Re: Rewriting SSA. Is This A Chance For GNU/Linux?

On 31/03/2025 04:08, pothead wrote:

On 2025-03-30, rbowman <bowman@montana.com> wrote:

On Sun, 30 Mar 2025 03:42:18 -0400, c186282 wrote:
>

    As for the SSA (and IRS) ... reports are that they're STILL using
    1960s computers for certain operations and thus really can't
    integrate anything.

>
https://arstechnica.com/tech-policy/2025/03/what-could-possibly-go-wrong-
doge-to-rapidly-rebuild-social-security-codebase/
>
I've never dealt with SSA other than as a consumer. However I have had to
deal with the DOI's IMARS (Incident Management Analysis and Reporting
System) that was deployed around 2012. Great fun!
>
Then there is the FBI's NIBRS (National Incident Based Reporting System)
that is meant to replace the UCR (Uniform Crime Reports) system that went
back to the 1920s.  The format for submitting data is a moving target and
crime in the US is under-reported since many agencies can't manage to
submit their data successfully.
>
Based on those two, I give the chance of updating the SSA system in less
than 10 years a very low probability. The US government's attention span
is more like 10 days than 10 years so good luck. Unlike the other two CFs
I have a personal interest in continuing to receive my checks every month.
>

Financial institutions are still running COBOL as well as other legacy
systems like TPF.

But probably on PCs running IBMS linux

They also run Linux on mainframes in logical partitions.
More recently is the adoption of the MongoDB platform which also runs on
'old iron' mainframes.
The once dead mainframe has made a huge comeback due to virtualization and
reliability due to being N+1 at a minimum.

It never went away really. Banks want their data *physically* secure in fully uninterruptible data centres.

However the big market is in storage devices and programming to support AI.
 the more things change, the more they seem to stay the same.
 

CPU power on the desk versus somewhere else has always been a compromise between cost and network bandwidth.
In terms of maintenance of one big corporate program the centralised single program scores heavily
  In terms of one man doing CAD or videos editing  a desktop with power scored every time.Although with gigabit the tradeoffs are changing
--
It’s easier to fool people than to convince them that they have been fooled.
Mark Twain