You might also be interested in
Replies (3)
Please login or register to join the discussion.
Small operators / large operators
Presumably the instances of prosecution involve small operators, for example single subpostoffice businesses. As such they would most likely lack the financial resource to adequately defend themselves.
Now, I wonder what would happen if a public company (eg. national chain of newsagents) received such demands from the Post Office in relation to it's operation of subpostoffice counters - would they just accept that Horizon is perfect and flawless so any alleged shortfalls must be deducted off bottom line profits? I think not. Presumably any such instances / negotiations (if they exist) would be kept quiet and free from public knowledge.
My concerns are are follows
If the issue is genuine it will affect large operators as well as small operatorsLarge operators would most likely not accept bearing the cost of unproven shortfalls just because an allegedly discredited software product states that there are shortfallsIs it possible that the Post Office has entered into secret unpublished agreements (under confidentiality agreements) with large operators that it would not do with smaller operators?
Basic first steps to check if computer problem …
Data is the key starting point - and being able to replicate the discrepancy/problem! Once you have this there can be no argument
Can someone explain why forensic accountants were used rather than computer specialists
Simple questions
What steps were taken to isolate the problem(s) at the time they occurred? In many cases these were not one-off incidents and all transactions for the branches in question should have been replicated into a test system so that they could be investigated
This fundamental approach would have provided copies of the actual offending data at the time these issues were raised and could be used for future testing to determine where any problems arose
In fact whilst Second Sight were conducting their 2 year audit did any PO branches encounter these problems? If they did then why did the auditors not insist on isolating the data from the offending branch (perhaps they did - so what were the results?) and running it through a test system to see if they could emulate the discrepancy
Once the test system can reproduce the problem then that is half the battle – all one subsequently needs to do is track down if/where the issues occur
Frankly any half competent computer person could have come up with a way of isolating the problem
After all it is only a glorified EPos system and a small PO will not be putting through thousands of transactions a day, although even if they did that should not be a problem