Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Home Office reports 27% of DNA matches have been made to more than one person.
252 views
Skip to first unread message
Great one
Oct 29, 2012, 3:17:20 AM10/29/12
to
the British Home
Office has reported that between 2001 and 2006, 27.6% of the matches reported from
searches of the UK National DNA Database (NDNAD) were to more than one person in
the database.
Office has reported that between 2001 and 2006, 27.6% of the matches reported from
searches of the UK National DNA Database (NDNAD) were to more than one person in
the database.
Great one
Oct 29, 2012, 3:21:21 AM10/29/12
to
Doc Insanity
Oct 29, 2012, 4:03:29 AM10/29/12
to
"Great one" wrote in message
news:ceb66daa-7e58-4ad0...@googlegroups.com...
Confirming a point I made a while ago! It just goes to show that the
barrack-room lawyers here also don't understand the application of
statistics either. No great surprise there!
Great one
Oct 29, 2012, 4:38:51 AM10/29/12
to
let me say that once again....1 in 3...so every 3 similar matches will yield 1 suspect...the bigger the data base the more jobs plod creates for himself trying to add legs to a coincidental match.
Great one
Oct 29, 2012, 4:42:00 AM10/29/12
to
Great one
Oct 29, 2012, 4:43:36 AM10/29/12
to
Doc Insanity
Oct 29, 2012, 5:15:44 AM10/29/12
to
"Great one" wrote in message
"John Doe"? statute of limitations?
Police forces don't perform forensic tests in this country.
Steerpike
Oct 29, 2012, 5:17:34 AM10/29/12
to
> Police forces known to be liars..now have a perfect foil, instead of a verbal they claim a 13 point loci match...when they have only got a partial..contaminated match.
Great one
Oct 29, 2012, 5:38:02 AM10/29/12
to
DocInsanity
Oct 29, 2012, 5:45:13 AM10/29/12
to
You're the fuckwit. The lab processes the sample, not the police - so how can the police realistically make claims for a sample that are not supported by the tests? What an idiot.
I think you need to have a psych work up.
Paul Nutteing
Oct 29, 2012, 5:44:24 AM10/29/12
to
Great one <tnh...@gmail.com> wrote in message
news:ceb66daa-7e58-4ad0...@googlegroups.com...
The Home Office would not divulge such as this.
(I am open to being proved wrong of course)
Not after leaning on my MP to make sure there is no reporting
of the lack of prosecutions from that Dutch CD of criminal DNA profiles
, from "matches" with the NDNAD. The FBI has a similar reporting lock down
on cross-state matches for the same reason.
The consequence of that CD fiasco ie all false matches, is that in a room of
12
people then probability greater than evens that someone in that room has a
10 locus, ie full NDNA DNA profile match, with an unrelated someone
else in the UK
What Special Branch don't want you to know about
DNA profiles
http://www.nutteing2.50megs.com/dnapr.htm
news:ceb66daa-7e58-4ad0...@googlegroups.com...
(I am open to being proved wrong of course)
Not after leaning on my MP to make sure there is no reporting
of the lack of prosecutions from that Dutch CD of criminal DNA profiles
, from "matches" with the NDNAD. The FBI has a similar reporting lock down
on cross-state matches for the same reason.
The consequence of that CD fiasco ie all false matches, is that in a room of
12
people then probability greater than evens that someone in that room has a
10 locus, ie full NDNA DNA profile match, with an unrelated someone
else in the UK
What Special Branch don't want you to know about
DNA profiles
http://www.nutteing2.50megs.com/dnapr.htm
Martin Brown
Oct 29, 2012, 7:21:38 AM10/29/12
to
On 29/10/2012 08:03, Doc Insanity wrote:
> "Great one" wrote in message
> news:ceb66daa-7e58-4ad0...@googlegroups.com...
>
> the British Home
> Office has reported that between 2001 and 2006, 27.6% of the matches
> reported from
> searches of the UK National DNA Database (NDNAD) were to more than one
> person in
> the database.
This seems to be the number submitted by GeneWatch to UK to parliament
> "Great one" wrote in message
> news:ceb66daa-7e58-4ad0...@googlegroups.com...
>
> the British Home
> Office has reported that between 2001 and 2006, 27.6% of the matches
> reported from
> searches of the UK National DNA Database (NDNAD) were to more than one
> person in
> the database.
in 2009 (so it would be worse by now). They do have an axe to grind...
http://www.publications.parliament.uk/pa/cm200910/cmselect/cmhaff/222/222we04.htm
(search for 27.6) p22 on their original submission document.
That is recent enough that it seems likely to be based on SGM Plus.
I think that they may be mis-stating the statistics to bolster their
case and that a fair proportion of these are actually partial match
searches on a limited number of loci with degraded DNA samples.
I would like confirmation that the ~30% multiple hit rate is a genuine
number provided by the Home Office and not one invented by GeneWatch.
Can anyone provide current unrelated persons adventitious match rates on
the existing 6M database (preferably for SGM Plus only)?
> ------------------------------------------------------------------
> Confirming a point I made a while ago! It just goes to show that the
> barrack-room lawyers here also don't understand the application of
> statistics either. No great surprise there!
infallible and pretending that DNA false positives or in lawyer speak
adventitious matches were all but unknown.
Iff the multiple match rate in the National DNAdatabase search really is
as high as 27.6% then the individual locus probabilities need to be
recalibrated using full Bayesian analysis and guidance to lawyers
rewritten to take account of these much higher systematic error rates.
The present situation is may be far *worse* than I had imagined. This
level of bogus matches goes way beyond the lower bound we established
from the published data on match probabilities. Worse by nearly two
orders of magnitude in fact. If it is true then this is *VERY* serious.
ISTR I found a 0.6% risk of a single false match in NDNAD assuming the
nominal FSS 1:10^9 match probability usually claimed per individual.
To find that in practice it is not unique at 28% on a 6M database means
that the nominal 1:10^9 match probability so often quoted by experts is
completely wrong for the UK population and a more realistic working
number based on these data is about 1:2x10^7.
I would like to see confirmation of the 27.6% multiple match claim from
Home Office sources, how they defined it and also the corresponding
figures for today.
This makes is even more unforgivable that the dozy bastards who locked
up that poor unfortunate rape suspect on the basis of DNA "evidence"
alone should have known that there was a 30% chance they were wrong!
He should be fully compensated and heads should roll!
--
Regards,
Martin Brown
DocInsanity
Oct 29, 2012, 7:27:36 AM10/29/12
to
Carry on living in cloud cuckoo land where you can believe that you understand DNA evidence by all means.
Big Les Wade
Oct 29, 2012, 10:33:38 AM10/29/12
to
Great one <tnh...@gmail.com> posted
>that between 2001 and 2006, 27.6% of the matches reported from
>searches of the UK National DNA Database (NDNAD) were to more than one
>person in
>the database.
>
--
Les
>the British Home
>Office has reported
Where?
>Office has reported
>that between 2001 and 2006, 27.6% of the matches reported from
>searches of the UK National DNA Database (NDNAD) were to more than one
>person in
>the database.
>
Les
Martin Brown
Oct 29, 2012, 10:41:34 AM10/29/12
to
On 29/10/2012 14:33, Big Les Wade wrote:
> Great one <tnh...@gmail.com> posted
>> the British Home
>> Office has reported
>
> Where?
I can't find where the Home Office have reported it at all.
> Great one <tnh...@gmail.com> posted
>> the British Home
>> Office has reported
>
> Where?
I did find it in the submission to Parliament by GeneWatch UK from 2009 at
http://www.publications.parliament.uk/pa/cm200910/cmselect/cmhaff/222/222we04.htm
(search for 27.6) p22 on their original submission document.
>
>> that between 2001 and 2006, 27.6% of the matches reported from
>> searches of the UK National DNA Database (NDNAD) were to more than one
>> person in
>> the database.
--
Regards,
>> searches of the UK National DNA Database (NDNAD) were to more than one
>> person in
>> the database.
--
Martin Brown
richardm...@googlemail.com
Oct 29, 2012, 10:43:04 AM10/29/12
to
Big Les Wade
Oct 29, 2012, 2:06:46 PM10/29/12
to
richardm...@googlemail.com posted
> Les posted:
>>>someone else posted:
>5EOYUZI.pdf&ei=4JWOUKCFOYWI0AWtiIDgBA&usg=AFQjCNGbraCeH75BrUPOwOLkJF8eto
>Y6RQ&sig2=N_SIgb8vqbt7C1ap8u9Bbw
Thanks. That document ("The Potential for Error in Forensic DNA Testing
and How That Complicates the Use of DNA Databases for Criminal
Identification" by William C. Thompson, Department of Criminology, Law &
Society, University of California, Irvine, August 12, 2008) states on
page 8: "the British Home Office has reported that between 2001 and
cases arose largely due to the significant proportion of crime scene
sample profiles that are partial.[13]"
and reference 13 is to "British Home Office, The National DNA Database
Annual Report 2005/6, p. 35,
http://www.homeoffice.gov.uk/documents/DNA-report2005-06.pdf "
However that URL does not seem to work any more.
--
Les
> Les posted:
>>>someone else posted:
>>the British Home
>>Office has reported
>
>>Where?
>
>>that between 2001 and 2006, 27.6% of the matches reported from
>>searches of the UK National DNA Database (NDNAD) were to more than one
>>person in
>>the database.
>http%3A%2F%2Fwww.councilforresponsiblegenetics.org%2FpageDocuments%2FH4T
>>Office has reported
>
>>Where?
>
>>that between 2001 and 2006, 27.6% of the matches reported from
>>searches of the UK National DNA Database (NDNAD) were to more than one
>>person in
>>the database.
>5EOYUZI.pdf&ei=4JWOUKCFOYWI0AWtiIDgBA&usg=AFQjCNGbraCeH75BrUPOwOLkJF8eto
>Y6RQ&sig2=N_SIgb8vqbt7C1ap8u9Bbw
Thanks. That document ("The Potential for Error in Forensic DNA Testing
and How That Complicates the Use of DNA Databases for Criminal
Identification" by William C. Thompson, Department of Criminology, Law &
Society, University of California, Irvine, August 12, 2008) states on
page 8: "the British Home Office has reported that between 2001 and
2006, 27.6% of the matches reported from
searches of the UK National DNA Database (NDNAD) were to more than one
person in the database. According to the report, the multiple-match
searches of the UK National DNA Database (NDNAD) were to more than one
cases arose largely due to the significant proportion of crime scene
sample profiles that are partial.[13]"
and reference 13 is to "British Home Office, The National DNA Database
Annual Report 2005/6, p. 35,
http://www.homeoffice.gov.uk/documents/DNA-report2005-06.pdf "
However that URL does not seem to work any more.
--
Les
Martin Brown
Oct 29, 2012, 3:51:49 PM10/29/12
to
However I did find this as a result of looking for it:
http://www.npia.police.uk/en/docs/NDNAD07-09-LR.pdf
You want page 28 which states since May 2001:
282,861 match groups
226,393 single subject match
56,468 two or more subjects (they don't explicitly show this number)
I get that to be about 20% multiple matches on trawling.
Still worryingly high! But they don't make clear how many of these are
due to partial match on degraded DNA. They are evasive on this point.
--
Regards,
Martin Brown
Norman Wells
Oct 29, 2012, 5:13:19 PM10/29/12
to
Martin Brown wrote:
> http://www.npia.police.uk/en/docs/NDNAD07-09-LR.pdf
>
> You want page 28 which states since May 2001:
> 282,861 match groups
> 226,393 single subject match
> 56,468 two or more subjects (they don't explicitly show this number)
>
> I get that to be about 20% multiple matches on trawling.
>
> Still worryingly high!
No it isn't. What it means is that in 80% of cases there is just a
> http://www.npia.police.uk/en/docs/NDNAD07-09-LR.pdf
>
> You want page 28 which states since May 2001:
> 282,861 match groups
> 226,393 single subject match
> 56,468 two or more subjects (they don't explicitly show this number)
>
> I get that to be about 20% multiple matches on trawling.
>
> Still worryingly high!
single match meaning everyone on the data-base apart from that one can
be eliminated from enquiries.
Of the 20% of cases where there is more than one match, assuming the
80:20 ratio is maintained, 80% of those (ie 16% of the total number of
cases) will have just two matches, meaning all but two on the data-base
can be eliminated from further enquiries.
You're looking at the results from the wrong direction. DNA testing is
not proof that someone did the deed, it's proof that the vast majority
didn't. To prove guilt, there _always_ has to be further evidence
besides a DNA match.
> But they don't make clear how many of these are
> due to partial match on degraded DNA. They are evasive on this point.
numbers not the methodology.
Paul Cummins
Oct 29, 2012, 8:11:00 PM10/29/12
to
We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
came up to me and whispered:
> DNA testing is
> not proof that someone did the deed, it's proof that the vast
> majority didn't. To prove guilt, there _always_ has to be
> further evidence besides a DNA match.
Fuck!
Norman said something profound and accurate.
He's obviously not a copper though.
--
Paul Cummins - Always a NetHead
Wasting Bandwidth since 1981
IF you think this http://bit.ly/u5EP3p is cruel
please sign this http://bit.ly/sKkzEx
---- If it's below this line, I didn't write it ----
came up to me and whispered:
> DNA testing is
> not proof that someone did the deed, it's proof that the vast
> majority didn't. To prove guilt, there _always_ has to be
> further evidence besides a DNA match.
Norman said something profound and accurate.
He's obviously not a copper though.
--
Paul Cummins - Always a NetHead
Wasting Bandwidth since 1981
IF you think this http://bit.ly/u5EP3p is cruel
please sign this http://bit.ly/sKkzEx
---- If it's below this line, I didn't write it ----
Great one
Oct 29, 2012, 8:43:52 PM10/29/12
to paul.c...@gstgroup.co.uk
On Tuesday, 30 October 2012 11:16:53 UTC+11, Paul Cummins wrote:
> We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
>
> came up to me and whispered:
>
>
>
> > DNA testing is
>
> > not proof that someone did the deed, it's proof that the vast
>
> > majority didn't. To prove guilt, there _always_ has to be
>
> > further evidence besides a DNA match.
>
>
>
> Fuck!
>
>
>
> Norman said something profound and accurate.
>
>
>
> He's obviously not a copper though.
>
>
>
> --
>
Not really..its a given that other evidence helps to convict, but, when the starting point is the DNA thread claiming 100% accuracy, other evidence is easily put aside. What Norman has missed is the contamination and degradation of any sample, plus the co incidental matching possibilities of samples, plus the frequency of similar matches within the sample size and the composition of each data base..and much more..at best Norman is a one legged pony.
> We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
>
> came up to me and whispered:
>
>
>
> > DNA testing is
>
> > not proof that someone did the deed, it's proof that the vast
>
> > majority didn't. To prove guilt, there _always_ has to be
>
> > further evidence besides a DNA match.
>
>
>
> Fuck!
>
>
>
> Norman said something profound and accurate.
>
>
>
> He's obviously not a copper though.
>
>
>
> --
>
nutherpe...@yahoo.co.uk
Oct 29, 2012, 9:55:10 PM10/29/12
to paul.c...@gstgroup.co.uk
On Tuesday, October 30, 2012 12:43:52 AM UTC, Great one wrote:
> On Tuesday, 30 October 2012 11:16:53 UTC+11, Paul Cummins wrote:
>
> > We were about to embark at Dover, when (Norman Wells)
> On Tuesday, 30 October 2012 11:16:53 UTC+11, Paul Cummins wrote:
>
>
> >
>
> > came up to me and whispered:
>
> >
>
> >
>
> >
>
> > > DNA testing is
>
> >
>
> > > not proof that someone did the deed, it's proof that the vast
>
> >
>
> > > majority didn't. To prove guilt, there _always_ has to be
>
> >
>
> > > further evidence besides a DNA match.
>
> >
>
> >
>
> >
>
> > Fuck!
>
> >
>
> >
>
> >
>
> > Norman said something profound and accurate.
>
> >
>
> >
>
> >
>
> > He's obviously not a copper though.
>
> >
>
> >
>
> >
>
> > --
>
> >
>
> Not really..its a given that other evidence helps to convict, but, when the starting point is the DNA thread claiming 100% accuracy, other evidence is easily put aside. What Norman has missed is the contamination and degradation of any sample, plus the co incidental matching possibilities of samples, plus the frequency of similar matches within the sample size and the composition of each data base..and much more..at best Norman is a one legged pony.
I find the co-incedence of convenience a much more believable scenario..
> >
>
> > came up to me and whispered:
>
> >
>
> >
>
> >
>
> > > DNA testing is
>
> >
>
> > > not proof that someone did the deed, it's proof that the vast
>
> >
>
> > > majority didn't. To prove guilt, there _always_ has to be
>
> >
>
> > > further evidence besides a DNA match.
>
> >
>
> >
>
> >
>
> > Fuck!
>
> >
>
> >
>
> >
>
> > Norman said something profound and accurate.
>
> >
>
> >
>
> >
>
> > He's obviously not a copper though.
>
> >
>
> >
>
> >
>
> > --
>
> >
>
> Not really..its a given that other evidence helps to convict, but, when the starting point is the DNA thread claiming 100% accuracy, other evidence is easily put aside. What Norman has missed is the contamination and degradation of any sample, plus the co incidental matching possibilities of samples, plus the frequency of similar matches within the sample size and the composition of each data base..and much more..at best Norman is a one legged pony.
.........
Great one
Oct 29, 2012, 11:58:28 PM10/29/12
to paul.c...@gstgroup.co.uk
nutherpe...@yahoo.co.uk
Oct 30, 2012, 1:06:13 AM10/30/12
to paul.c...@gstgroup.co.uk
On Tuesday, October 30, 2012 3:58:28 AM UTC, Great one wrote:
>
....
sold on the idea of a unique DNA signature for every individual..
I have tried to find answers to questions re possible familial
matches,, results in murder cases..
It seems that there are more possible results as one widens the
data base search...
Brothers sisters sons and daughters,,
are the only ones that ever lead an investigation to an arrest...
There are too many possibles if they try to widen the search
any farther,, to include cousins for example..
(my particular interest)
I look on the false conviction of Humble,,
supported by DNA if acctually not used in court,,
Plus
The posthumous conviction of chris smith,,
(Preston 75)
also, supposedly confirmed by DNA...
The stretch of imagination required
to believe that three entirely separate individuals
who just happened to have the same rare blood type B
was the "real" solution to Oldfields ripper case
is mindboggling..
And yet all tha nonsense was seriously reported,,
and never questioned...
Sutty's secret bum-chum in Sunderland
wot never got caught,,cetra cetra..
Now,, supposedly disproved by Humbles arrest
and the subsequent Chris Smith Posthumas confession
dna match..
Just as Sutty's arrest,, supposedly proved
the letters and tape from Sunderland were really a hoax..
All water under the bridge but not forgotten,,
DNA magic bullet comes along,,Whopee,, Humble Smyth...
Case closed,, closure at last spin..
And now were asked to swaller DNA errors on a national
data base,, just by some freakish co-incidence
just happened to identify suitable culprits
in the right areas,, with the right sort of background,,
in the right periods of time..
Give over will you..
Cops tell lies,,and the press report them..
DNA is nothing more than convenient persuasion,,
for those lacking integrity
Simple as that..
..............
>
....
> > > Not really..its a given that other evidence helps to convict, but, when the starting point is the DNA thread claiming 100% accuracy, other evidence is easily put aside. What Norman has missed is the contamination and degradation of any sample, plus the co incidental matching possibilities of samples, plus the frequency of similar matches within the sample size and the composition of each data base..and much more..at best Norman is a one legged pony.
>
> > I find the co-incidence of convenience a much more believable scenario..
>
>
> > .........
>
>
>
> What ever turns u on I suppose.....but the arguments for all flaws taken together when compared to one sample and one accused are much more profound, in debunking this crap.
I am not able to argue the science involved,,, but i have been
> > .........
>
>
>
> What ever turns u on I suppose.....but the arguments for all flaws taken together when compared to one sample and one accused are much more profound, in debunking this crap.
sold on the idea of a unique DNA signature for every individual..
I have tried to find answers to questions re possible familial
matches,, results in murder cases..
It seems that there are more possible results as one widens the
data base search...
Brothers sisters sons and daughters,,
are the only ones that ever lead an investigation to an arrest...
There are too many possibles if they try to widen the search
any farther,, to include cousins for example..
(my particular interest)
I look on the false conviction of Humble,,
supported by DNA if acctually not used in court,,
Plus
The posthumous conviction of chris smith,,
(Preston 75)
also, supposedly confirmed by DNA...
The stretch of imagination required
to believe that three entirely separate individuals
who just happened to have the same rare blood type B
was the "real" solution to Oldfields ripper case
is mindboggling..
And yet all tha nonsense was seriously reported,,
and never questioned...
Sutty's secret bum-chum in Sunderland
wot never got caught,,cetra cetra..
Now,, supposedly disproved by Humbles arrest
and the subsequent Chris Smith Posthumas confession
dna match..
Just as Sutty's arrest,, supposedly proved
the letters and tape from Sunderland were really a hoax..
All water under the bridge but not forgotten,,
DNA magic bullet comes along,,Whopee,, Humble Smyth...
Case closed,, closure at last spin..
And now were asked to swaller DNA errors on a national
data base,, just by some freakish co-incidence
just happened to identify suitable culprits
in the right areas,, with the right sort of background,,
in the right periods of time..
Give over will you..
Cops tell lies,,and the press report them..
DNA is nothing more than convenient persuasion,,
for those lacking integrity
Simple as that..
..............
nutherpe...@yahoo.co.uk
Oct 30, 2012, 1:13:57 AM10/30/12
to paul.c...@gstgroup.co.uk
> Cops tell lies,,and the press report them..
>
> DNA is nothing more than convenient persuasion,,
>
> for those lacking integrity
>
> Simple as that..
>
I'd like yo add a minor correction,,,
>
> DNA is nothing more than convenient persuasion,,
>
> for those lacking integrity
>
> Simple as that..
>
> Cops tell lies,,and the press report them..
>
>
> for those lacking integrity
>
> Simple as that..
..................
> for those lacking integrity
>
> Simple as that..
Many cases will of course be trustworthy,,,
The are not all telling lies..
.......................
Great one
Oct 30, 2012, 1:16:09 AM10/30/12
to paul.c...@gstgroup.co.uk
Now what you get is this.
Cops are trolling those data banks in the states and coming up with full profile matches in every state...now these full profile matches are for different people...not the same person using a different name, but a totally different person is being shown to be a perfect full profile match.
Now....you can repeat that for the UK cept they don't do it by county, they do in on the national DNA data base...now plod in both the states and UK have been caught out fabricating convictions based on a conviction that DNA doesn't lie.
nutherpe...@yahoo.co.uk
Oct 30, 2012, 2:13:31 AM10/30/12
to paul.c...@gstgroup.co.uk
On Tuesday, October 30, 2012 5:16:10 AM UTC, Nuther one wrote:
.....
right we agree there...
But
My understanding of DNA is that only identical twins share the same
full DNA signature..
I do not understand your assertions that many others may
share the same full dna matches..
................
.....
> > Cops tell lies,,and the press report them..
>
> >
>
> > DNA is nothing more than convenient persuasion,,
>
> >
>
> > for those lacking integrity
>
> >
>
> > Simple as that..
>
> >
>
> >
>
> You are missing the point...let me make this very simple for you....every state in the US is making a DNA data bank....you could say just like every county in the UK contributes to the national DNA data bank.
>
>
>
> Now what you get is this.
>
>
>
> Cops are trolling those data banks in the states and coming up with full profile matches in every state...now these full profile matches are for different people...not the same person using a different name, but a totally different person is being shown to be a perfect full profile match.
>
>
>
> Now....you can repeat that for the UK cept they don't do it by county, they do in on the national DNA data base...now plod in both the states and UK have been caught out fabricating convictions based on a conviction that DNA doesn't lie.
Plod caught out fabricating DNA evidence to enable convictions,,
>
> >
>
> > DNA is nothing more than convenient persuasion,,
>
> >
>
> > for those lacking integrity
>
> >
>
> > Simple as that..
>
> >
>
> >
>
> You are missing the point...let me make this very simple for you....every state in the US is making a DNA data bank....you could say just like every county in the UK contributes to the national DNA data bank.
>
>
>
> Now what you get is this.
>
>
>
> Cops are trolling those data banks in the states and coming up with full profile matches in every state...now these full profile matches are for different people...not the same person using a different name, but a totally different person is being shown to be a perfect full profile match.
>
>
>
> Now....you can repeat that for the UK cept they don't do it by county, they do in on the national DNA data base...now plod in both the states and UK have been caught out fabricating convictions based on a conviction that DNA doesn't lie.
right we agree there...
But
My understanding of DNA is that only identical twins share the same
full DNA signature..
I do not understand your assertions that many others may
share the same full dna matches..
................
Great one
Oct 30, 2012, 4:13:20 AM10/30/12
to paul.c...@gstgroup.co.uk
Explaining Cold Hits to a Jury
People have been prosecuted based on cold hits to partial profiles. Earlier this
year, for example, a man named John Puckett was convicted of murdering a San
Francisco woman in 1972.[26] The key evidence against him (uncovered in a recent
search of the state offender database) was that his DNA profile matched a partial evidentiary profile found on the dead victim. The estimated frequency of the matching characteristics was 1 in 1.1 million, hence one would expect to find more than 250 peoplein the United States with a matching profile. In the absence of other persuasive evidenceof the defendant’s guilt, the probability of a false incrimination in a case like this wouldappear to be high. According to press accounts, the only evidence of Puckett’s guilt wasthe DNA match, the fact that he had lived in San Francisco and been old enough tocommit the crime in 1972, and the fact that he had pleaded guilty to two rapes and a
sexual assault in 1977 (although he had been free and had a clean record since
1988).[27,28]
There has been considerable debate over the manner in which DNA evidence
should be presented to juries in cold hit cases. Two reports of the National ResearchCouncil took the position that the random match probabilities presented in connection15with a cold hit should be adjusted to take into account the greater likelihood of finding acoincidental match when searching a database. A 1992 NRC report suggested that once acold hit is obtained the government test additional genetic loci and then, if they match,present statistics based only on these new matching loci.[29] A 1996 NRC reportsuggested presenting what it called a “database match probability,” which reflects theprobability of finding a match, by coincidence in the database.[30] The database matchprobability is determined by multiplying the estimated frequency of the matching DNAprofile by the number of different samples in the database. In the Puckett case the coldhit occurred during the search of a California offender database, which contained 338,000 samples, hence the database match probability was approximately 1 in 3.[28]
Hyde of course is well aware of partial matches being the same as a 1 in 3 shot so UK plod routinely say they have a full match and lie there heads of with other supporting evidence
Norman Wells
Oct 30, 2012, 5:06:01 AM10/30/12
to
Great one wrote:
> On Tuesday, 30 October 2012 11:16:53 UTC+11, Paul Cummins wrote:
>> We were about to embark at Dover, when h...@unseen.ac.am (Norman
>> Wells)
>> came up to me and whispered:
>>
>>> DNA testing is
>>> not proof that someone did the deed, it's proof that the vast
>>> majority didn't. To prove guilt, there _always_ has to be
>>> further evidence besides a DNA match.
>> Fuck!
>> Norman said something profound and accurate.
>> He's obviously not a copper though.
> On Tuesday, 30 October 2012 11:16:53 UTC+11, Paul Cummins wrote:
>> We were about to embark at Dover, when h...@unseen.ac.am (Norman
>> Wells)
>> came up to me and whispered:
>>
>>> DNA testing is
>>> not proof that someone did the deed, it's proof that the vast
>>> majority didn't. To prove guilt, there _always_ has to be
>>> further evidence besides a DNA match.
>> Fuck!
>> Norman said something profound and accurate.
>> He's obviously not a copper though.
> Not really..its a given that other evidence helps to convict,
I didn't say it helps to convict, I said it's absolutely necessary.
> but,
> when the starting point is the DNA thread claiming 100% accuracy,
> other evidence is easily put aside. What Norman has missed is the
> contamination and degradation of any sample, plus the co incidental
> matching possibilities of samples, plus the frequency of similar
> matches within the sample size and the composition of each data
> base..and much more..at best Norman is a one legged pony.
vast majority of the population from any consideration whatsoever.
Against the few that are left, for any conviction to result, there
always has to be sufficient further evidence.
All DNA analysis does is reduce the number of _possible_ suspects down
to a manageable number, probably in single figures. Most of those could
not reasonably have committed the crime for one reason or another, eg
age or distance from the event. That cuts the number down even further,
probably down to one, or at most two. Then it becomes a matter for all
the other evidence that there might be, including identification,
alibis, circumstances, credibility etc etc. Guilt will _never_ be
established here unless there is compelling additional evidence linking
the accused to the crime.
Big Les Wade
Oct 30, 2012, 5:02:04 AM10/30/12
to
Paul Cummins <uset...@stedtelephone.invalid> posted
just about any other fact as further evidence against the suspect. The
fact that he was an adult male and lived in England at the time of the
crime will do, if nothing else emerges.
--
Les
>We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
>came up to me and whispered:
>
>> DNA testing is
>> not proof that someone did the deed, it's proof that the vast
>> majority didn't. To prove guilt, there _always_ has to be
>> further evidence besides a DNA match.
>
>Fuck!
>
>Norman said something profound and accurate.
Not really. Once the police have a "DNA match", then they will regard
>came up to me and whispered:
>
>> DNA testing is
>> not proof that someone did the deed, it's proof that the vast
>> majority didn't. To prove guilt, there _always_ has to be
>> further evidence besides a DNA match.
>
>Fuck!
>
>Norman said something profound and accurate.
just about any other fact as further evidence against the suspect. The
fact that he was an adult male and lived in England at the time of the
crime will do, if nothing else emerges.
--
Les
Message has been deleted
Martin Brown
Oct 30, 2012, 5:25:57 AM10/30/12
to
On 29/10/2012 21:13, Norman Wells wrote:
> Martin Brown wrote:
>
>> http://www.npia.police.uk/en/docs/NDNAD07-09-LR.pdf
>>
>> You want page 28 which states since May 2001:
>> 282,861 match groups
>> 226,393 single subject match
>> 56,468 two or more subjects (they don't explicitly show this number)
>>
>> I get that to be about 20% multiple matches on trawling.
>>
>> Still worryingly high!
>
> No it isn't. What it means is that in 80% of cases there is just a
> single match meaning everyone on the data-base apart from that one can
> be eliminated from enquiries.
So far so good.
> Martin Brown wrote:
>
>> http://www.npia.police.uk/en/docs/NDNAD07-09-LR.pdf
>>
>> You want page 28 which states since May 2001:
>> 282,861 match groups
>> 226,393 single subject match
>> 56,468 two or more subjects (they don't explicitly show this number)
>>
>> I get that to be about 20% multiple matches on trawling.
>>
>> Still worryingly high!
>
> No it isn't. What it means is that in 80% of cases there is just a
> single match meaning everyone on the data-base apart from that one can
> be eliminated from enquiries.
>
> Of the 20% of cases where there is more than one match, assuming the
> 80:20 ratio is maintained, 80% of those (ie 16% of the total number of
> cases) will have just two matches, meaning all but two on the data-base
> can be eliminated from further enquiries.
Also true. And it is quite possible that one of them did it.
> Of the 20% of cases where there is more than one match, assuming the
> 80:20 ratio is maintained, 80% of those (ie 16% of the total number of
> cases) will have just two matches, meaning all but two on the data-base
> can be eliminated from further enquiries.
>
> You're looking at the results from the wrong direction. DNA testing is
> not proof that someone did the deed, it's proof that the vast majority
> didn't. To prove guilt, there _always_ has to be further evidence
> besides a DNA match.
Yeah right! You mean like the now infamous case of the unfortunate Adam
> You're looking at the results from the wrong direction. DNA testing is
> not proof that someone did the deed, it's proof that the vast majority
> didn't. To prove guilt, there _always_ has to be further evidence
> besides a DNA match.
Scott where he was banged up for 5 months on the basis of faulty DNA
sample handling at LGC and a trawl of the NDNAD that matched. This was
actually a correct match in the sense that the blistering incompetence
at LGC allowed his DNA to contaminate a rape victims sample.
http://www.homeoffice.gov.uk/publications/agencies-public-bodies/fsr/dna-contam-report?view=Binary
His biggest problem was that his DNA sample had been taken and that he
could not prove his whereabouts on the day of the crime (although his
mobile phone was clearly in Bristol). See 6.10 p10 & weep.
Where were *YOU* on 27th April 2011 9pm? Can't answer? - lock him up.
It beggars belief that LGC knew at the time of running the samples that
they *had* a contamination problem but still presented these data to
Manchester Police as strong statistical "evidence". In fact we now know
that it was complete bollocks and yet Mr Scott has no legal redress
against the incompetents that locked him up for something he did not do.
This could just as easily have happened to an innocent man with an
adventitious match on the now large database instead of a lab cock-up.
That has also happened before and in a rape case conviction.
In this case they did eventually figure out their mistake but the guy
doesn't get his life back together so easily.
>> But they don't make clear how many of these are
>> due to partial match on degraded DNA. They are evasive on this point.
>
> It doesn't matter. The conclusions I've outlined above result from the
> numbers not the methodology.
reported then it calls into question the 1:10^9 figure that is so often
quoted by expert witnesses as the individual match probability. It could
actually be as low as 1:10^6 though more likely is it around 2x10^7
(which is my best estimate from the limited published data).
It would be necessary to eliminate familiar DNA partial matches and
degraded DNA trawls with high multiple matches to get a correct
realistic value for the empirical statistical probability of a match.
I used to think that NDNAD matching was trustworthy and had operational
integrity. But now I will *never* volunteer to provide a DNA sample.
Don't get me wrong. It is a very useful tool and most of the time it
gives the right answer, but the public need to be protected from and
compensated for the times when police, CPS and forensic incompetence
results in a serious travesty of justice.
And I count being innocent and held in police custody for an extended
period of time as being almost as bad as being incorrectly found guilty.
--
Regards,
Martin Brown
Norman Wells
Oct 30, 2012, 5:28:41 AM10/30/12
to
Big Les Wade
Oct 30, 2012, 5:34:41 AM10/30/12
to
Great one <tnh...@gmail.com> posted
well have been the decision of the trial judge, over the objection of
defense counsel, to exclude any mention of the fact that the DNA match
was discovered in a database search.[33] Jurors were told about the DNA
match, but not that it arose from a cold hit. The judge excluded
testimony by an expert called by the defendant to explain the
statistical implications of a cold hit. The judge also excluded any
mention of a database match probability. During their deliberations, the
jurors sent a note to the judge asking for information on how Puckett
had been identified as a suspect. The judge told the jury that that
information was irrelevant to their decision and should not be
considered.[28,33]"
The lesson here is that lawyers are too stupid (and perhaps, depending
on their personal role in the criminal justice system, too determined to
obtain a conviction) to be allowed to consider DNA evidence. Quite
simply, the Puckett judge should have been prosecuted for perverting the
course of justice, and imprisoned for many years. So should the
prosecuting counsel, who presumably encouraged him in his stupidity.
--
Les
>Explaining Cold Hits to a Jury
>People have been prosecuted based on cold hits to partial profiles.
>Earlier this
>year, for example, a man named John Puckett was convicted of murdering a San
>Francisco woman in 1972.[26] The key evidence against him (uncovered in
>a recent
>search of the state offender database) was that his DNA profile matched
>a partial evidentiary profile found on the dead victim. The estimated
>frequency of the matching characteristics was 1 in 1.1 million, hence
>one would expect to find more than 250 peoplein the United States with
>a matching profile. In the absence of other persuasive evidenceof the
>defendant’s guilt, the probability of a false incrimination in a case
>like this wouldappear to be high. According to press accounts, the only
>evidence of Puckett’s guilt wasthe DNA match, the fact that he had
>lived in San Francisco and been old enough tocommit the crime in 1972,
>and the fact that he had pleaded guilty to two rapes and a
>sexual assault in 1977 (although he had been free and had a clean record since
>1988).[27,28]
It goes on ... "A key factor in the conviction of John Puckett might
>People have been prosecuted based on cold hits to partial profiles.
>Earlier this
>year, for example, a man named John Puckett was convicted of murdering a San
>Francisco woman in 1972.[26] The key evidence against him (uncovered in
>a recent
>search of the state offender database) was that his DNA profile matched
>a partial evidentiary profile found on the dead victim. The estimated
>frequency of the matching characteristics was 1 in 1.1 million, hence
>one would expect to find more than 250 peoplein the United States with
>a matching profile. In the absence of other persuasive evidenceof the
>defendant’s guilt, the probability of a false incrimination in a case
>like this wouldappear to be high. According to press accounts, the only
>evidence of Puckett’s guilt wasthe DNA match, the fact that he had
>lived in San Francisco and been old enough tocommit the crime in 1972,
>and the fact that he had pleaded guilty to two rapes and a
>sexual assault in 1977 (although he had been free and had a clean record since
>1988).[27,28]
well have been the decision of the trial judge, over the objection of
defense counsel, to exclude any mention of the fact that the DNA match
was discovered in a database search.[33] Jurors were told about the DNA
match, but not that it arose from a cold hit. The judge excluded
testimony by an expert called by the defendant to explain the
statistical implications of a cold hit. The judge also excluded any
mention of a database match probability. During their deliberations, the
jurors sent a note to the judge asking for information on how Puckett
had been identified as a suspect. The judge told the jury that that
information was irrelevant to their decision and should not be
considered.[28,33]"
The lesson here is that lawyers are too stupid (and perhaps, depending
on their personal role in the criminal justice system, too determined to
obtain a conviction) to be allowed to consider DNA evidence. Quite
simply, the Puckett judge should have been prosecuted for perverting the
course of justice, and imprisoned for many years. So should the
prosecuting counsel, who presumably encouraged him in his stupidity.
--
Les
Big Les Wade
Oct 30, 2012, 5:41:02 AM10/30/12
to
Kim Bolton <nos...@all.invalid> posted
>or you don't. You can't have a partial sample.
Yes you can, see p4 of the Williams paper: "Evidentiary samples from
crime scenes often produce incomplete or partial DNA profiles. Limited
quantities of DNA, degradation of the sample, or the presence of
inhibitors (contaminants) can make it impossible to determine the
genotype at every locus. In some instances the test yields no
information about the genotype at a particular locus; in some instances
one of the two alleles at a locus will “drop out” (become
undetectable). "
>
>You could, however, have a partial match (not the same thing) due to
>contamination or multiple profiles - perhps this is what they are
>getting at.
>
No, that's a separate issue, and is yet another source of error.
--
Les
>
>Big Les Wade wrote:
>
>>"The Potential for Error in Forensic DNA Testing
>>and How That Complicates the Use of DNA Databases for Criminal
>>Identification" by William C. Thompson, Department of Criminology, Law &
>>Society, University of California, Irvine, August 12, 2008 states on
>Big Les Wade wrote:
>
>>"The Potential for Error in Forensic DNA Testing
>>and How That Complicates the Use of DNA Databases for Criminal
>>Identification" by William C. Thompson, Department of Criminology, Law &
>>page 8: "the British Home Office has reported that between 2001 and
>>2006, 27.6% of the matches reported from
>>searches of the UK National DNA Database (NDNAD) were to more than one
>>person in the database. According to the report, the multiple-match
>>cases arose largely due to the significant proportion of crime scene
>>sample profiles that are partial.[13]"
>
>This seems to be somewhat disingenuous. You either have a DNA sample,
>>2006, 27.6% of the matches reported from
>>searches of the UK National DNA Database (NDNAD) were to more than one
>>person in the database. According to the report, the multiple-match
>>cases arose largely due to the significant proportion of crime scene
>>sample profiles that are partial.[13]"
>
>or you don't. You can't have a partial sample.
Yes you can, see p4 of the Williams paper: "Evidentiary samples from
crime scenes often produce incomplete or partial DNA profiles. Limited
quantities of DNA, degradation of the sample, or the presence of
inhibitors (contaminants) can make it impossible to determine the
genotype at every locus. In some instances the test yields no
information about the genotype at a particular locus; in some instances
one of the two alleles at a locus will “drop out” (become
undetectable). "
>
>You could, however, have a partial match (not the same thing) due to
>contamination or multiple profiles - perhps this is what they are
>getting at.
>
No, that's a separate issue, and is yet another source of error.
--
Les
Norman Wells
Oct 30, 2012, 5:47:23 AM10/30/12
to
always be with us. That's an unfortunate fact of life, and they occur
in all areas. But it's a red herring. We're talking about cases where
there is in fact a proper DNA match.
>>> But they don't make clear how many of these are
>>> due to partial match on degraded DNA. They are evasive on this
>>> point.
>>
>> It doesn't matter. The conclusions I've outlined above result from
>> the numbers not the methodology.
>
> Yes. It does. If these adventitious positives are as high as are being
> reported then it calls into question the 1:10^9 figure that is so
> often quoted by expert witnesses as the individual match probability.
> It could actually be as low as 1:10^6 though more likely is it around
> 2x10^7 (which is my best estimate from the limited published data).
if it is as low as 1 in a million, which most would dispute as being far
too high, that means just 60 people in the country could possibly have
committed the crime. If they're not too young, or too old, or the wrong
sex, most will have easily verifiable alibis meaning they could not
possibly have been at the scene of the crime at the time it was
committed. Most will in fact live a considerable distance from it, and
there will be absolutely no other evidence connecting them in any way
with it.
The number of actual suspects will therefore rapidly fall to one or at
most two. Further conventional evidence, I suggest, will rapidly reduce
it, in the vast majority of cases, to one.
> It would be necessary to eliminate familiar DNA partial matches and
> degraded DNA trawls with high multiple matches to get a correct
> realistic value for the empirical statistical probability of a match.
>
> I used to think that NDNAD matching was trustworthy and had
> operational integrity. But now I will *never* volunteer to provide a
> DNA sample.
> Don't get me wrong. It is a very useful tool and most of the time it
> gives the right answer, but the public need to be protected from and
> compensated for the times when police, CPS and forensic incompetence
> results in a serious travesty of justice.
> And I count being innocent and held in police custody for an extended
> period of time as being almost as bad as being incorrectly found
> guilty.
Norman Wells
Oct 30, 2012, 5:49:21 AM10/30/12
to
appeal would have been successful. It's how it works.
Martin Brown
Oct 30, 2012, 6:43:36 AM10/30/12
to
On 30/10/2012 09:15, Kim Bolton wrote:
exact-1, exact-2, exact-3, exact-M etc and number of matches for a few
hundred representative targets of different ethnicities this would go a
long way to providing some reassurance that the NDNAD is not being
abused. Having read the technical reports I worry that DNA evidence is
being presented as "better" than it really is in too many cases.
The degenerate match multiplicities go up alarmingly from exact:
1, 20, 190, 1140, 4845, 15504 for M=0,1,2,3,4,5
We now have one very high profile example where not only was the DNA
database match abused, but dodgy contaminated DNA "evidence" was used on
its *own* to lock someone up for an extended period.
The DNA evidence in the Adams case was a 17 of 20 allele match on one
sample, no match at all on a second sample, and a match on the control
blank! This should have rung alarm bells at the outset but it didn't :(
--
Regards,
Martin Brown
>
> Big Les Wade wrote:
>
>> "The Potential for Error in Forensic DNA Testing
>> and How That Complicates the Use of DNA Databases for Criminal
>> Identification" by William C. Thompson, Department of Criminology, Law &
>> Society, University of California, Irvine, August 12, 2008 states on
> Big Les Wade wrote:
>
>> "The Potential for Error in Forensic DNA Testing
>> and How That Complicates the Use of DNA Databases for Criminal
>> Identification" by William C. Thompson, Department of Criminology, Law &
>> page 8: "the British Home Office has reported that between 2001 and
>> 2006, 27.6% of the matches reported from
>> searches of the UK National DNA Database (NDNAD) were to more than one
>> person in the database. According to the report, the multiple-match
>> cases arose largely due to the significant proportion of crime scene
>> sample profiles that are partial.[13]"
>
>> 2006, 27.6% of the matches reported from
>> searches of the UK National DNA Database (NDNAD) were to more than one
>> person in the database. According to the report, the multiple-match
>> cases arose largely due to the significant proportion of crime scene
>> sample profiles that are partial.[13]"
>
> This seems to be somewhat disingenuous. You either have a DNA sample,
> or you don't. You can't have a partial sample.
>
> or you don't. You can't have a partial sample.
>
> You could, however, have a partial match (not the same thing) due to
> contamination or multiple profiles - perhps this is what they are
> getting at.
If they published the allele match statistics working back from exact,
> contamination or multiple profiles - perhps this is what they are
> getting at.
exact-1, exact-2, exact-3, exact-M etc and number of matches for a few
hundred representative targets of different ethnicities this would go a
long way to providing some reassurance that the NDNAD is not being
abused. Having read the technical reports I worry that DNA evidence is
being presented as "better" than it really is in too many cases.
The degenerate match multiplicities go up alarmingly from exact:
1, 20, 190, 1140, 4845, 15504 for M=0,1,2,3,4,5
We now have one very high profile example where not only was the DNA
database match abused, but dodgy contaminated DNA "evidence" was used on
its *own* to lock someone up for an extended period.
The DNA evidence in the Adams case was a 17 of 20 allele match on one
sample, no match at all on a second sample, and a match on the control
blank! This should have rung alarm bells at the outset but it didn't :(
--
Regards,
Martin Brown
Paul Nutteing
Oct 30, 2012, 7:51:03 AM10/30/12
to
Martin Brown <|||newspam|||@nezumi.demon.co.uk> wrote in message
news:r7Bjs.1610$Uu4...@newsfe12.iad...
archived on
http://web.archive.org/web/20081124203539/http://www.homeoffice.gov.uk/docum
ents/DNA-report2005-06.pdf
Also pertenant to this discussion, but not referred to in those reports
the error rate for reference (buccal cell) and crime scene DNA profiles
http://gednap.forensischegenetik.de/Information/IJLM_GEDNAP_II.pdf
Results for Gednap 20 & 21 which show 0.7 per cent on 13,868 aggregated
single
loci tests gives
an error rate of 1 in 11 expressed for CODIS profiles, 1 in 14 for UK FSS 10
loci profiles.
(for worst case that only one locus per profile was in error , "best" case
would
be if some labs returned totally erroneous results ie all loci wrong)
As blind and not double blind trials, unlikely any fully wrong results would
have been returned to the collators.
They were only able to improve the error rates by moving the simulated
sample-types further away from the sort of stains found at crime-scenes.
So you have the situation of wrongly recorded NDNAD reference profiles
falsely matching true crime scene full profiles AND also erroneous crime
scene profiles
falsely matching correct full reference profiles.
What Special Branch don't want you to know about
DNA profiles
http://www.nutteing2.50megs.com/dnapr.htm
news:r7Bjs.1610$Uu4...@newsfe12.iad...
http://web.archive.org/web/20081124203539/http://www.homeoffice.gov.uk/docum
ents/DNA-report2005-06.pdf
Also pertenant to this discussion, but not referred to in those reports
the error rate for reference (buccal cell) and crime scene DNA profiles
http://gednap.forensischegenetik.de/Information/IJLM_GEDNAP_II.pdf
Results for Gednap 20 & 21 which show 0.7 per cent on 13,868 aggregated
single
loci tests gives
an error rate of 1 in 11 expressed for CODIS profiles, 1 in 14 for UK FSS 10
loci profiles.
(for worst case that only one locus per profile was in error , "best" case
would
be if some labs returned totally erroneous results ie all loci wrong)
As blind and not double blind trials, unlikely any fully wrong results would
have been returned to the collators.
They were only able to improve the error rates by moving the simulated
sample-types further away from the sort of stains found at crime-scenes.
So you have the situation of wrongly recorded NDNAD reference profiles
falsely matching true crime scene full profiles AND also erroneous crime
scene profiles
falsely matching correct full reference profiles.
What Special Branch don't want you to know about
DNA profiles
http://www.nutteing2.50megs.com/dnapr.htm
Martin Brown
Oct 30, 2012, 8:05:24 AM10/30/12
to
> archived on
> http://web.archive.org/web/20081124203539/http://www.homeoffice.gov.uk/docum
> ents/DNA-report2005-06.pdf
>
Thanks. But when I try that site I got/get the following hard error msg:
> http://web.archive.org/web/20081124203539/http://www.homeoffice.gov.uk/docum
> ents/DNA-report2005-06.pdf
>
Wayback Exception
An unknown exception has occurred. Read timed out :
http://webarchive.nationalarchives.gov.uk/+/http://www.homeoffice.gov.uk/documents/DNA-report2005-06.pdf?view=Binary
--
Regards,
Martin Brown
Big Les Wade
Oct 30, 2012, 8:20:06 AM10/30/12
to
Martin Brown <|||newspam|||@nezumi.demon.co.uk> posted
--
Les
>Thanks. But when I try that site I got/get the following hard error
>msg:
>
>Wayback Exception
>
>An unknown exception has occurred. Read timed out :
>http://webarchive.nationalarchives.gov.uk/+/http://www.homeoffice.gov.uk
>/documents/DNA-report2005-06.pdf?view=Binary
It did work for me, though I didn't expect it to :)
>msg:
>
>Wayback Exception
>
>An unknown exception has occurred. Read timed out :
>http://webarchive.nationalarchives.gov.uk/+/http://www.homeoffice.gov.uk
>/documents/DNA-report2005-06.pdf?view=Binary
--
Les
Martin Brown
Oct 30, 2012, 8:32:58 AM10/30/12
to
On 30/10/2012 09:47, Norman Wells wrote:
> Martin Brown wrote:
>> On 29/10/2012 21:13, Norman Wells wrote:
>>> You're looking at the results from the wrong direction. DNA testing
>>> is not proof that someone did the deed, it's proof that the vast
>>> majority didn't. To prove guilt, there _always_ has to be further
I agree with what you say here, but the police usage of NDNAD does not
> Martin Brown wrote:
>> On 29/10/2012 21:13, Norman Wells wrote:
>>> You're looking at the results from the wrong direction. DNA testing
>>> is not proof that someone did the deed, it's proof that the vast
>>> majority didn't. To prove guilt, there _always_ has to be further
bear this out - indeed CPS have guidelines that look pretty iffy wrt
uncorroborated DNA evidence. How else do you explain this rape case?...
DNA match. I take it to mean a good match on at least 17 out of 20
alleles as happened in this particularly bad case.
Enough that the evidence was presented as statistically "strong". When
in fact it was incredibly weak and due to systematic error in the lab.
You cannot know after a database trawl that you have the right person.
Only that their DNA appears to match the sample well enough for them to
be worthy of further consideration as a suspect.
If it was a safe breaking or burglary and the profile matched the right
sort of local suspect I would certainly give that considerable weight.
But what if the match is for a bloke two hundred miles away guilty only
of some minor misdemeanor but unlucky enough to share a 17 of 20 allele
match with the real criminal. The size of the database is such now that
these sorts of bogus adventitious matches are increasingly likely.
Matches generated by PLS against NDNAD are even more iffy since linear
combinations of more than one DNA profile provides additional degrees of
freedom to generate entirely spurious pseudo matches.
>>>> But they don't make clear how many of these are
>>>> due to partial match on degraded DNA. They are evasive on this
>>>> point.
>>>
>>> It doesn't matter. The conclusions I've outlined above result from
>>> the numbers not the methodology.
>>
>> Yes. It does. If these adventitious positives are as high as are being
>> reported then it calls into question the 1:10^9 figure that is so
>> often quoted by expert witnesses as the individual match probability.
>> It could actually be as low as 1:10^6 though more likely is it around
>> 2x10^7 (which is my best estimate from the limited published data).
>
> Any of which eliminates the vast majority of people at that stage. Even
> if it is as low as 1 in a million, which most would dispute as being far
> too high, that means just 60 people in the country could possibly have
> committed the crime. If they're not too young, or too old, or the wrong
> sex, most will have easily verifiable alibis meaning they could not
> possibly have been at the scene of the crime at the time it was
> committed. Most will in fact live a considerable distance from it, and
> there will be absolutely no other evidence connecting them in any way
> with it.
remember where they were at a particular date is not the way to proceed.
> The number of actual suspects will therefore rapidly fall to one or at
> most two. Further conventional evidence, I suggest, will rapidly reduce
> it, in the vast majority of cases, to one.
"We have the DNA evidence we know you are guilty why not confess?"
I don't understand why his defence did not demand a re-examination of
the forensic evidence early on though. Can someone explain?
>
>> It would be necessary to eliminate familiar DNA partial matches and
>> degraded DNA trawls with high multiple matches to get a correct
>> realistic value for the empirical statistical probability of a match.
>>
>> I used to think that NDNAD matching was trustworthy and had
>> operational integrity. But now I will *never* volunteer to provide a
>> DNA sample.
>> Don't get me wrong. It is a very useful tool and most of the time it
>> gives the right answer, but the public need to be protected from and
>> compensated for the times when police, CPS and forensic incompetence
>> results in a serious travesty of justice.
>
> I don't think there's anyone here who would disagree.
I can think of at least one Insane Doc that would.
>> It would be necessary to eliminate familiar DNA partial matches and
>> degraded DNA trawls with high multiple matches to get a correct
>> realistic value for the empirical statistical probability of a match.
>>
>> I used to think that NDNAD matching was trustworthy and had
>> operational integrity. But now I will *never* volunteer to provide a
>> DNA sample.
>> Don't get me wrong. It is a very useful tool and most of the time it
>> gives the right answer, but the public need to be protected from and
>> compensated for the times when police, CPS and forensic incompetence
>> results in a serious travesty of justice.
>
> I don't think there's anyone here who would disagree.
>
>> And I count being innocent and held in police custody for an extended
>> period of time as being almost as bad as being incorrectly found
>> guilty.
>
> No it isn't. But of course it shouldn't happen unreasonably.
It did though and it appears that he has no legal redress.
>> And I count being innocent and held in police custody for an extended
>> period of time as being almost as bad as being incorrectly found
>> guilty.
>
> No it isn't. But of course it shouldn't happen unreasonably.
--
Regards,
Martin Brown
richardm...@googlemail.com
Oct 30, 2012, 9:39:05 AM10/30/12
to
As my knowledge of DNA is very limited
and to help me understand the issue
please correct the following:
If we take the DNA profile from
http://dna-view.com/profile.htm
It suggests in their example that out
a database of 428-432 there is a
1/7000 that another persons left
the DNA.
If in the above example there was
27% DNA matches would this not
suggest that the database is biased
or that the biology is incorrect. i.e.
if a 6 sided die you would expect a 1/6
chance of a 2 or a 3 etc where as in fact
it may infact be a loaded die and thus a
propensity towards one or more number.
and to help me understand the issue
please correct the following:
If we take the DNA profile from
http://dna-view.com/profile.htm
It suggests in their example that out
a database of 428-432 there is a
1/7000 that another persons left
the DNA.
If in the above example there was
27% DNA matches would this not
suggest that the database is biased
or that the biology is incorrect. i.e.
if a 6 sided die you would expect a 1/6
chance of a 2 or a 3 etc where as in fact
it may infact be a loaded die and thus a
propensity towards one or more number.
Cynic
Oct 30, 2012, 10:07:32 AM10/30/12
to
On Tue, 30 Oct 2012 09:06:01 -0000, "Norman Wells" <h...@unseen.ac.am>
wrote:
>No. Whatever the circumstances, if there's no match it eliminates the
>vast majority of the population from any consideration whatsoever.
Not always. That depends on whether the evidential DNA came from the
only person who could have committed the crime or not. It also
depends on whether the test performed on the evidential DNA did in
fact succeed in obtaining the necessary markers.
>Against the few that are left, for any conviction to result, there
>always has to be sufficient further evidence.
That is indeed how it should work. In practice however there have
been quite a few convictions where the only significant evidence is
the DNA match itself, and a few convictions where the other evidence
tended to show innocence rather than guilt.
--
Cynic
wrote:
>No. Whatever the circumstances, if there's no match it eliminates the
>vast majority of the population from any consideration whatsoever.
only person who could have committed the crime or not. It also
depends on whether the test performed on the evidential DNA did in
fact succeed in obtaining the necessary markers.
>Against the few that are left, for any conviction to result, there
>always has to be sufficient further evidence.
been quite a few convictions where the only significant evidence is
the DNA match itself, and a few convictions where the other evidence
tended to show innocence rather than guilt.
--
Cynic
DocInsanity
Oct 30, 2012, 10:13:29 AM10/30/12
to
(sigh) if only you could actually understand the law, you might have a chance of making a sensible comment about it. What a prat.
Cynic
Oct 30, 2012, 10:19:45 AM10/30/12
to
On Tue, 30 Oct 2012 12:32:58 +0000, Martin Brown
<|||newspam|||@nezumi.demon.co.uk> wrote:
>I don't understand why his defence did not demand a re-examination of
>the forensic evidence early on though. Can someone explain?
I suspect that the answer will be that legal aid did not cover such a
<|||newspam|||@nezumi.demon.co.uk> wrote:
>I don't understand why his defence did not demand a re-examination of
>the forensic evidence early on though. Can someone explain?
re-test, and the defendent could not afford to pay for it himself.
Legal aid these days is such that if you are earning a very modest
salary, you will have to pay for your own defence. Even if you can
get legal aid, it may not cover forensic analysis or an expert
witness. Sure, you will get it back (or at least some of it) if you
are found not guilty, but that's a big gamble and you still have to
find aa 5 figure sum of money up front.
The punishment that will be imposed if you plead guilty may well be
preferable to taking the gamble of defending the charge and ending up
a homeless bankrupt.
--
Cynic
Steve Firth
Oct 30, 2012, 11:05:06 AM10/30/12
to
<richardm...@googlemail.com> wrote:
> If in the above example there was
> 27% DNA matches would this not
> suggest that the database is biased
> or that the biology is incorrect.
No. It indicates, as previously explained at length, that statistics works,
> If in the above example there was
> 27% DNA matches would this not
> suggest that the database is biased
> or that the biology is incorrect.
the birthday paradox is a real effect and the chances of a match against a
complete stranger increase rapidly as the size of the NDNAD increases.
--
<•DarWin><|
_/ _/
Big Les Wade
Oct 30, 2012, 11:26:13 AM10/30/12
to
richardm...@googlemail.com posted
say Mr P, who is not the suspect, left the DNA at the crime scene.
The chance that *any* other person left the DNA depends on how many
people there are in the suspect population. It gets more likely as the
suspect population gets larger.
>
>If in the above example there was
>27% DNA matches would this not
>suggest that the database is biased
>or that the biology is incorrect. i.e.
I can't work out what you are asking I am afraid. What do you mean by
"If in the above example there was 27% DNA matches"? Where does the 27%
come from?
>if a 6 sided die you would expect a 1/6
>chance of a 2 or a 3 etc where as in fact
>it may infact be a loaded die and thus a
>propensity towards one or more number.
True, but in the web page you quote the guy is calculating allele
frequencies from observation, not from symmetry considerations. So if
there is a "loaded die" in the alleles his data will show it, just as
you would be able to detect a loaded die by throwing it lots of times.
--
Les
>As my knowledge of DNA is very limited
>and to help me understand the issue
>please correct the following:
>
>If we take the DNA profile from
>http://dna-view.com/profile.htm
>
>
>It suggests in their example that out
>a database of 428-432 there is a
>1/7000 that another persons left
>the DNA.
There is a 1/7000 chance that another *particular* individual, let us
>and to help me understand the issue
>please correct the following:
>
>If we take the DNA profile from
>http://dna-view.com/profile.htm
>
>
>It suggests in their example that out
>a database of 428-432 there is a
>1/7000 that another persons left
>the DNA.
say Mr P, who is not the suspect, left the DNA at the crime scene.
The chance that *any* other person left the DNA depends on how many
people there are in the suspect population. It gets more likely as the
suspect population gets larger.
>
>If in the above example there was
>27% DNA matches would this not
>suggest that the database is biased
>or that the biology is incorrect. i.e.
"If in the above example there was 27% DNA matches"? Where does the 27%
come from?
>if a 6 sided die you would expect a 1/6
>chance of a 2 or a 3 etc where as in fact
>it may infact be a loaded die and thus a
>propensity towards one or more number.
frequencies from observation, not from symmetry considerations. So if
there is a "loaded die" in the alleles his data will show it, just as
you would be able to detect a loaded die by throwing it lots of times.
--
Les
Big Les Wade
Oct 30, 2012, 11:28:13 AM10/30/12
to
DocInsanity <docin...@googlemail.com> posted
you're killfiled.
--
Les
>(sigh) if only you could actually understand the law, you might have a
>chance of making a sensible comment about it. What a prat.
You have one more chance to make a constructive comment. After that,
>chance of making a sensible comment about it. What a prat.
you're killfiled.
--
Les
LebesgueMeasure
Oct 30, 2012, 11:29:29 AM10/30/12
to
"Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message
news:G2Njs.11688$Be.1...@newsfe13.iad...
officer asking "no". He asked me why. I asked him how many unconvicted
people his force had removed from the DNA Criminal database. He didn't
know. I told him to come back when 100% of innocent people had been removed
from the database. He hasn't come back and that was 4 years ago.
news:G2Njs.11688$Be.1...@newsfe13.iad...
>
> I used to think that NDNAD matching was trustworthy and had operational
> integrity. But now I will *never* volunteer to provide a DNA sample.
>
>
Same here. I was asked once to volunteer a DNA sample. I firmly told the
> I used to think that NDNAD matching was trustworthy and had operational
> integrity. But now I will *never* volunteer to provide a DNA sample.
>
>
officer asking "no". He asked me why. I asked him how many unconvicted
people his force had removed from the DNA Criminal database. He didn't
know. I told him to come back when 100% of innocent people had been removed
from the database. He hasn't come back and that was 4 years ago.
DocInsanity
Oct 30, 2012, 11:31:01 AM10/30/12
to
On Monday, October 29, 2012 7:17:20 AM UTC, Great one wrote:
> the British Home
>
> Office has reported that between 2001 and 2006, 27.6% of the matches reported from
>
> searches of the UK National DNA Database (NDNAD) were to more than one person in
>
> the database.
If this shocks you, try looking at the science of fingerprinting matching.
> the British Home
>
> Office has reported that between 2001 and 2006, 27.6% of the matches reported from
>
> searches of the UK National DNA Database (NDNAD) were to more than one person in
>
> the database.
LebesgueMeasure
Oct 30, 2012, 11:31:28 AM10/30/12
to
"DocInsanity" <docin...@googlemail.com> wrote in message
news:078890ca-172b-4083...@googlegroups.com...
news:078890ca-172b-4083...@googlegroups.com...
>
> -------------------------------
> (sigh) if only you could actually understand the law, you might have a
> chance of making a sensible comment about it. What a prat.
>
>
I have yet to see a sensible comment from you on any subject.
> -------------------------------
> (sigh) if only you could actually understand the law, you might have a
> chance of making a sensible comment about it. What a prat.
>
>
LebesgueMeasure
Oct 30, 2012, 11:34:38 AM10/30/12
to
"Norman Wells" <h...@unseen.ac.am> wrote in message
news:96Njs.58252$pg2....@fx18.am4...
It is the police who decide to abandon the investigation when they have a
DNA match and nothing else.
news:96Njs.58252$pg2....@fx18.am4...
DNA match and nothing else.
DocInsanity
Oct 30, 2012, 11:35:02 AM10/30/12
to
DocInsanity
Oct 30, 2012, 11:35:57 AM10/30/12
to
LebesgueMeasure
Oct 30, 2012, 11:42:16 AM10/30/12
to
"DocInsanity" <docin...@googlemail.com> wrote in message
news:adeddc85-ae92-478d...@googlegroups.com...
>
> I don't read any of your tripe.
>
>
Good. I'd rather creepy nonce like you didn't read my posts.
> I don't read any of your tripe.
>
>
DocInsanity
Oct 30, 2012, 11:43:22 AM10/30/12
to
LebesgueMeasure
Oct 30, 2012, 11:44:46 AM10/30/12
to
"DocInsanity" <docin...@googlemail.com> wrote in message
news:0db19dd9-403d-4dec...@googlegroups.com...
>>
>> DNA match and nothing else.
>
> You can come out with this crap, but it doesn't mean it's true
>
>
You said you didn't read my posts. Lying again.
>> DNA match and nothing else.
>
> You can come out with this crap, but it doesn't mean it's true
>
>
DocInsanity
Oct 30, 2012, 11:46:27 AM10/30/12
to
That's the trouble with the internet, retards like you can express an opinion even though you know fuck all.
LebesgueMeasure
Oct 30, 2012, 11:54:20 AM10/30/12
to
"DocInsanity" <docin...@googlemail.com> wrote in message
news:123abbd8-b451-42a1...@googlegroups.com...
>
> How do you manage to turn on a computer?
>
>
There's usually a switch on the side or a button on the front.
> How do you manage to turn on a computer?
>
>
Don't you know anything?
Janitor of Lunacy
Oct 30, 2012, 12:03:28 PM10/30/12
to
"LebesgueMeasure" <lebe...@gmail.com> wrote in message
news:ksSjs.51$Sa6...@en-nntp-06.am2.easynews.com...
evidence and decide whether they think it's enough to convince a jury. They
might culturally have a less cynical approach to DNA evidence than some, but
ultimately they will have to convince a judge and posssibly a jury that DNA
evidence alone is sufficient BRD.
Big Les Wade
Oct 30, 2012, 12:51:27 PM10/30/12
to
LebesgueMeasure <lebe...@gmail.com> posted
jury. Having got that far they can play up the
"DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to persuade
the jury that no other incriminating evidence is necessary.
>It is the police who decide to abandon the investigation when they have a
>DNA match and nothing else.
They'd never do that.
--
Les
>"Norman Wells" <h...@unseen.ac.am> wrote in message
>news:96Njs.58252$pg2....@fx18.am4...
>> Big Les Wade wrote:
>>> Paul Cummins <uset...@stedtelephone.invalid> posted
>>>> We were about to embark at Dover, when h...@unseen.ac.am (Norman
>>>> Wells) came up to me and whispered:
>>>>
>>>>> DNA testing is
>>>>> not proof that someone did the deed, it's proof that the vast
>>>>> majority didn't. To prove guilt, there _always_ has to be
>>>>> further evidence besides a DNA match.
>>>>
>>>> Fuck!
>>>>
>>>> Norman said something profound and accurate.
>>>
>>> Not really. Once the police have a "DNA match", then they will regard
>>> just about any other fact as further evidence against the suspect. The
>>> fact that he was an adult male and lived in England at the time of the
>>> crime will do, if nothing else emerges.
>>
>> It's not the police who decide, however. It's the jury.
It's the police (or the CPS) who decide whether to put the case before a
>news:96Njs.58252$pg2....@fx18.am4...
>> Big Les Wade wrote:
>>> Paul Cummins <uset...@stedtelephone.invalid> posted
>>>> We were about to embark at Dover, when h...@unseen.ac.am (Norman
>>>> Wells) came up to me and whispered:
>>>>
>>>>> DNA testing is
>>>>> not proof that someone did the deed, it's proof that the vast
>>>>> majority didn't. To prove guilt, there _always_ has to be
>>>>> further evidence besides a DNA match.
>>>>
>>>> Fuck!
>>>>
>>>> Norman said something profound and accurate.
>>>
>>> Not really. Once the police have a "DNA match", then they will regard
>>> just about any other fact as further evidence against the suspect. The
>>> fact that he was an adult male and lived in England at the time of the
>>> crime will do, if nothing else emerges.
>>
>> It's not the police who decide, however. It's the jury.
jury. Having got that far they can play up the
"DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to persuade
the jury that no other incriminating evidence is necessary.
>It is the police who decide to abandon the investigation when they have a
>DNA match and nothing else.
--
Les
Norman Wells
Oct 30, 2012, 12:59:46 PM10/30/12
to
should drop the case.
LebesgueMeasure
Oct 30, 2012, 1:02:58 PM10/30/12
to
"Norman Wells" <h...@unseen.ac.am> wrote in message
news:5JTjs.110818$Xv.5...@fx07.am4...
Norman Wells
Oct 30, 2012, 1:02:33 PM10/30/12
to
Big Les Wade wrote:
> LebesgueMeasure <lebe...@gmail.com> posted
>> "Norman Wells" <h...@unseen.ac.am> wrote in message
>> news:96Njs.58252$pg2....@fx18.am4...
>>> Big Les Wade wrote:
>>>> Paul Cummins <uset...@stedtelephone.invalid> posted
>>>>> We were about to embark at Dover, when h...@unseen.ac.am (Norman
>>>>> Wells) came up to me and whispered:
>>>>>
>>>>>> DNA testing is
>>>>>> not proof that someone did the deed, it's proof that the vast
>>>>>> majority didn't. To prove guilt, there _always_ has to be
>>>>>> further evidence besides a DNA match.
>>>>>
>>>>> Fuck!
>>>>>
>>>>> Norman said something profound and accurate.
>>>>
>>>> Not really. Once the police have a "DNA match", then they will
>>>> regard just about any other fact as further evidence against the
>>>> suspect. The fact that he was an adult male and lived in England
>>>> at the time of the crime will do, if nothing else emerges.
>>>
>>> It's not the police who decide, however. It's the jury.
>
> It's the police (or the CPS) who decide whether to put the case
> before a jury. Having got that far they can play up the
> "DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to
> persuade the jury that no other incriminating evidence is necessary.
And the defence lawyer says it isn't.
> LebesgueMeasure <lebe...@gmail.com> posted
>> "Norman Wells" <h...@unseen.ac.am> wrote in message
>> news:96Njs.58252$pg2....@fx18.am4...
>>> Big Les Wade wrote:
>>>> Paul Cummins <uset...@stedtelephone.invalid> posted
>>>>> We were about to embark at Dover, when h...@unseen.ac.am (Norman
>>>>> Wells) came up to me and whispered:
>>>>>
>>>>>> DNA testing is
>>>>>> not proof that someone did the deed, it's proof that the vast
>>>>>> majority didn't. To prove guilt, there _always_ has to be
>>>>>> further evidence besides a DNA match.
>>>>>
>>>>> Fuck!
>>>>>
>>>>> Norman said something profound and accurate.
>>>>
>>>> Not really. Once the police have a "DNA match", then they will
>>>> regard just about any other fact as further evidence against the
>>>> suspect. The fact that he was an adult male and lived in England
>>>> at the time of the crime will do, if nothing else emerges.
>>>
>>> It's not the police who decide, however. It's the jury.
>
> It's the police (or the CPS) who decide whether to put the case
> before a jury. Having got that far they can play up the
> "DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to
> persuade the jury that no other incriminating evidence is necessary.
It's how it works.
richardm...@googlemail.com
Oct 30, 2012, 1:03:56 PM10/30/12
to
Thanks Steve and Les
Norman Wells
Oct 30, 2012, 1:14:26 PM10/30/12
to
Cynic
Oct 30, 2012, 1:20:10 PM10/30/12
to
On Tue, 30 Oct 2012 16:03:28 -0000, "Janitor of Lunacy"
<zo...@zonk.com> wrote:
>It's much more likely to be the CPS- they will look at the totality of the
>evidence and decide whether they think it's enough to convince a jury. They
>might culturally have a less cynical approach to DNA evidence than some, but
>ultimately they will have to convince a judge and posssibly a jury that DNA
>evidence alone is sufficient BRD.
Due to the fact that the average person (and hence the average jury
<zo...@zonk.com> wrote:
>It's much more likely to be the CPS- they will look at the totality of the
>evidence and decide whether they think it's enough to convince a jury. They
>might culturally have a less cynical approach to DNA evidence than some, but
>ultimately they will have to convince a judge and posssibly a jury that DNA
>evidence alone is sufficient BRD.
member) regards DNA evidence as irrefutable, and the fact that the
prosecution expert will no doubt state that the probability of a false
match is umpty billion to one, the probability of a conviction on DNA
evidence alone is pretty high.
--
Cynic
Norman Wells
Oct 30, 2012, 1:25:38 PM10/30/12
to
Cynic
Oct 30, 2012, 1:42:08 PM10/30/12
to
On Tue, 30 Oct 2012 17:25:38 -0000, "Norman Wells" <h...@unseen.ac.am>
wrote:
>> Due to the fact that the average person (and hence the average jury
>> member) regards DNA evidence as irrefutable, and the fact that the
>> prosecution expert will no doubt state that the probability of a false
>> match is umpty billion to one, the probability of a conviction on DNA
>> evidence alone is pretty high.
>It's why we have defence lawyers too.
Whose client often cannot afford to pay for a dissenting expert, which
is the only way to rebut the prosecution expert's assertion. Besides
which the faith in DNA evidence has become too strong to easily change
within the space of a trial.
--
Cynic
wrote:
>> Due to the fact that the average person (and hence the average jury
>> member) regards DNA evidence as irrefutable, and the fact that the
>> prosecution expert will no doubt state that the probability of a false
>> match is umpty billion to one, the probability of a conviction on DNA
>> evidence alone is pretty high.
>It's why we have defence lawyers too.
is the only way to rebut the prosecution expert's assertion. Besides
which the faith in DNA evidence has become too strong to easily change
within the space of a trial.
--
Cynic
Cynic
Oct 30, 2012, 1:46:36 PM10/30/12
to
On Tue, 30 Oct 2012 17:02:33 -0000, "Norman Wells" <h...@unseen.ac.am>
wrote:
>> It's the police (or the CPS) who decide whether to put the case
>> before a jury. Having got that far they can play up the
>> "DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to
>> persuade the jury that no other incriminating evidence is necessary.
>
>And the defence lawyer says it isn't.
A lawyer is not qualified to rebut an expert's assertion, he can only
attempt to question its robustness. But to do that requires some
expertise in the field of DNA or statistics, which very few lawyers
will have.
>It's how it works.
How it works is that the defendent is told that if he forks out circa
Ł6000 he could have his own expert in the witness box to challenge the
prosecution expert's conclusions, but there is no guarantee he will be
found not guilty, and no guarantee that he will get his Ł6000 back
even if he is.
--
Cynic
wrote:
>> It's the police (or the CPS) who decide whether to put the case
>> before a jury. Having got that far they can play up the
>> "DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to
>> persuade the jury that no other incriminating evidence is necessary.
>
>And the defence lawyer says it isn't.
attempt to question its robustness. But to do that requires some
expertise in the field of DNA or statistics, which very few lawyers
will have.
>It's how it works.
Ł6000 he could have his own expert in the witness box to challenge the
prosecution expert's conclusions, but there is no guarantee he will be
found not guilty, and no guarantee that he will get his Ł6000 back
even if he is.
--
Cynic
Janitor of Lunacy
Oct 30, 2012, 2:48:13 PM10/30/12
to
"Cynic" <cyni...@yahoo.co.uk> wrote in message
news:509010e8...@127.0.0.1...
had any difficulty in getting a LA representation certificate extended to
cover such an expert. He also briefs counsel experienced in such matters (of
whom there are plenty in Liverpool), so at least can put up a decent fight
in court.
Doc Insanity
Oct 30, 2012, 3:40:02 PM10/30/12
to
"Cynic" wrote in message news:5090117f...@127.0.0.1...
On Tue, 30 Oct 2012 17:02:33 -0000, "Norman Wells" <h...@unseen.ac.am>
wrote:
>> It's the police (or the CPS) who decide whether to put the case
>> before a jury. Having got that far they can play up the
>> "DNA-says-its-a-billion-to-one-chance-he's-innocent" routine to
>> persuade the jury that no other incriminating evidence is necessary.
>
>And the defence lawyer says it isn't.
A lawyer is not qualified to rebut an expert's assertion, he can only
attempt to question its robustness. But to do that requires some
expertise in the field of DNA or statistics, which very few lawyers
will have.
>It's how it works.
How it works is that the defendent is told that if he forks out circa
prosecution expert's conclusions, but there is no guarantee he will be
found not guilty, and no guarantee that he will get his �6000 back
even if he is.
-------------------------------------------------------------
Absolute rubbish. Contrary to what you're asserting, it's still trial by
jury, not trial by expert. Having spoken to a number of experienced expert
witnesses, I can assure you that they have come across advocates that are
able to pounce on their mistakes. The jury is entitled to ignore an expert's
opinion, and a lawyer can be perfectly well qualified to point out flaws in
expert evidence.
"Armchair experts" like yourself don't realise how the system operates.
richardm...@googlemail.com
Oct 30, 2012, 3:47:39 PM10/30/12
to
On Tuesday, October 30, 2012 5:03:56 PM UTC, richardm...@googlemail.com wrote:
> Thanks Steve and Les
The penny did drop!
The 1 in 7000 is a kin to the 1 in 6 of a die
and the 27% if akin to rolling 2 sixes in a row
> Thanks Steve and Les
The penny did drop!
The 1 in 7000 is a kin to the 1 in 6 of a die
and the 27% if akin to rolling 2 sixes in a row
Cynic
Oct 30, 2012, 3:52:59 PM10/30/12
to
On Tue, 30 Oct 2012 18:48:13 -0000, "Janitor of Lunacy"
<zo...@zonk.com> wrote:
>> Whose client often cannot afford to pay for a dissenting expert, which
>> is the only way to rebut the prosecution expert's assertion. Besides
>> which the faith in DNA evidence has become too strong to easily change
>> within the space of a trial.
<zo...@zonk.com> wrote:
>> Whose client often cannot afford to pay for a dissenting expert, which
>> is the only way to rebut the prosecution expert's assertion. Besides
>> which the faith in DNA evidence has become too strong to easily change
>> within the space of a trial.
>I've occasionally done work for a solicitor friend, and he says he has never
>had any difficulty in getting a LA representation certificate extended to
>cover such an expert. He also briefs counsel experienced in such matters (of
>whom there are plenty in Liverpool), so at least can put up a decent fight
>in court.
How long ago was your experience? The legal aid rules have changed
>had any difficulty in getting a LA representation certificate extended to
>cover such an expert. He also briefs counsel experienced in such matters (of
>whom there are plenty in Liverpool), so at least can put up a decent fight
>in court.
significantly over the past few years. In magistrates' court, hardly
anyone who has a job will qualify for *any* legal aid whatsoever. In
Crown Court the view is often taken that only one expert witness is
required (because they are supposed to be neutral fact-finders), and
aid is not granted to cover the appointment of a second, defence
expert. There is huge political pressure to cut back of LA funding.
If the defendent pays his own legal costs and wins the case (is found
NG), the court will very possibly only refund a portion of his costs,
because a judge can state (with hindsight) that it was not necessary
to hire an expert witness, or an expensive barrister etc., and will
pay the defendent only those costs that it is considered would have
been funded by legal aid had the defendent qualified.
*Many* defendents are made bankrupt by the experience, in addition to
any fines or other punishment imposed by the court.
--
Cynic
LebesgueMeasure
Oct 30, 2012, 5:02:01 PM10/30/12
to
"Doc Insanity" <docin...@gmail.com> wrote in message
news:k6paeo$2it$1...@dont-email.me...
news:k6paeo$2it$1...@dont-email.me...
>
> Having spoken to a number of experienced expert witnesses,
>
Saving on lawyers fees by defending yourself?
> Having spoken to a number of experienced expert witnesses,
>
Great one
Oct 30, 2012, 5:16:41 PM10/30/12
to
1. The Court will not take your word for something.
(Unlike sol/bar. you have not given a professional oath to the court to tell the truth.)
2. If all your evidence is challenged because your opponents claim it has not been independently verified..well you can see the obvious problem..I hope.
3. The court is NOT interested in the truth.....only what you can get a jury to believe.
4...Appeal courts maybe interested in truth...but the court that puts you down isnt..so..you could end up in jail and then have to pay out a fortune to get out......plus you collect a crim record for life.
Norman Wells
Oct 30, 2012, 5:31:57 PM10/30/12
to
richardm...@googlemail.com wrote:
> On Tuesday, October 30, 2012 5:03:56 PM UTC,
> richardm...@googlemail.com wrote:
>> Thanks Steve and Les
>
> The penny did drop!
>
> The 1 in 7000 is a kin to the 1 in 6 of a die
Is it really? How do you figure that then?
> On Tuesday, October 30, 2012 5:03:56 PM UTC,
> richardm...@googlemail.com wrote:
>> Thanks Steve and Les
>
> The penny did drop!
>
> The 1 in 7000 is a kin to the 1 in 6 of a die
> and the 27% if akin to rolling 2 sixes in a row
was 1/36, or rather less than 3%.
Great one
Oct 30, 2012, 5:40:45 PM10/30/12
to
In one roll, the probability of rolling a 6 is 1/6.
For two rolls, there is a 1/6 probability of rolling a six on the first roll.
If this occurs, we've satisfied our condition. There is a 5/6 probability
that the first roll is not a 6. In that case, we need to see if the second
roll is a 6. The probability of the second roll being a 6 is 1/6, so our
overall probability is 1/6 + (5/6)*(1/6) = 11/36. Why did I multiply the
second 1/6 by 5/6? Because I only need to consider the 5/6 of the time that
the first roll wasn't a 6. As you can see the probability is slightly less
than 2/6.
For three rolls, there is a 1/6 probability of rolling a six on the first
roll. There is a 5/6 probability that the first roll is not a 6. In that
case, we need to see if the second roll is a 6. The probability of the second
roll being a 6 is 1/6, giving us a probability of 11/36. There is a 25/36
probability that neither of the first two rolls was a 6. In that case, we
need to see if the third roll is a 6. The probability of the third roll being
a 6 is 1/6, giving us a probability of 1/6 + (5/6)*(1/6) + (25/36)*(1/6) =
91/216. Again, this is less than 3/6.
The general formula for rolling at least one 6 in n rolls is 1 - (5/6)^n.
One final note: the probability of rolling _exactly_ one 6 in n rolls is
different, and a little more difficult to compute.
Norman Wells
Oct 30, 2012, 5:46:13 PM10/30/12
to
> The general formula for rolling at least one 6 in n rolls is 1 -
> (5/6)^n.
rolling two sixes in a row. You can check that by reading the above.
> One final note: the probability of rolling _exactly_ one 6 in n rolls
> is different, and a little more difficult to compute.
things.
Paul Cummins
Oct 30, 2012, 6:10:00 PM10/30/12
to
We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
came up to me and whispered:
came up to me and whispered:
> You haven't studied probability,
So what? You haven't studied anything...
--
Paul Cummins - Always a NetHead
Wasting Bandwidth since 1981
IF you think this http://bit.ly/u5EP3p is cruel
please sign this http://bit.ly/sKkzEx
---- If it's below this line, I didn't write it ----
Norman Wells
Oct 30, 2012, 6:19:24 PM10/30/12
to
Paul Cummins wrote:
> We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
> came up to me and whispered:
>
>> You haven't studied probability,
>
> So what? You haven't studied anything...
Yeah? And what do you know?
> We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
> came up to me and whispered:
>
>> You haven't studied probability,
>
> So what? You haven't studied anything...
Martin Brown
Oct 30, 2012, 6:27:00 PM10/30/12
to
Officers on p30 start of p31. I quote here:
"Police must therefore be aware of the consequences
of adventitious matches and that DNA evidence is not
used in criminal prosecution without careful consideration
of the surrounding circumstances.
Indeed, this was one of the findings in the
Appeal Court hearing of R. v Lashley (2000), where DNA
evidence had been relied upon for a conviction. There
was no evidence that the appellant had been in the
vicinity of where the crime had been committed and
hence the DNA (SGM only) profile was just as likely to
have been left by any one of five or six other men in
the United Kingdom."
--
Regards,
Martin Brown
Paul Cummins
Oct 30, 2012, 7:38:00 PM10/30/12
to
We were about to embark at Dover, when h...@unseen.ac.am (Norman Wells)
came up to me and whispered:
>
came up to me and whispered:
>
> Yeah? And what do you know?
I could discuss my qualifications and expertise, but I see no purpose, as
you can't afford to employ me.
Paul Nutteing
Oct 31, 2012, 4:33:31 AM10/31/12
to
Cynic <cyni...@yahoo.co.uk> wrote in message
news:50900bc1...@127.0.0.1...
You can always tell when an "expert witness" is lying to
the court. He/she refers to "a billion to one" - whatever the
figure is in a particular circumstance
its never going to be a billion to one almost every time.
Just monitor court reports for a few weeks and you will soon
realise how much corruption there is going on, as simple as that.
I thought perjury in courts was a criminal offence.
What Special Branch don't want you to know about
DNA profiles
http://www.nutteing2.50megs.com/dnapr.htm
Great one
Oct 31, 2012, 4:51:35 AM10/31/12
to
richardm...@googlemail.com
Oct 31, 2012, 5:07:37 AM10/31/12
to
Is it really? How do you figure that then?
> and the 27% if akin to rolling 2 sixes in a row
Is it really? I'd have thought the chance of rolling
2 sixes in a row
was 1/36, or rather less than 3%.
-----
> and the 27% if akin to rolling 2 sixes in a row
Is it really? I'd have thought the chance of rolling
2 sixes in a row
was 1/36, or rather less than 3%.
It is to do with relationships, just like Steve
said with the room full of 30 people and a pair
sharing a birthday.
If we were to compare say birthdays between 2
people there is one relationship
i.e one roll of the die.
with 3 people there are 3 relationships,
4 people it is 6 etc.
The formula is r=n(n-1)/2,
where r is relationship(s) and n is the
number of people. So
n r
1 0
2 1
3 3
4 6
5 10
6 15
. .
. .
30 435
The 435 represents the number of goes or rolls of
the die. Although there are 365 days in the year it
does not mean that there will be a pair who share a
birthday. It is equivalent of attempting to get a 6
in a die and having circa 7 goes.
You may get 0,1,2,3,4,5,6 or 7 sixes.
If you have a million people on a database there
are circa 4.99x10 power 11 relationships therefore
if there is a 1 in a million chance of a DNA match
then there is a fair chance of a hit.
There a more to it that this however it is the
basis.
Big Les Wade
Oct 31, 2012, 6:00:19 AM10/31/12
to
Martin Brown <|||newspam|||@nezumi.demon.co.uk> posted
posted.)
--
Les
>
>It is discussed in the Guide to DNA for Lawyers and Investigating
>Officers on p30 start of p31. I quote here:
Is this document publicly available? (Sorry if it's already been
>It is discussed in the Guide to DNA for Lawyers and Investigating
>Officers on p30 start of p31. I quote here:
posted.)
--
Les
Martin Brown
Oct 31, 2012, 6:12:39 AM10/31/12
to
On 30/10/2012 19:47, richardm...@googlemail.com wrote:
> On Tuesday, October 30, 2012 5:03:56 PM UTC, richardm...@googlemail.com wrote:
>> Thanks Steve and Les
>
> The penny did drop!
Unfortunately it fell onto the floor.
> On Tuesday, October 30, 2012 5:03:56 PM UTC, richardm...@googlemail.com wrote:
>> Thanks Steve and Les
>
> The penny did drop!
>
> The 1 in 7000 is a kin to the 1 in 6 of a die
Only in a very weak sense. It is much more like throwing an incredibly
> The 1 in 7000 is a kin to the 1 in 6 of a die
biassed coin that gives heads on average only once in every 7000 throws.
>
> and the 27% if akin to rolling 2 sixes in a row
*NO!* - It is like *any* combination of sixes in two throws.
> and the 27% if akin to rolling 2 sixes in a row
1-6,2-6,3-6,4-6,5-6,6-6 and 6-1, 6-2, 6-3, 6-4, 6-5 (11 in all)
1 - (5/6)^2 = 11/36 = 30.6%
This is basic statistics rather than DNA.
You should also lookup the "Birthday Paradox" which isn't a paradox at
all but is a great way to take money off the innumerate if you are so
inclined. The bet that two people share the same birthday in a room with
23 people in it is about even odds. In fact they are slightly better
because birthdays are also clustered by seasonal factors.
http://en.wikipedia.org/wiki/Birthday_problem
and
http://www.panix.com/~murphy/bday.html
I will try and explain this first away from the DNA controversy.
Lets consider a simple binary trial heads or tails
- only two possible outcomes.
Probability of heads x = 1/7000
Probability of tails 1-x = 1 - 1/7000 = 6999/7000
I am now writing "x" as shorthand for 1/7000
Mathematicians are lazy (although I am a physicist).
So for a single trial the outcomes are
H x
T 1-x
For two trials the possible outcomes are
HH x.x
HT x(1-x)
TH (1-x)x
TT (1-x)^2
Now we don't care if it was HT or TH so in shorthand
HH x^2
HT 2x(1-x)
TT (1-x)^2
And for three trials the possible outcomes are
HHH
HHT
HTH
HTT
THH
THT
TTH
TTT
Rather than deal with the clutter HHT, HTH, THH and HTT, THT, TTH merged
HHH x.x.x
HHT 3x.x(1-x)
HTT 3x(1-x)(1-x)
TTT (1-x)(1-x)(1-x)
These patterns are from the binomial theorem and Pascals triangle.
http://en.wikipedia.org/wiki/Binomial_expansion
Now the crucial point is that if there are N trials we can very easily
compute the probability of never seeing any heads.
T....T (N times in a row) = (1-x)^N
So taking your concrete numerical example and 27% having at least one or
more heads we know that 73% had no heads at all in a run of length N.
So to figure out how big N needs to be we solve for N
(6999/7000)^N = 0.73
I get this to be for N = 2203
So you get 27% with at least one or more heads occuring when there are
about 2200 individual trials at odds of 1:7000.
Put another way
1:7000 is pretty good if you have a pulled guy you already suspect for
other reasons and *then* use DNA to place him at the crime scene.
But it is lousy if you trawl the 6M NDNAD records and get 857 suspects
and go and grab them in turn until you can find one without an alibi. It
is this latter mode of failure that is now occurring.
The odds in the total UK population of 60M are worse still. At matching
odds of 1:7000 there are 8570 possible matches in the whole population.
The actual match probability number in DNA work is officially 1:10^9
according to expert witnesses, but based on the published data I have my
doubts. It is important for us to know the correct value!
I think the only way to resolve the diminishing trust resulting form the
mishandling of DNA evidence would be to turn an anonymised copy of the
NDNA database over for academic statistical scrutiny. I really want to
know what the empirical values of match probability are in practice and
also how they vary with ethnicity.
--
Regards,
Martin Brown
Martin Brown
Oct 31, 2012, 6:17:47 AM10/31/12
to
On 30/10/2012 14:13, DocInsanity wrote:
> On Tuesday, October 30, 2012 12:33:01 PM UTC, Martin Brown wrote:
>> On 30/10/2012 09:47, Norman Wells wrote:
[long snip]
>>> I don't think there's anyone here who would disagree.
>>
>>
>>
>> I can think of at least one Insane Doc that would.
>>
>>>
>>
>>>> And I count being innocent and held in police custody for an extended
>>
>>>> period of time as being almost as bad as being incorrectly found
>>
>>>> guilty.
>>
>>>
>>
>>> No it isn't. But of course it shouldn't happen unreasonably.
>>
>>
>>
>> It did though and it appears that he has no legal redress.
>>
>>
>>
>> --
>>
>> Regards,
>>
>> Martin Brown
> -------------------------------
> (sigh) if only you could actually understand the law, you might have a chance of making a sensible comment about it. What a prat.
You quote the entirety of my post double-spaced to add a lame single
line of trolling dross announcing that you are a prat.
You are a third rate fuckwit with delusions of adequacy.
Get back under your bridge or I will use Troll-B-Gone.
--
Regards,
Martin Brown
> On Tuesday, October 30, 2012 12:33:01 PM UTC, Martin Brown wrote:
>> On 30/10/2012 09:47, Norman Wells wrote:
[long snip]
>>> I don't think there's anyone here who would disagree.
>>
>>
>>
>> I can think of at least one Insane Doc that would.
>>
>>>
>>
>>>> And I count being innocent and held in police custody for an extended
>>
>>>> period of time as being almost as bad as being incorrectly found
>>
>>>> guilty.
>>
>>>
>>
>>> No it isn't. But of course it shouldn't happen unreasonably.
>>
>>
>>
>> It did though and it appears that he has no legal redress.
>>
>>
>>
>> --
>>
>> Regards,
>>
>> Martin Brown
> -------------------------------
> (sigh) if only you could actually understand the law, you might have a chance of making a sensible comment about it. What a prat.
You quote the entirety of my post double-spaced to add a lame single
line of trolling dross announcing that you are a prat.
You are a third rate fuckwit with delusions of adequacy.
Get back under your bridge or I will use Troll-B-Gone.
--
Regards,
Martin Brown
richardm...@googlemail.com
Oct 31, 2012, 6:25:50 AM10/31/12
to
Thanks Martin
Its slowly coming back to me. Such a long time
ago!
Its slowly coming back to me. Such a long time
ago!
Martin Brown
Oct 31, 2012, 6:29:31 AM10/31/12
to
http://www.cps.gov.uk/legal/assets/uploads/files/lawyers'%20DNA%20guide%20KSWilliams%20190208%20(i).pdf
--
Regards,
Martin Brown
Paul Nutteing
Oct 31, 2012, 7:55:01 AM10/31/12
to
Martin Brown <|||newspam|||@nezumi.demon.co.uk> wrote in message
news:tQ6ks.1724$Uu4....@newsfe12.iad...
I draw your attention to this diclosure
<http://www.nlada.org/Defender/forensics/for_lib/Documents/1148592247.61/M
yers%20CAC%20Presentation.pdf>
Kathryn Troyer was too low in the forensic hierarchy to know
she is not supposed to divulge such anonymised material.
A black guy's DNA profile matching a white guy's on
10 loci in a database of only 65,493 people.
Not the same 10 as used by the NDNAD and 10 of 13 but still disturbing to
the keepers of these numbers.
The substructuring (partial matches) of that disclosure still cannot be
explained assuming Bayesian independence.
The substructuring pattern also gets over the problem of duplicate
profiles from people arrested under different aliases and
erroroneous personal details as a reason for non-disclosure of
all the anaonymised DNA profiles on these databases.
news:tQ6ks.1724$Uu4....@newsfe12.iad...
<http://www.nlada.org/Defender/forensics/for_lib/Documents/1148592247.61/M
yers%20CAC%20Presentation.pdf>
Kathryn Troyer was too low in the forensic hierarchy to know
she is not supposed to divulge such anonymised material.
A black guy's DNA profile matching a white guy's on
10 loci in a database of only 65,493 people.
Not the same 10 as used by the NDNAD and 10 of 13 but still disturbing to
the keepers of these numbers.
The substructuring (partial matches) of that disclosure still cannot be
explained assuming Bayesian independence.
The substructuring pattern also gets over the problem of duplicate
profiles from people arrested under different aliases and
erroroneous personal details as a reason for non-disclosure of
all the anaonymised DNA profiles on these databases.
Martin Brown
Oct 31, 2012, 10:07:59 AM10/31/12
to
On 31/10/2012 11:55, Paul Nutteing wrote:
> Martin Brown <|||newspam|||@nezumi.demon.co.uk> wrote in message
> news:tQ6ks.1724$Uu4....@newsfe12.iad...
>>
[snip]
> Martin Brown <|||newspam|||@nezumi.demon.co.uk> wrote in message
> news:tQ6ks.1724$Uu4....@newsfe12.iad...
>>
>> The actual match probability number in DNA work is officially 1:10^9
>> according to expert witnesses, but based on the published data I have my
>> doubts. It is important for us to know the correct value!
>>
>> I think the only way to resolve the diminishing trust resulting form the
>> mishandling of DNA evidence would be to turn an anonymised copy of the
>> NDNA database over for academic statistical scrutiny. I really want to
>> know what the empirical values of match probability are in practice and
>> also how they vary with ethnicity.
>>
>> --
>> Regards,
>> Martin Brown
>
> I draw your attention to this diclosure
> <http://www.nlada.org/Defender/forensics/for_lib/Documents/1148592247.61/M
> yers%20CAC%20Presentation.pdf>
>
Thanks for this. It will take a little bit a of reading. I have scanned
>> according to expert witnesses, but based on the published data I have my
>> doubts. It is important for us to know the correct value!
>>
>> I think the only way to resolve the diminishing trust resulting form the
>> mishandling of DNA evidence would be to turn an anonymised copy of the
>> NDNA database over for academic statistical scrutiny. I really want to
>> know what the empirical values of match probability are in practice and
>> also how they vary with ethnicity.
>>
>> --
>> Regards,
>> Martin Brown
>
> I draw your attention to this diclosure
> <http://www.nlada.org/Defender/forensics/for_lib/Documents/1148592247.61/M
> yers%20CAC%20Presentation.pdf>
>
through it and it does suggest again to me that the official figures for
match "uniqueness" are more than a little optimistic.
> Kathryn Troyer was too low in the forensic hierarchy to know
> she is not supposed to divulge such anonymised material.
> A black guy's DNA profile matching a white guy's on
> 10 loci in a database of only 65,493 people.
> Not the same 10 as used by the NDNAD and 10 of 13 but still disturbing to
> the keepers of these numbers.
> The substructuring (partial matches) of that disclosure still cannot be
> explained assuming Bayesian independence.
> The substructuring pattern also gets over the problem of duplicate
> profiles from people arrested under different aliases and
> erroroneous personal details as a reason for non-disclosure of
> all the anaonymised DNA profiles on these databases.
taking place since that it the method most likely to reveal any
unexpected systematic quirks.
>
> What Special Branch don't want you to know about
> DNA profiles
> http://www . nutteing2 . 50megs . com/dnapr.htm
>
>
following warning msg
Web Page: http://www . nutteing2 . 50megs . com/dnapr.htm
Detected viruses: JS:Trojan.Crypt.JD
Could be a false positive (wouldn't be the first time)
URLs broken with spaces to invalidate them
--
Regards,
Martin Brown
Doc Insanity
Oct 31, 2012, 2:49:01 PM10/31/12
to
"Big Les Wade" wrote in message news:YYfX0UKNI$jQF...@obviously.invalid...
DocInsanity <docin...@googlemail.com> posted
>(sigh) if only you could actually understand the law, you might have a
>chance of making a sensible comment about it. What a prat.
You have one more chance to make a constructive comment. After that,
>chance of making a sensible comment about it. What a prat.
you're killfiled.
--------------------------------------------------------------
Pointing out the invincible ignorance of some people IS being constructive.
All the armchair experts who make pronouncements here, it really makes you
shake your head in despair.
Jon Ribbens
Oct 31, 2012, 3:15:31 PM10/31/12
to
On 2012-10-31, Doc Insanity <docin...@gmail.com> wrote:
> "Big Les Wade" wrote in message news:YYfX0UKNI$jQF...@obviously.invalid...
>
> DocInsanity <docin...@googlemail.com> posted
>>(sigh) if only you could actually understand the law, you might have a
>>chance of making a sensible comment about it. What a prat.
>
> You have one more chance to make a constructive comment. After that,
> you're killfiled.
> --------------------------------------------------------------
> Pointing out the invincible ignorance of some people IS being constructive.
No, it isn't. I had already kill-filed you, why am I seeing this post?
> "Big Les Wade" wrote in message news:YYfX0UKNI$jQF...@obviously.invalid...
>
> DocInsanity <docin...@googlemail.com> posted
>>(sigh) if only you could actually understand the law, you might have a
>>chance of making a sensible comment about it. What a prat.
>
> You have one more chance to make a constructive comment. After that,
> you're killfiled.
> --------------------------------------------------------------
> Pointing out the invincible ignorance of some people IS being constructive.
Ah, I see... you've morphed your name slightly.
Big Les Wade
Oct 31, 2012, 5:27:24 PM10/31/12
to
Doc Insanity <docin...@gmail.com> posted
a useful contribution, you have to state exactly where you think people
are going wrong, and put them right, if necessary citing sources they
can check. You never even attempt to do that, which is why your time is
running out.
--
Les
>
>
>"Big Les Wade" wrote in message news:YYfX0UKNI$jQF...@obviously.invalid...
>
>DocInsanity <docin...@googlemail.com> posted
>>(sigh) if only you could actually understand the law, you might have a
>>chance of making a sensible comment about it. What a prat.
>
>You have one more chance to make a constructive comment. After that,
>you're killfiled.
>--------------------------------------------------------------
>Pointing out the invincible ignorance of some people IS being
>constructive.
Just stating "You are ignorant" is not constructive. If you want to make
>
>"Big Les Wade" wrote in message news:YYfX0UKNI$jQF...@obviously.invalid...
>
>DocInsanity <docin...@googlemail.com> posted
>>(sigh) if only you could actually understand the law, you might have a
>>chance of making a sensible comment about it. What a prat.
>
>You have one more chance to make a constructive comment. After that,
>you're killfiled.
>--------------------------------------------------------------
>Pointing out the invincible ignorance of some people IS being
>constructive.
a useful contribution, you have to state exactly where you think people
are going wrong, and put them right, if necessary citing sources they
can check. You never even attempt to do that, which is why your time is
running out.
--
Les
Doc Insanity
Nov 1, 2012, 10:49:52 AM11/1/12
to
"Great one" wrote in message
news:6fe335ef-3b71-4609...@googlegroups.com...
Doc Insanity
Nov 1, 2012, 10:51:58 AM11/1/12
to
"Great one" wrote in message
news:6fe335ef-3b71-4609...@googlegroups.com...
That comment just makes no sense. How could you save on lawyers fees by
speaking to expert witnesses? Expert witness statement is only available to
expert witness, self-evidently.
The guy is an imbecile.
Defending myself? It's called research.
Doc Insanity
Nov 1, 2012, 5:33:50 PM11/1/12
to
"Big Les Wade" wrote in message news:xvEtJqN8...@obviously.invalid...
Discussion is not a rational response to invincible ignorance, is it?
I certainly do explain the law and sources. Your rebuttal of my point about
probabilities has been disproved, any comment to make?
It is loading more messages.
0 new messages