A font represents
a collection of shapes organized in a well defined manner. In scalable
fonts, the shapes are defined mathematically so that a shape can be scaled.
The mathematical description of a shape is given in terms of a sequence
of curves which form the outline of the shape. The shapes relate to the
letters of the alphabet and special symbols used in print.
Bit mapped fonts differ from
scalable fonts in that the shapes are specified through bitmaps which are
essentially digitized forms of the letter of the alphabet. Bit mapped fonts
are specified through a matrix of ones and zeros giving one the impression
of the shape when viewed at a proper distance.
One important aspect of fonts
is the Encoding method which relates the code of a character to a shape
through appropriate naming conventions. Encoding specifies how a given
code (the numerical value assigned to a letter of the alphabet, punctuation
or special symbol) is to be mapped to a particular shape. Different encoding
standards exist and the differences relate to the language (the alphabet
corresponding to a language), the computer platform as well as the number
of shapes recognized correctly.
Eight bit fonts permit upto
256 shapes to be specified but for several practical reasons, this set
is restricted to about 180-220 shapes. The differences in Encoding usually
result in text getting rendered differently on different systems though
the Roman alphabet (English letters and punctuation) are more or less guaranteed
to be displayed without error on most systems. This set corresponds to
the standard 96 character ASCII.
Vagaries in font encoding
as well as rendering often result in display of text that differs from
the originally intended form. This is usually the result of fonts designed
according to arbitrary encodings or that the system rendering text cannot
properly handle the encoding. Systems based on Microsoft Windows follow
one encoding while the MacIntosh, rich in its graphic features understands
a totally different encoding which accommodates many more shapes. Unix
systems have traditionally restricted displays to simple English (standard
ASCII). With the introduction of X-Windows, Unix systems have provided
Graphic support but have handled mostly the Latin Character set as a basic
standard.
Fonts used in the Seventies
and Eighties were essentially Bit Mapped until, the Truetype standard as
well as the PostScript standard were introduced. During the eighties, Prof.
Knuth had introduced TeX, a typesetting application which used high quality
fonts defined through a standard devised by him called MetaFont. Metafont
is an independent approach to defining fonts and accommodates almost a
full complement of 256 shapes. Typesetting in Indian languages through
TeX was popular in the eighties and continue to be used today since TeX
has been ported to many different platforms.
These are the standards which
have been in vogue for nearly two decades.
Bit mapped fonts (bdf standard)
Metafont (Introduced as
part of TeX)
Portable Compiled Format
(pcf on Unix Systems)
ISO-8859 Encoding
standard for different European languages
MacIntosh Truetype fonts
MS Windows Truetype Fonts
Adobe PostScript Fonts (Used
with Page Printing applications)
Fonts for Chinese, Japanese
and Korean
Unicode Fonts
OpenType Fonts
It may be noted that Fonts
for Indian languages have never been standardized, for it has not been
possible to specify a suitable encoding for any of the Indian languages.
In the absence of suitable standards, it has not been possible to handle
Indian language text on computer systems in a uniform manner. This is unlikely
to change, for even among the scholars and professionals in India, there
is little agreement on what should go into a standard for writing systems
that are based on syllables rather than the letters of the alphabet.
