valenz@bnl.UUCP (Greg Valenzuela) (01/21/84)
Although word processors are very common now, there isn't(and maybe
never will be) a standard code for special characters(e.g. greek characters,
subscripts,superscripts,etc) similar to ASCII. It would be nice if there
was some software which could translate a file produced by one type of WP
into the characterset of another type.
We deal with documents written on Philip's MICOM system, the CP/T system,
and MASS11 running under VMS. The default procedure would be for us to try
to get the character codes from the respective companies and write a
translator which would run on the VAX. I would be interested in hearing
from anyone who has had to transfer files between the MICOM and either
the CP/T or MASS11.
Is this a lost cause? We can at least transfer the ASCII parts of the
files, but it would be nice to be able to read equations too.
Greg Valenzuela
USENET: {harpo|seismo,decvax|linus}|philabs|sbcs|bnl|valenz
ARPA: valenz@bnlguy@rlgvax.UUCP (Guy Harris) (01/23/84)
I remember hearing of an extension to ASCII to support various special
characters, which the DEC personal computers were using. Does anybody know
whether this is an official ANSI X3 standard, and what the number is?
(Unfortunately, even though most word processing software probably uses ASCII
for the regular characters, they all had to choose their own extension to
ASCII for the special characters, so such a standard may be closing the
barn door later than would be desired.)
For things like equations, it may not help, as some systems may build up
formulae out of regular characters while others may choose a special
internal representation, as equations are more complicated beasts than text.
Doing it "right" means implementing a structured implementation, where instead
of just saying "a sigma appears here" the sigma is coupled with the summation
limits and with the term of the sum - sort of like the difference between
doing it directly in "troff" and doing it with something like EQN or TeX.
A company called Altertext, in Boston, makes a Z80-based box with two rather
flexible floppy-disk drives (which can read several disk formats) and a
communications link, and they have software that translates between various
document formats by going into and out of an internal code; you may be able
to get what you want with their box.
Guy Harris
{seismo,ihnp4,allegra}!rlgvax!guylmc@denelcor.UUCP (01/23/84)
ANSI has published a standard on how to extend ascii (X3.41-1974: Code Extension Techniques for Use with the 7-bit Coded Character Set of ASCII) and a standard on a specific extension (X3.64-1979: Additional Contols for Use with ASCII), but it is only control characters (Cursor Up, Absolute Positioning, etc). As far as I know, that's it for standards. -- Lyle McElhaney (hao,brl-bmd,nbires,csu-cs,scgvaxd)!denelcor!lmc
steveg@hammer.UUCP (Steve Glaser) (01/30/84)
There's also an ANSI standard somewhere in the mill (nowhere near approval yet, I don't even thing there is a DRAFT out). I forget the number they assigned for it. The Navy apparently is trying to get an interchange format that they can require from vendors. The standard is supposed to address things like subscript representation, charactersets, and such. Also, there is a revision of X3.4 undersay (the "ASCII" standard) to produce an encoding for the upper half of the 8-bit ASCII set so that we can handle national alphabets reasonably. Ever try to run unix on a terminal where the vertical bar comes out as an 'o' with an umlat? Unforthnately, there appears to be no chance for changing the current situation of overloaded bit encodings, but at least we may have one set that can "express" all the characters (albeit not in the places that the various countries would like them). Steve Glaser tektronix!steveg steveg.tektronix@rand-relay