Another strange mailing from the ISO8859 list. I note that I always seem to get these aberrations from BYUADMIN. Warning, there are unprintable characters at the start of what follows. I note that I always seem to get these aberrations from BYUADMIN. Do you think there's a chance that something is flaky with their server or might there be something flaky in the setup of this particular list? /June To: LSTSRV-L(LSTSRV-L@POLYGRAF) FORWARDED MESSAGE 04/05/88 08:44 FROM [log in to unmask]: INMR02 A Dear list subscribers Mr. Melcher's letter raises some important points, which are not yet answered in a satisfactory way. The primary question is: Should the EBCDIC/ISO8859 conversion table be either code to code, or graphic to graphic? If the second is chosen, as Mr. Melcher seems to do, there are serious consequences: -- separate and different tables for Western Europe, Eastern Europe, Greece, Turkey, Israel, and the Arabic countries. -- necessary change of a table (not only the code page), as soon as ISO makes modification in a part of 8859 (in which they are completely free in doing). -- other persisting complexities with national character sets as shown in Mr. Melcher's letter. What we need is a stable and uniform conversion table (that can be implemented into hardware), independent of the meaning of the bytes in any national environment. Code pages can vary, according to local need. Such a table exists just now, (and from about 1972), and should be changed only after proper justification has been given (which I have not seen as yet). It has been suggested that CP500 was necessary for including all the 8859-1 characters. This is NOT true. Distribution can be done according to the present table. We should realize that changing the table could result in a number of very expensive modifications, without diminishing confusion at all. I hope that the matter will be discussed thoroughly at the next SEAS meeting in Davos which I am not able to attend. Yours faithfully, Johan van Wingen