Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!aplcen!uakari.primate.wisc.edu!brutus.cs.uiuc.edu!jarthur!dhosek
From: dhosek@jarthur.Claremont.EDU (D.A. Hosek)
Newsgroups: comp.fonts
Subject: What Metafont is (was Re: Custom Fonts...?)
Message-ID: <3242@jarthur.Claremont.EDU>
Date: 27 Nov 89 03:10:16 GMT
References: <1533@cc.helsinki.fi>
Reply-To: dhosek@jarthur.UUCP (D.A. Hosek)
Organization: Pitzer College, Claremont, CA 91711
Lines: 41

In article <1533@cc.helsinki.fi> STICKLER@cc.helsinki.fi writes:
>Also, what is the METAFONT format used in TEX? How does one
>define a METAFONT?

Metafont is a system which takes a description of the curves and
lines composing a character and uses those to produce the font in
bitmap form (actually, there is a working prototype of a MF->PS
system, and a second MF->PS program in the works). The basic
principle is to use a number of parameters to describe assorted
characteristics of the typeface (e.g., the width of a "thick"
stroke in an uppercase letter or the height of lower case letters).
The programs for characters, then, rather than use constants to
indicate the appropriate dimensions, use the parameters. In this 
way, a wide variety of typefaces can be created from a single
character program.

As indicated by the abovementioned MF->PS conversion, Metafont and 
PostScript are not incompatible. In fact the relationship between
them can be seen in much the same light as the relationship 
between TeX and PostScript. MF provides a somewhat nicer interface
for the production of the PostScript code than "doing it by hand".
However, the Metafont motif has been slow to gain acceptance by 
font designers. Sumner Stone, in a conversation with me a month ago,
commented that type designers are unwilling to abandon the visual
approach to letterform design that a true meta-font would require.
Richard Southall (I believe) did some experiments with Metafont
design a couple of years back and had a rather lukewarm response
(his comments were published in one of the European text-processing
conference proceedings). Personally, I believe that the true
potential of Metafont has yet to be exploited. For example, in an
undertaking like the Quixote Oriental Fonts Project, visual design
of the more than 50,000 characters involved would take an extravagant
amount of time. A procedural approach seems to be the only rational 
way to approach the development of the characters.

-dh
-- 
"Odi et amo, quare id faciam, fortasse requiris?
   nescio, sed fieri sentio et excrucior"          -Catullus
D.A. Hosek.                        UUCP: uunet!jarthur!dhosek
                               Internet: dhosek@hmcvax.claremont.edu