Specifications

CONSTRUCTION
F
or displaying text of Hindi, En-
glish, or any other Indian language
on a TV-like screen, as may be
required for public announcements or for
educative programs etc, presently there
are two possible ways:
1. Use a personal computer (PC), with
all its hardware, such as the hard disk,
monitor etc, and develop or buy a suit-
able software to display such text on its
screen using its keyboard.
2. Develop a dedicated low-cost mi-
croprocessor based system employing a
CRT controller circuit, with suitable firm-
ware for each of the languages.
This article provides the software for
use with a PC based system as well as
use of a dedicated microprocessor based
system, complete with circuitry and firm-
ware programs. Incidentally, it introduces
an important aspect concerning coding of
text characters for Indian languages and
provides an efficient solution. The soft-
ware developed for both the above
schemes of display is based on the pro-
posed simplified coding solution.
ASCII codes and Indian languages
The typewriter for the English language,
along with its mechanism, has been al-
ready adopted for almost all of our In-
dian languages with practically no change
in its layout. The positions of keys and
their operation remain unchanged. It has
the same four rows of keys, including shift
and space keys, with top row for numer-
als, and so on.
The combination vowels in Indian lan-
guages such as ‘oo’ are made as separate
‘hook’ characters, which upon stroke are
non-space-moving. Persons involved in
development and adoption of the English
keyboard have cleverly tackled the prob-
lem of typing the large number of charac-
ters involved in most of the Indian lan-
guages, in contrast to a mere 52 (2 x 26)
characters in English language. In spite of
the fact that Hindi, or for that matter
Tamil or Telugu etc, have to deal with a
large number of basic consonants, which
combine, singly or doubly, with a similarly
large set of vowels, the four-row keyboard
deals with all of them adequately to en-
sure fast typing. Our trained typists are
able to make up to 40 strokes per minute
(approximately 15 words per minute) in
the most intricate of Indian languages.
Today, there is both a concern and
talk in several circles, e.g. computer, tele-
communications, and other hi-tech indus-
tries, to develop a new type of keyboard
layout for Indian languages, with the aim
of making the software development task
easy enough. In this context, phonetic key-
boards have been proposed and are also
in vogue already, with a great deal of soft-
ware available commercially. These key-
boards do not make use of hook charac-
ters, but then such a keyboard is not well
suited for training.
Generally, typewriting is a process
based on direct eye-to-limb reflex signal
generation, with little thinking going on
deep down in the brain. If the typist looks
at a letter ‘hu’, he presses the ‘ha’ key
first, and as he sees a hook ‘oo’ below it,
he strikes the corresponding hook key,
and so on. Thus the process can be speeded
up with practice and is not easily forgot-
ten. We know of language typists, who
even after 30 years of work, continue to
type as fast as they did when they were
young. Some of them can even talk while
typing without missing anything.
Now, consider the phonetic keyboard
typist. He has to split each letter men-
tally into its vowel and consonant parts;
find out the consonant key and the hook
key, and then press them in proper se-
quence. Here, the direct eye-to-finger re-
flex does not take place, because there is
a thinking process involved. For example,
for typing ‘hoom’, he has to know the
grammer to split that into a ‘ha’, a ‘oo’,
and an ‘mm’. Thus the typist does not
pick up speed even after considerable
practice, and as a result fatigue sets in
quickly for him.
As mentioned earlier, the Indian lan-
guages have more characters than those
in the English language. The well-known
ASCII codes for English are just 128 in
number, including several control charac-
ters and punctuation marks. Each code
occupies one byte and hence the total code
space is 7FH for the complete English set.
With our Indian languages, we have var-
ied sets of characters. Consonants are quite
many, and therefore it is not easy to ac-
commodate all characters within the same
set space of 128. For this purpose, the
author proposed a scheme of forming such
ASCII-like codes for Hindi as well as other
Indian languages, which occupy a space of
128 bytes only for each. Based on the stan-
dard typewriter format for English, Hindi,
Tamil etc, a method for typing text of
these and other languages, all simulta-
neously, is described in this article.
This proposed scheme of coding would
not cause any disturbance to the present
typists of these languages. They do not
have to undergo fresh training for using
the proposed keyboard.
The method of making the ASCII code
set for any Indian language is based on
the typist’s existing keyboard. For ex-
ample, in English, the letter ‘d’ has its
code as 64H. So, the code for the Hindi
letter ‘ka’ (d) is also 64 hex. For Tamil
ASCII code 64 hex is used for ‘na’, and so
on… for the other languages. A table of
such codes for the three languages Hindi,
English, and Tamil is shown in Table I.
Character generator for Indian
languages. While characters of any lan-
guage need a character generator, which
puts dots in a rectangular matrix to de-
pict the shape of the character on the
screen, the English language, in its sim-
plest form of display, manages to write
all its characters within a 5 x 7 matrix.
Therefore, within an 8 x 8 matrix, there
is enough gap to allow for inter-character
and inter-row space. But, in Hindi and
DISPLAY SCHEMES FOR INDIAN
LANGUAGESPART I
(HARDWARE AND SOFTWARE)
K. PADMANABHAN, S. ANANTHI, K. CHANDRASEKHARAN,
AND P. SWAMINATHAN
120