Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO
From: Doug Scott (dwscott_at_ieee.org)
Date: 01/02/04
- Next message: Doug Scott: "Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO"
- Previous message: docdwarf_at_panix.com: "Re: No panic - but .. Y2K + 4"
- In reply to: Judson McClendon: "Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO"
- Next in thread: Peter E.C. Dashwood: "Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Fri, 02 Jan 2004 00:27:34 GMT
Judson,
> Even when things can be done real time you may not want to, because it
> complicates the rounding error problem, for example.
Nah - since you have to carry an error from one calculation to the next,
that can be held on a database or somewhere convenient. But I'm not
suggesting real time for /everything/ as a practical solution, come to
think of it - it's one of my beefs about some of the newer kids on the
block that they try to do it all online and end up with hideous designs,
simply because of lack of expertise in designing batch systems. They
have their own rules, but they have the magnificent feature that a batch
run is 100% repeatable - if designed properly.
I picked up an insurance system (HUON) at one stage, and it regularly
used to update DB/2 databases in batch without any commits at all, until
the end of program. If a program failed for any reason (and they did
regularly), all updates were backed out and the program could be rerun
without any database restore.
Of course, it might have been better to design the system so that such a
huge consumption of resources could be better used, but it was good for
the standby programmers.
--- Doug dwscott@ieee.org
- Next message: Doug Scott: "Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO"
- Previous message: docdwarf_at_panix.com: "Re: No panic - but .. Y2K + 4"
- In reply to: Judson McClendon: "Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO"
- Next in thread: Peter E.C. Dashwood: "Re: (not entirely...) OT: OPINION... chicken entrails, runic stones, and crystal balls... WAS CoBOL moved to OO"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|