> when they run both systems both systems in parallel for
> 6 months, and try to find the cause of the two systems giving different
> results

I see issues there--the Horizon scandal in UK suggests a different
possibility, where they either don't run the 2 systems in parallel and
simply throw the switch over, or they do, but blame any difference on
people after they get some rent-a-mouth to state the code is working fine.

Roops



On Fri, 4 Apr 2025, 18:15 Colin Paice, <
[email protected]> wrote:

> They may be able to automate changing the code to a fancier language -
> that's the "easy bit"
>
> I can see the problems occurring when they run both systems in parallel for
> 6 months, and try to find the cause of the two systems giving different
> results - this is the hard bit.
> I remember having problems when we changed the version of the compiler...
> it exposed a timing window - and this was on the same code base!
> This situation is when you need tracing etc in the programs, which you will
> not from the automatic conversion.
> Colin
>
>
>

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to