Re: Source file names (was Using prototypes...)

Lim Hock-Chai wrote:

I wonder if it would be easier for IBM to simply increase the 10-character limit to say 30-char. It is not very often for
developer to name a program/procedure that is more than 30-char
long. I know it is a bit short sighted. But I really think that
the reason for iseries being so easy to work with is because of
the way the library list and object works.

Any change to the object name length would be a massively costly investment irrespective of the name length. Thus IMO, as a niche OS, there is almost zero possibility the limitation would change. There would be a tiny possibility that there could be both a short name and a long name for every /QSYS.LIB object, perhaps [and IMO best] implemented with an additional index object for the *LIB object type. Such a change could then /replace/ the database-only implementation of the SQL long names.

The system needs too many significant changes to ever possibly become mainstream, but no significant changes are likely to transpire until [even if] it becomes mainstream. If the system became mainstream without significant changes, that should make obvious, that no significant changes were required and sufficiently justifies never implementing them. Such is the life of an OS without massive investments behind it.

Regards, Chuck