Re: RDi 9.5 announcement

On Thu, Sep 3, 2015 at 9:16 AM, Roche, Bob <broche@xxxxxxxxxxxxxxxxx> wrote:

Why do you not like sequence numbers? When there is a problem in the code, isn't it nice to have the system tell you where it is? It's not like RDI make syou look at them.

I think folks that don't like sequence numbers are specifically
against *stored* sequence numbers. Personally, I consider them to be
mostly harmless, but completely useless. I have auto-renumbering on,
and compiler listings have their own sequence number anyway.

When I say "mostly harmless", the very small harm I see is threefold:
(1) costs a small amount of storage, (2) costs a minuscule amount of
CPU to do recalculation, (3) creates a slight amount of confusion
(does a line number refer to the stored sequence number or the
compiler-assigned sequence number?). I cannot stress enough that these
are very, very small costs. But for me, there is zero benefit. Display
of sequence numbers should be handled dynamically by the editor, and
error messages should report the ordinal line number accordingly. You
can see in RPG IV compiler listings and dumps that IBM had precisely
this in mind. And folks programming in other languages on other
platforms had been doing without stored line numbers (happily!) for
decades prior to RPG IV.

John Y.