Post Office Creates CTO Role To Support 'Extensive and Complex' Plans (computerweekly.com) 10
The UK Post Office has appointed Paul Anastassi as interim CTO amid efforts to replace its controversial Horizon IT system, which led to hundreds of wrongful convictions of subpostmasters due to software errors since 1999.
The appointment, the news of which an anonymous reader shared, comes as the Post Office grapples with its $1.25 billion over-budget New Branch IT project, which was recently paused after being deemed "unachievable" in a government report. The organization is reportedly considering purchasing the Horizon system from Fujitsu while combining it with in-house and commercial software solutions.
The appointment, the news of which an anonymous reader shared, comes as the Post Office grapples with its $1.25 billion over-budget New Branch IT project, which was recently paused after being deemed "unachievable" in a government report. The organization is reportedly considering purchasing the Horizon system from Fujitsu while combining it with in-house and commercial software solutions.
they've obviously been reading (Score:2)
the output from the coloured pencils department.
you can't make this stuff up!
I can help ... (Score:1)
Re: what the hell actually went wrong? (Score:3, Interesting)
Re:what the hell actually went wrong? (Score:4, Interesting)
There were bugs in the system that lost transactions or incorrectly calculated things, compounded by the developers and PO staff trying to fix the issues by manually correcting balances.
Re:what the hell actually went wrong? (Score:5, Informative)
No, it wasn't "lost" transactions. it was duplicated transactions. If you accidentally hit "Enter" one too many times, it could result in a transaction being duplicated. If it stalls out when you hit "Enter", you might hit "Enter" again, and now you created two duplicate transactions. Or if it's hung, you might hit a few keys to see if it's truly stuck, and that can cause more transactions to be created.
The problem was, every transaction was incrementing how much money was supposed to be in the till at the end of the day. And because the system was buggy and slowdowns were common, you could easily be out thousands of UKP at the end of the day (i.e., the computer said you should have thousands more pounds in the register than you actually had).. With no obvious reason why it happened, the postmaster of each post office was accused to stealing that money, and in small towns, they were often ostracized by the public for theft. This lead to some committing suicide because they couldn't take the public accusing them of theft when none took place.
And this, is why even boring software can kill. You don't have to be designing a safety critical system where failure can lead to injury of death for it to occur. Even a regular normal accounting application can cause death. This should be the real lesson to every software developer out there. Even though your software may seem banal, it can kill. You might be writing the world's more boring word processor, but there can be instances where failure of such can lead to death.
Re: what the hell actually went wrong? (Score:2)
It's GBP by the way not UKP.
Re: what the hell actually went wrong? (Score:1)
Re: (Score:2)
When a transaction was entered by clicking on “send”, the screen froze, so the operator clicked “send” again. Now there were two transactions in the system for the same amount. At the end-of-day account reconciliation, there was a short fall. Sometimes the software people would remotely login to the Post Office terminal and adjust the numbers. Sometimes sub-postmasters would add their own money. Rather rick be
Re: (Score:2)
The problem was made much worse by the fact that the UK Post Office is allowed to bring its own criminal prosecutions. If they had to do what everyone else does - ask the police to investigate, and the police then have to convince the Crown Prosecution Service that there's a reasonable chance of getting a conviction - many of these cases would've been thrown out for lack of evidence.