One could add that ANSI is the character set used by Windows but not by Dos or the Mac. Just to give an idea of the differences: ANSI uses the range 192 to 255 for accented characters - neatly laid out with each lowercase character exactly 32 above its uppercase equivalent. The range 128 to 191 contains miscellaneous characters - the copyright symbol for instance. The Dos 437 character set puts accented characters between 128 and 175, line drawing characters between 176 & 223, Greek characters and other oddities from 224 to 255. As an example, character code 233 is an acute e é character under ANSI but a Greek theta character under Dos. In addition, Hewlett Packard - for one - has sets of its own - Roman 8 and Latin 1 are examples - in addition to the standard ones. If one wants to print characters which aren't in the character set one's using, there are two basic substitution possibilities, depending on what one's printer has in its quiver. - one is to manufacture the accented character using other characters - to print an e acute é, one might print the e, tell the printer to go back one position then print the character which most closely resembles the accent in question (as in Michael Norman's quote of Norman Havens). (Alternatively, one can use a language like PCL or Postscript to draw any shape one wants: somewhere in Xy documentation is a description of how Xyquest used this method in their program documentation to print a character unavailable in any character set. The Postscript character set I believe has very few accented characters as such; it has a full set of accents though and Postscript printer drivers use this method to manufacture accented characters at print time (someone will no doubt correct me if this isn't the case).) - the other is to find a character set in the printer's repertoire which has the accented character in question and get Xy's substitution table to issue the commands for switching to that character set, printing the accented character in question then switching back to the default character set. In each case, one has to choose a fall-guy - some character which is available in the on-screen character set but which will be converted to the wanted character during printing. If though one simply wants to transfer text from dos to windows, a simple xpl program can do the conversions (where, that is, the characters exist in both sets: Dos doesn't have a copyright symbol, ANSI doesn't have Greek characters. And neither has Postscript's set of accents-on-their-own.) Unicode holds out the promise of an end to this chaos. It isn't limited to 255 characters - it has 65000+. So everyone in the world will find what they want in one and the same character set. With the tradeoff that file sizes will tend to double. John _____________________________________________________________________ GRAND JEU SMS : Pour gagner un NOKIA 7650, envoyez le mot IF au 61321 (prix d'un SMS + 0.35 euro). Un SMS vous dira si vous avez gagné. Règlement : http://www.ifrance.com/_reloc/sign.sms