5.0403 Multilinqual Applications (1/59)
Elaine Brennan & Allen Renear (EDITORS@BROWNVM.BITNET)
Mon, 21 Oct 1991 23:50:24 EDT
Humanist Discussion Group, Vol. 5, No. 0403. Monday, 21 Oct 1991.
Date: Mon, 21 Oct 91 12:07:39 CDT
From: email@example.com (John Baima)
Subject: Multilingual Apps
Sorry for the long delay in picking up this thread, but I
would like to offer a somewhat different perspective.
John Hughes writes, "Practically speaking, would
Microsoft or Apple or NeXT recoup their costs if they
decided to provide HUMANISTS with the kind of operating
system and interface we want?". John goes on to give a
very gloomy assessment about the prospects of such an OS.
However, Microsoft is committed to providing a full
Unicode implementation in Windows/NT. Apple's "Pink OS"
is designed around Unicode. NeXT will provide Unicode
support in a future OS as well. I do not know what other
plans IBM and Apple have, although I have some hope that
both will provide Unicode before Windows/NT or Pink. Mac
and OS/2 2.0 support 16 bit character sets. Unicode will
will come along with time. With the recent decisions in
the ISO community, it seems very likely that ISO will
endorse the Unicode way of doing things next year.
Whenever ISO 10646 becomes an official standard, it will
then eventually become a purchasing requirement for many
I must also note that I read John's description of how
one must go about developing multilingual DOS apps and
printer drivers with some horror, "The only 100%
foolproof solution to these firmware woes is for software
developers to get _all_ the old versions of the firmware
for _each_ printer they wish to support and make their
initial printers backwards compatible for each supported
printer and then to get all the new firmware revisions
for each supported printer and continually issue upgraded
printer drivers that take account of each new firmware
revision. Neither of these courses is practical. . . .
Finally, one of the most cost-intensive aspects of
software development for multilingual DOS applications is
creating the printer drivers." This sounds like a good
reason to call DOS a dinosaur to me :-).
I would like to suggest that a much less expensive
alternative would be to use an OS/Windowing environment
that uses Type 1 fonts (maybe TrueType in the future). On
PC's, this includes Windows 3.0 (+ATM), Mac (+ATM, soon
to be built in) or OS/2 (built in). Here, if someone
wants to build an app, they only need to design one font.
This same font can be used to display and print. If one
does not have the ideal of a unified imaging model (and
all the NeXT people said, "Amen!"), one should at least
have a unified font model for screen display and
printing. This seems to me to be the only rational, low
cost way to do things. Bitmapped fonts and writing
printer drivers should be avoided at all cost. Or rather
it will cost you plenty in time and money to do it "the
old fashioned way" of bitmaps and device drivers.