Re: What counts as technically slick?

Scott,

I understand your viewpoint.  We hedged our bets and put all
SETLL/READ/CHAIN opcodes in a separate subroutine by file and use /COPY
to incorporate them into the programs.

We figured this way if SQL record level access becomes mandated or we
need to access a file on another platform (hasn't happened yet), we can
just change the  contents of the /COPY member and recompile.  If server
programs are to be used, again, we change the contents of the /COPY and
off we go.

IMHO, this satisfies the need to isolate the application programmer from
the mechanics of DB I/O without adding any complexity.  It's real easy
to get a programmer to do EXSR CUSTGET instead of CHAIN CUSTREC.


Scott Klement wrote:
> 
<<snip>>
> 
> I guess I could see doing this if I thought the files would soon be
> located on another computer, on another platform, where I'd save the
> work of re-writing all of the programs involved.   But how often does
> something like this ever happen?   It never has, for me.
>
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---