There are quite a few IBM mainframe systems using EBCDIC on the
Internet now. The problem with EBCDIC is that there are a lot of
different versions of it and no agreement on which one to use. There
are country specific versions that substitute various national
characters for things like braces. This is similar to the national
variations of the ISO 7 bit code (ASCII being the US version). But
even within the US, for example, there are different variations.
Relatively recently EBCDIC variations have been developed that have the
same character set as ISO 8859-1. Unfortunately, IBM chose to use
different code assignments for different countries in order to preserve
upward compatibility with the older country specific versions.
A group within SHARE and other IBM mainframe user groups has been
working (with considerable success) to get IBM to adopt a single EBCDIC
that uses the ISO 8859-1 (Latin 1) character set. However, it will be
quite a while before conversion to such a version of EBCDIC can happen.
ASCII-EBCDIC conversion can usually be made fairly consistent within an
installation. It's a pain though, as translate tables in various
products have to be adjusted. But exchanging EBCDIC data using more
than a mimimal set of characters among installations can be a problem.
One of the advantages of TCP/IP is supposed to be interoperability
among different types of computers. A narrow view of what the internal
code of a machine ought to be does not seem to be in line with that
philosophy.
Roger Fajman Telephone: +1 301 402 1246
National Institutes of Health BITNET: RAF(_at_)NIHCU
Bethesda, Maryland, USA Internet:
RAF(_at_)CU(_dot_)NIH(_dot_)GOV