Using IBM i? Need to create Excel, CSV, HTML, JSON, PDF, SPOOL reports? Learn more about the fastest and least expensive tool for the job: SQL iQuery.
It is 2012 and IBM's QSYSINC members still suck. What is going on here!
New Rule: Nothing new ships until the APIs are prototyped in all languages and the corresponding supporting data structure are correctly prototyped and tested.
I almost never use them directly using /COPY, for many of the reasons you listed. They are sometimes helpful for getting the offsets right when creating my own copy members. A few more things to add to the list...
Far too many /copy sources, both IBM and published in the wild, quite frankly suck. Do any of these people revisit their work and update it for modern standards? Do any of them realise you can now use Length/OVERLAY rather than From/To?
And we wonder why nobody learns RPG anymore, when trainers seem to be using techniques that are depricated.
Personally if I do manage to find a DS or plist that meets my needs - I rework it to a more modern coding style. And what's with IBM publishing most new prototypes in C only??
The only reason I have found to look at the QSYSINC copy books was to convert them into something we could actually use and read. The field names they define are horrendous, but that is mostly because they don't use QUALIFIED and have to make all the names unique.