On 17/07/2017 11:25, Tim Watts wrote:
> On 17/07/17 09:15, Norman Wells wrote:
>> On 16/07/2017 21:41, Vir Campestris wrote:
>>> On 15/07/2017 23:33, Graham. wrote:
>>>> The logical progression of this is DNA taken at birth, anyone in
>>>> favour, after all, nothing to hide, nothing to fear.
>>>
>>> I fear false positives. Most jurors will not understand statistics.
>>
>> Nor will they need to. If statistics are involved, they will be
>> explained to them in terms that they can understand.
>>
>> But any systems in place should be robust enough to make false
>> positives almost impossible. It's in the development of such systems
>> that statisticians need to be involved.
>>
>> But it has to be remembered that DNA evidence is only relevant in a
>> very small percentage of crimes anyway, and may well help the defence
>> in most of those as it can eliminate the possibility that the accused
>> did it. It's far easier to show there is no match than to show the
>> existence of one. And if it helps identify the true perpetrator,
>> what's not to like?
>>
>> And in other non-criminal activities where proof of identity is
>> needed, what better than your DNA which is absolutely unique?
>
> 1) Identical twins have posed a problem until very recently as their DNA
> varies by a very subtle amount, mostly due to lifestyle related mutations.
Yes, they pose a particular problem, but they're not very common and
they're not a reason to throw the baby out with the bathwater.
> 2) Does not the standard DNA profiling only sample a subset of the DNA?
> Which leaves the possibility of two different samples having the same
> signature profile?
Yes, but the number of markers and their prevalence in humans is such
that a full match means only one in a billion, ie about 7 people on
earth. If you have any corroborative evidence at all, that clinches it
beyond any reasonable doubt.
> All of this is fine IF the judge and jurors understand the limits of the
> test.
That's what barristers are for. It's their job.