[SystemSafety] Post Office Horizon System

[Olwen Morgan]

PBL is to be much thanked for posting this. ... A POS it was indeed!

Sadly, this is not the only example of people being convicted on the 
basis of unreliable computer-abstracted evidence. A couple of years ago, 
I attended a lecture - under Chatham House rules - that described 
similar weaknesses in a healthcare system. The lecturer is himself on 
this list but I will not presume to second-guess him as regards what may 
be publicly discussed about that system.

Thank God for a return to sanity after posting ping-pong re the 737 MAX.


Happidaze,

Olwen


On 16/12/2020 16:26, Peter Bernard Ladkin wrote:
> A couple of decades ago, the Post Office introduced a new WAN'd POS 
> called Horizon. Various of its franchisees in the early years were 
> accused of fraud; many were prosecuted and convicted; some were unable 
> to offer any plausible defence except to say "I didn't do it; really I 
> didn't; there must be some mistake".
>
> A class action Bates et al v Post Office was heard in 2019 before Mr. 
> Justice Fraser. Fraser J conducted a superb judicial-forensic 
> investigation from the bench, which determined more or less that the 
> Horizon system was in fact itself capable of much of the erroneous 
> behaviour which had been attributed to franchisee malfeasance in many, 
> possibly even most, of the criminal cases which had been prosecuted. 
> Subsequent to Fraser J's ruling, the Criminal Cases Review Commission 
> determined quickly that many of the convictions were unsafe.
>
> Some of the problem lies in how UK law treats "electronic"/digital 
> evidence, which includes evidence about the behaviour of digital 
> systems. There is a history to this. Barrister Paul Marshall recently 
> approached the Ministry of Justice about it. A paper with 
> recommendations on how the law should approach this kind of evidence 
> in the future was submitted on request to the Ministry in November 
> 2020 and a slightly modified version has been published in an open 
> access journal (indirect link below).
>
> Some of us on this list have been involved. I must say I am proud to 
> have been able to help. I am sure my colleagues feel the same way. Pro 
> bono publico. There is nothing more rewarding.
>
> https://ials.blogs.sas.ac.uk/2020/12/16/recommendations-for-the-probity-of-computer-evidence-requested-by-the-ministry-of-justice-now-published/ 
>
>
> PBL
>
> Prof. Peter Bernard Ladkin, Bielefeld, Germany
> ClaireTheWhiteRabbit RIP
> Tel+msg +49 (0)521 880 7319  www.rvs-bi.de
>
>
>
>
>
>
> _______________________________________________
> The System Safety Mailing List
> systemsafety at TechFak.Uni-Bielefeld.DE
> Manage your subscription: https://lists.techfak.uni-bielefeld.de/mailman/listinfo/systemsafety
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/pipermail/systemsafety/attachments/20201216/b0865eb0/attachment.html>