnfont1, vn-nhac, butviet, more bdf fonts for X,
         vietxrn, vprint, thitap1.

With  the  newly  added  Windows  and  TrueType  supports,  one  can  generate
"beautiful" and "professional" publications easily, such as ≡úc san, bßo xuΓn,
bφch bßo, sßch, ...  All data files are identical across Unix, DOS and Windows
platforms, with the same keyboard entry style. If you use TriChlor software to
generate these publications, we would be glad to receive a compliment/souvenir
copy.

Most of TriChlor software provides a foundation upon which  other  Vietnamese-
dedicated applications can be built.   Following is a short description of all
programs.   All are Viet-Std VISCII 1.1 & VIQR  compliant.   Unless  otherwise
specified,   most   software   is   written/ported   by   C▀╢ng   Tñn   Nguy¡n
([e-mail]) and C▀╜ng Minh B∙i ([e-mail]).


1. FOR UNIX & X-WINDOWS:
~~~~~~~~~~~~~~~~~~~~~~~

include:  Include files for software developers.

lib:      Library routines for software developers.

fonts:    Contains X-Window bitmap fonts,  METAFONT sources for TeX,  pre-made
          TeX bitmap .pk fonts, and scalable outline Type-1 fonts.

          METAFONT can be used to generate different font sizes and  typefaces
          for TeX bitmap .pk fonts.

          X-Window .bdf fonts include:
             vn10x20.bdf, vnlucida18      [NEW]
                    designed by Thu V. Vu ([e-mail])
             vn9x15.bdf, vn-r14.bdf.

                                         61
                                                                        62
          Type 1 fonts: include all the TrueType  fonts  in  vnfont1  archived
          under VN/WINDOWS

vietterm:
          A modified version of X-Window xterm.   It allows entry and  display
          of Vietnamese characters, at the same time stays compatible with US-
          ASCII  environment,  and  uses  existing  8-bit  clean  Unix  tools.
          Changes include 8-bit Viet-Std,  support for both Athena  and  MOTIF
          widgets, radio buttons to switch/display keyboard and screen modes.

vnterm:
          A modified version of X-Window xterm.   It allows entry and  display
          of Vietnamese characters, at the same time stays compatible with US-
          ASCII  environment,  and  uses  existing  8-bit  clean  Unix  tools.
          Changes include 8-bit Viet-Std.

vn7to8:
          Filters for conversion between 8-bit Viet-Std VISCII codes and 7-bit
          Viet-Std (VIQR), for printing, posting or mailing.  SCV articles and
          Viet-Net mails (7-bit) can be printed by converting  them  to  8-bit
          Viet-Std.  Algorithm by H≈c ╨. Ng⌠ ([e-mail])

vnpstext:
          Prints 8-bit Viet-Std file to PostScript printers.   Uses PostScript
          Type 3 Courier font.  Both regular and bold typefaces are supported.
          Default to font size 10.  Prolog files can be modified for different
          font size.  Also available for DOS.

vietxmail:
          A  modified  version  of  X  Window  xmail.    Used  together   with
          "vnmailtool" to allow one to read/write 8-bit  Vietnamese  chars  in
          7-bit channel mail transparently.   Updates include  8-bit  Viet-Std
          VISCII codes.   Allows choice of different mail-agents (sendmail) or
          filters.  Needs utilities from "vnmailtool" to work.

vnmailtool:
          A set of conversion filters,  when set  up  properly,  will  process
          8-bit Viet-Std to 7-bit Viet-Std mails and vice-versa transparently.
          Allows choice of different mail-agents (sendmail) or  filters.   Can
          be used  with  vietxmail  or  vietterm/vnterm.   Also  included  are
          viet7to8 and viet8to7 for  conversion  between  8-bit  Viet-Std  and
          7-bit Viet-Std for posting or mailing.

vnelvis:
          A modified "elvis1.4" ("vi" editor clone),  tuned for 8-bit Viet-Std
          & C0 space characters.   Need vietterm or vnterm.   Works with  most
          Unix system. Also available on DOS.  Needs vietdos on DOS.

vpp8:
          A modified version of vpp, a troff preprocessor, used to print 8-bit
          Viet-Std characters.   Original program written by Hµu,  ported to C
          by NhΓn TrÑn ([e-mail]),  ported to 8-bit Viet-Std by
                                                                        63
          TriChlor.  Support for Italic added.  troff is standard on most Unix
          system.  Scalable fonts.

vnfortune:
          Same as Unix "fortune" program.  Displays Vietnamese poems, t°c ngµ,
          ca dao,  ... randomly.   Users  can  add  or  modify  the  database.
          Database consists of around 100  phrases,  contributed  by  SCV  and
          Vietnetters,    most    notably,     Nguy¡n    Th╕    Ca-Dao    (ca-
          [e-mail]).
             Author:  QuΓn TrÑn ([e-mail])

locale.sun:
          VISCII locale for SUN running OS 4.1.   When  installed  will  allow
          direct use of Sun's "vi",  "ls" and other tools that support  locale
          to work also on VISCII 8-bit Vietnamese chars.

vnless:
          A modified "less" to work with 8 bit Viet-Std Very powerful pager.

vietsc:
          A modified "sc",  a popular public domain Unix spreadsheet. Includes
          mortgage calculation examples.  May need vncurses.

vncurses:
          Vietnamese capable curses library.   Unix's curses from BSD  4.3  or
          SVR3 is not 8 bit clean.

lambai:
          A sample application for teaching Vietnamese.  Teachers can generate
          coded quiz in Vietnamese.   Students answer  in  either  English  or
          Vietnamese.  Also available on IBM PC.

vietxrn:
          Vietnamese capable version of xrn 6.17.   For posting,  reading  and
          replying using Vietnamese 8-bit  Viet-Std.   Automatic,  transparent
          conversion to Viet-Std 7-bit  VIQR  for  posting  and  mailing.   To
          compose and print in 8-bit,  needs vietterm or  vnterm,  vnelvis  or
          locale,  and vpp8 or  vnpstext.   Supports  both  MOTIF  and  Athena
          widgets.


2. FOR MS-WINDOWS
~~~~~~~~~~~~~~~~~

winvnkey:  [NEW]
          Allows entry of Vietnamese chars in Microsoft Windows  applications.
          Works with most Windows existing applications such as Word, Notepad,
          Paintbrush, ... Viet-Std compliant.
                                                                        64
vnfont1:  [NEW]
          Set of 4 Vietnamese TrueType fonts, VISCII encoding.
               HeoMay   (vn-helvetica)
               HoangYen (vn-present-script)
               MinhQuan (vn-courier)
               UHoai    (vn-utopia)
          From these  fonts,  Windows  can  generated  many  point  sizes  and
          variations such as Italic, Bold and Bold-Italic.

monmoi:    [NEW]
          Add Vietnamese description  to  icon.   Window's  File  Manager  has
          trouble doing this.

thitap1:   [NEW]
          Electronic poetry collection of 11 poems by Quang D√ng.  Display and
          print poems using Window Help facility.



3. FOR MS-DOS
~~~~~~~~~~~~~

vietdos:
          Allows entry and display  of  Vietnamese  chars.   Works  with  most
          existing software such as vi,  vnelvis,  pctools,  xtree,  MS  Edit,
          WordPerfect, brief, Paradox, Word, C, Pascal, ASM, ...  Requires EGA
          or VGA.  Use vprint or vlaser to print.

vnelvis:
          "elvis1.4",  a "vi" editor clone,  tuned for Vietnamese. This is the
          same version of vnelvis on Unix, ported to DOS.

viet78:
          filters for 7-bit (used by SCV,  vietnet) to 8-bit  Viet-Std.   Also
          from 8-bit Viet-Std to 7-bit.

less:
          public domain "less" pager,  better than  "more",  with  environment
          variable for charset tuned for 8-bit Viet-Std

lambai:
          same version  of  lambai  on  Unix,  ported  to  DOS. Dedicated  for
          teaching/testing using Vietnamese language. Required vietdos.

vlaser2:
          printing 8-bit Viet-Std Vietnamese  on  laser  printers. Supports  4
          typefaces.
             Author: H≥a Nguy¡n ([e-mail])
             He is also author of viet1p6, a Vietnamese word processor.

                                                                        65
testdata.std:
          A  file  of  VISCII-1.1  8-bit  Vietnamese  characters  for  testing
          software.
             Author: H≥a Nguy¡n ([e-mail])

vn-nhac:         [NEW]
          A colorful program that display collection of Vietnamese songs, menu
          driven.   Currently, there are four volumes of 200 songs contributed
          by SCVers and Viet-Netters.
             Author: H░ Ph▀╛c H∙ng ([e-mail])


butviet:         [NEW]
          A Vietnamese editor, Viet-Std VISCII and VIQR compliant, can be used
          with CGA, Hercules, EGA, VGA.  Has multi-windows and mouse support.
             Author: Nguy¡n Doπn V▀■ng ([e-mail])

vprint:          [NEW]
          VISCII 1.1 printer driver for Epson and IBM dot-matrix printer.  Can
          be invoked from a pipe.
             Author: Nguy¡n Doπn V▀■ng ([e-mail])

convert 1.3:     [NEW]
          Convert between many Vietnamese coding formats, including VISCII and
          VIQR.
             Author: Vietnamese Professional Society
                     (H╡i ChuyΩn Gia Vi«t Nam)
                     Email: [e-mail]



4. WORK IN PROGRESS
~~~~~~~~~~~~~~~~~~~

vietvu:
          similar to winvnkey, can also use clipboard besides message hook.
             Author: V√ V. ChΓu ([e-mail])

thitap:
          a bookshelf of poetry volumes,  currently has thitap1 (Quang  D√ng).
          Will have thitap2 (Nguy¡n Bφnh),  thitap3 (nhi½u tßc giΣ),  ... Many
          poems were  posted  by  Viet-Netters  and  SCV-ers. Needs  volunteer
          editors! Contact [e-mail] if interested.



5. PROPOSED PROJECTS (Need volunteers!)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

TrueType fonts:
          Additional fonts are  needed.   Should  be  public  domain  and  re-
          distributable.
                                                                        66
text2speech:
          to encode Vietnamese speech and implement text to speech  conversion
          for Vietnamese text files.

multimedia application:

vietspell:
          a dictionaryless speller based on lex/yacc.



6. GETTING SOFTWARE
~~~~~~~~~~~~~~~~~~~

There is an archive site at sonygate for anonymous ftp,  or if you don't  have
ftp,  there is a mailserver at saigon.com.   Pre-compiled binaries for  IBM-PC
and Sun 4/SPARC are also available via floppy disks.   Binaries  are  compiled
under Sun OS 4.1, X11R4, OpenWindow 3.0, Motif 1.1.


6.1 Using Anonymous ftp
~~~~~~~~~~~~~~~~~~~~~~~

Here are the instructions for accessing the anonymous FTP archive.


  Site name: sonygate.Sony.Com [192.65.138.2] (try 192.65.138.240 if the
                                               first one does not work)
  Login as anonymous, password is anything (please use your email name).

    % mkdir VN          # where you will be compiling everything
    % cd VN
    % ftp sonygate.sony.com
    Connected to sonygate.sony.com
    220 sonygate.sony.com FTP server ready.
    Name (sonygate.sony.com:<your_login>): anonymous
    Password (sonygate.sony.com:anonymous): <type_your_name>
    331 Guest login ok, send ident as password.
    230 Guest login ok, access restrictions apply.
    ftp> cd tin/VN
    250 CWD command successful.
    ftp> type image  # (or "binary")
    200 Type set to I
    ftp> get <file>.tar.Z
    ftp> quit

    If you have question/problem accessing the archive, please contact
    Tφn LΩ ([e-mail])

                                                                        67
6.2 Using Modem or MailServer at Saigon.COM
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    From: [e-mail]  (Tφn LΩ)
    Subject: [TECH] Instruction for accessing Viet files
    Date: Tue, 4 Aug 92 00:19:56 GMT


The mailserver at Saigon.COM has been updated to contain the latest files from
the anonymous FTP archive at sonygate.   This is for those of you without  FTP
access.  You can retrieve the files by sending email requests.

You may also dialup Saigon.COM directly and download the files. The front  end
to it is Station Zebra,  a BBS.   It is located in the San Francisco Bay Area.
The phone number is 408-739-1520.  Login with user id of BBS.

   [ MailServer Instructions ]

Vietnet and Vietnamese Software MailServer Archive
-------------------------------------------------

This is the MailServer archive for Vietnet and Vietnamese software.   It is  a
duplicate of the anonymous FTP archive.   The purpose of this mailserver is to
allow those without ftp access to retrieve these files.

The MailServer looks for instructions in the body of the  message. It  ignores
the subject line.  Following are some of the instructions that it understands:

        LIST directory_name
        DIR directory_name
        CD directory_name
        GET filename
        HELP

Please make sure that your email address in the header of the message is good.
The MailServer uses that to send the files to you.   I have gotten  a  lot  of
bounced messages from those of you with unreachable email address.

Here is an example session:

        $ mail [e-mail]
        Subject: <anything here>
        DIR viet.net
        GET viet.net/vnterm.tar.Z
        EOF (type a Control-D)

The email address of the server is [e-mail] Some of you  who  are
in the UUCP domain might have  problem  sending  mail  to  an  Internet  style
address.  You could try the following bang paths:

        uunet!sonyusa!szebra!mailserver
        claris!szebra!mailserver
        zorch!szebra!mailserver
                                                                        68
6.3 Using Modem to Phsys.COM
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

   VIETNET BBS - phsys.com (PH. Systems)
        Phone No: 1-213-734-8528  (Modem: Hayes V32-V42, 8N1, 300-38400bps)
           Login: NEW  (then fill out questionaire for account)
           Owner: H∙ng P. H░, [e-mail]  (Los Angeles, California, USA)
           Mail Server will be in operation starting 2nd week of Oct 92,
           please mail to [e-mail] for more info

The BBS is a 24-hour public system with Vietnamese newsgroups,  Job  listings,
limited Usenet News,  mail connections to many other nets including  Internet.
Archive site of all Vietnamese public-domain software and technical  documents
from  Viet-Std,  Vietnamese  Professional  Society,  TriChlor,  MAT/CAT,  etc.
Graphics section has been updated with *.gif from the Viet  Nam  War  era. You
are  welcome  to  login  and  download  software/gif.    Please  leave  him  a
message/mail so he will give you full access.


6.4 Using Post Office
~~~~~~~~~~~~~~~~~~~~~

For DOS/Windows send $5.00 to cover the set of 2  floppy  disks,  and  specify
your disk type (5-inch disk or 3.5-inch disk).   The disks come with very easy
to set up and install program, that will put executables in proper place,  and
installed to proper Windows group.

For Sun Sparc/Sun4/X-Windows binaries,  send $10.00 to cover a set of 2 floppy
disks containing the following programs: vietterm,  vnpstext,  vpp8,  vietxrn,
vnelvis, locale, ...  and bdf fonts.

Make check payable to C▀╢ng T. Nguy¡n (acting accountant) and send to:

                TriChlor Software
                3388 Burgundy Drive
                San Jose, CA 95132, U.S.A.

We definitely are not going to profit from  this.   The  $5  covering  mailing
expenses does  not  come  anywhere  close  to  representing  the  huge  effort
everyone,  named and unnamed,  has put into  this,  both  in  terms  of  time,
keyboard stress,  head strain,  and out-of-pocket expenses to  buy  compilers,
font converters, hardware, etc. to bring VISCII/VIQR compliant software to the
Vietnamese community. If you like what we are doing and would like to to  help
us meet expenses---which are quite real and  considerable---your  contribution
in any amount is always welcome, but is never required.


[ For an experimental  period,  precompiled  binaries  for  a  select  set  of
TriChlor software will be available via anonymous  FTP  to  Haydn.Stanford.EDU
(36.22.0.47).   Although limits are not enforced,  please try to schedule your
access between 7PM - 6AM Pacific time.  Depending on the resulting load on the
system, this service may become permanent, off-hours, or altogether removed.
                                                                        69
Binaries are currently  available  for  DecStation  3100,  Sun-3,  and  Sun-4.
Knowledgeable users with binaries compiled for these and other  platforms  are
encouraged   to   make   them   available   to   other   netters.    Trichlor-
[e-mail] is a mailing list for that  circle  of  volunteers
who participate in this distribution system.   If you have  binaries  and  are
interested in helping,  please send a "SIGN-ON" message to
          [e-mail]
 ]


7. COPYRIGHT
~~~~~~~~~~~~

All software products are copyrighted by respective authors  or  the  TriChlor
group.  All rights reserved.  You may copy, distribute, and use these software
for any purpose, as long as you do not charge a fee for doing so,  and include
this copyright notice.   All software is provided "as is",  without express or
implied warranty.

All software is provided free of charge as copyrighted  freeware  (that  means
you can have it for free and make as many copies for as many  friends  as  you
like,   but  we  reserve  the  copyright  in  order  to  prevent  unauthorized
modifications and to prevent people from selling it to unsuspecting users---we
want to keep the freeware free).


8. ABOUT THE TRICHLOR GROUP
~~~~~~~~~~~~~~~~~~~~~~~~~~~

TriChlor is a group of  volunteers  whose  common  interests  are  to  provide
quality and free utilities,  codes,  libraries for  use  with  the  Vietnamese
language,   and  to  integrate  Vietnamese  language  to   current   computing
environments.

If you would like to pitch in and help, we can be contacted at
         [e-mail]
If you wish, you may get on the mailing list and participate in discussions by
sending a message to
         [e-mail]
with the subject
         "SIGN-ON".

But please be prepared to donate your time :-).
All works, unless specified, will be placed in public domain.




The TriChlor Group