Character conversion in MQ from ASCII to EBCDIC or the other way around is so easy that I did not bother to write a tip about it, but on the other hand... Actually I have mostly worked with MVS to MVS communication in MQ so until recently I did not have any need for character conversion.

MQ performs character conversion when you issue a MQGET and you remember to ask for it. If you do not, well then your displayable characters does not form the result you were expecting. Please note that binary data by nature does not need to be converted (like PDF). If you are mixing both binary and displayable data in your transmissions you are looking for trouble. To carry out the conversion you must use the get message option MQGMO_CONVERT (in COBOL MQGMO-CONVERT) and nothing else.

In the MQ Application programming Reference there is a lot of description about CodedCharSetId and Encoding that confuses more than it enlightens you. My best advise is to try and ignore it unless you face problems with your MQ character conversion. The hard way I have learned that if a queue contains data created by MQPUT from an ASCII based platform and also data created from an EBCDIC based platform, then a MQGET with a MQGMO_CONVERT will convert data created in another format than the receiver while the data created in the same format as the receiver will not be converted. In other words you do not have to worry about whether data in a queue is in EBCDIC or in ASCII. You must of course worry about the conversion of binary values, but if all data is displayable you can always use MQGMO_CONVERT. I think this is pretty neat.