Some of the html (and html practice) threads have discussed the "coding by character number" business in lengthy detail. Some caution is advised with some of them, particularly with older ones, but the full "ANSI character set" defines "a whole bunch" of characters - far more than can be coded (easily) with Alt-0xxx. That system, in "off the shelf" systems can only handle the 128 ASCII characters, or - if you load ANSI.sys - the 256 characters of the "extended ASCII set(s). The difficulty there is that there are several "extended ASCII sets. IBM tried to enforce one that still survives in a few places, and a more "universal" extended set is probably pretty much a Microsoft product.

People do frequently refer to "extended ASCII" as "ANSI" but it's really just a (broken) fragment of ANSI.

All of this has been somewhat superceded by availability of font sets, and tools to use them, for the full range of ANSI characters. With enough add-ons, one can theoretically enter all(*?) the Chinese and other "oriental" languages, Hebrew and others, with top-to-bottom and/or right-to-left line parsing, etc. An Alt-Uxxxxx coding exists, but is unlikely to work with installed programs and on the machines that many of us have.

* Some languages include large numbers of "composite characters" that require superimposing several "primitive glyph forms" to form the actual character(s) of the language, but in principle this can be handled by the new systems.

The characters themselves can be viewed at the ANSI web sites. An early "definitive volume" on the whole subject of using them (and other aspects of talking between languages and cultures) was Developing International Software: For Windows 95 and Windows NT, Nadine Kano, ©1995 by Microsoft Press, but with an earlier ©1968 in the authors name. Unfortunately, neither Win95 nor WinNT was really adapted to using the full ANSI character set, among other things because the "native word length" didn't have enough bits to represent a character in "one binary word."

With these "primitive" systems, software was "localized" to provide the ability to use the local language, but no tools were generally available to make an English (US or WestEu) talk to a Chinese one.

WinXP and probably Win2K(?) are the first Microsoft OSs with theoretical full ANSI capability, but even there it's extremely rare to find a machine set up to use "exotic mixes" of languages. The unconfirmed impression is that software intended for Chinese use has some limited "English" capability, but equipping an English system to speak Chinese appears to be available only to users of the "WinXP Server" versions. I've been unable to find whether "Win2K Server" versions can be suitably equipped.

The update to the above book, Developing International Software 2, by "Dr. Software," ©2003, Microsoft Press, gives some good info, but is still rather vague about what really is needed to write "all the languages." It's even more vague about what's available. At the time of it's release, Microsoft's print shops and other production people did not have the capability of handling "full set ANSI" characters, so all of the "foreign" characters were converted to "images" that were pasted into the text as separate objects, i.e. as "pictures" - not as "text characters."

There are setup and printing programs with full ANSI capability. TEX is one, if you get all the right plugins and fonts, but the few programs that can be upgraded now are pretty "learning intensive," so common workarounds are probably more suitable for most users. You find, or buy, a font that has the characters you want, and you cut and paste (Insert symbol or use CharMap) to get individual characters.

As long as you stick to "Western European" languages, and possibly Kanji written left to right, recent Windows versions can probably provide a font that will work for places where a few characters suffice. WinXP, probably Win2K, and possibly WinME can be equipped to read "right to left" if absolutely necessary, to handle Hebrew - sort of. Other similar languages tend to have large numbers of "composite characters" that are an additional challenge. "Pictograph" languages like Chinese are another step beyond. But we're promised "it will all be easy - - - someday."

The bottom line now is that if the "character" you want is on your machine, you can probably insert it wherever you want to, using one of the methods already given. If you're communicating digitally, and the person receiving your stuff has the character in a standard font set, you can "code" the ASCII/ANSI value of the character, and the recipient will see the character when they get the message. If neither of you has the character - then who cares?

John