Continuation Part 2 - Discussion of the Amanda Knox case

[RandyN]

The haplotype jokes are coming thick and fast today

I have to say that while certain erstwhile forum-associates have invested very large amounts of time and energy in a partisan attempt to question the DNA report, it's rather clear that they don't know what they are talking about. They are also focussing (probably deliberately) on one very narrow part of the report: the interpretation of the electropherograms.

But that's just one small part of this whole sorry story. GIGO means that all the errors that preceded the interpretation phase imply that the interpreted results cannot be trusted to be accurate or reliable - even if the interpretation was carried out absolutely properly. And part of that is directly Stefanoni's fault. After all, she was the one who attempted low template analysis in a lab environment that was not sufficient for such a task, and she herself was not trained (let alone accredited) in conducting low template analysis. SO her lab procedures (or, more accurately, lack of procedures) opened the door wide for contamination before the interpretation was ever even attempted.

I happen to believe that Stefanoni made serious (and provable) errors in the interpretation of the results. But she also made other massive errors - as did virtually everyone else throughout the process of collection, handling, storage and transportation of these items - that in themselves render these items inadmissible as evidence.

Please just one more........Halfa$$lotype........

Well <200picograms means special handling is required which we all know in this lab was impossible. Maybe this is the part Marasca means where Italy shows the world how to do DNA.

Ive read the expert report translated by someone I trust. The knife is out. The bra clasp is out. No if's and's or but's about it. They are both out!

The mixed DNA samples were not examined but they were meaningless anyway. The one questionable sample from Filomenas room "blob" is spilled juice... that sticky substance that collects DNA like a magnet. DNA dragged there by roaming CSI wearing dust collector shoe covers as they trapsed from room to room in the six weeks and no less than three or four different visits by 3 or 4 seperate police units. A halfa$$edinvestigation...x p alodocious.................hummmdiddlediddle...Im drunk gnite.

 

[Kevin_Lowe]

I'm sorry to see you got punted from idiots.org, Fine.

I'm not the least bit surprised it happened though - if anything I'm surprised they let you post that long after you broke ranks.

 

[Mary_H]

I'm sorry to see you got punted from idiots.org, Fine.

I'm not the least bit surprised it happened though - if anything I'm surprised they let you post that long after you broke ranks.

Wow. And Fine was expelled in the ugliest of ways -- based on rumor:

It has been pointed out here before that Fine takes dictation from Bruce Fisher TM and posts it here. This is tiresome and adds nothing to the discussion. In fact, it is disrespectful. Thanks for stopping by, Fine.

I agree, it is tiresome and adds nothing to the discussion to point out such things, especially if they are not true. It appears nobody even checked with Fine about the veracity of these reports before she was abruptly offed on account of them. Again, I agree -- how disrespectful.

Coincidentally, Fine "stopped by" for just about exactly two years -- her join date was Sun Jul 19, 2009.

Reminds me of a saying that some site administrators are going to wish they had heeded in the days to come: "Be good to the people you meet on your way up, because you will meet them again on your way down."

 

[RoseMontague]

Just a random thought. I was looking in the AAH forum for something else and found posts from this thread that were over a year old. Long as far as post counts go, yes I knew, but going on over a year? Wow, just wow.

Are these archived somewhere? It seems AAH has gotten smaller than I remember.

 

[RoseMontague]

Wow. And Fine was expelled in the ugliest of ways -- based on rumor:



I agree, it is tiresome and adds nothing to the discussion to point out such things, especially if they are not true. It appears nobody even checked with Fine about the veracity of these reports before she was abruptly offed on account of them. Again, I agree -- how disrespectful.

Coincidentally, Fine "stopped by" for just about exactly two years -- her join date was Sun Jul 19, 2009.

Reminds me of a saying that some site administrators are going to wish they had heeded in the days to come: "Be good to the people you meet on your way up, because you will meet them again on your way down."

Skeptical Bystander&#153; is not interested in finding the truth, in my opinion.

 

[LondonJohn]

I think it's worth repeating at this point that there are two separate issues worth of discussion in regard to this case: a) a discussion of what really happened on the night of November 1st/2nd 2007 in Perugia (aka "the truth"); and b) a discussion about the criminal trials of Knox and Sollecito (and Guede).

I'm primarily interested in the second of these discussions. And that's because I think not only that it's far more relevant to current events, but also because a search for "the truth" is esoteric, unprovable and semantic.

Remember that a criminal trial is not a search for the truth per se. It's an examination of whether certain defendants can be proven beyond a reasonable doubt* to be guilty of the crimes with which they are charged. And it has to be this way. It's the only way of holding people accountable in a fair criminal justice system. The inquisitorial system of criminal justice has been shown to be a failure. Like communism, it's one of those ideas which may sound good on paper, but which is completely unworkable (and indeed actively destructive) in application.

The only conceivable way in which "the truth" can even be attempted to be discovered in such cases is by way of separate judicial inquiries, or coroners' inquiries. Such inquiries do not concern themselves with assigning individual (or group) blame to a criminal standard. Coroners' inquests are solely concerned with establishing how and why someone died, and judicial inquiries are usually set up to establish what really happened so that lessons might be learned.


* Again, for clarity, this phrase means that guilt must be proven beyond all doubt that a reasonable person (i.e. a person of sound mind and normal world experiences) could reasonably hold. It does not mean anything along the lines of "guilt must be proven beyond a reasonable level of doubt" - which incorrectly implies that jurors are allowed to have a "reasonable" level of doubt (i.e. some doubt) and still vote to convict. In fact, if they have any doubt that is based on reason, they must acquit.

 

[LondonJohn]

They've quite obviously deliberately spoiled the bra-clasp so it can't be re-tested. DNA is gathered from crime scene material in cases that happened years ago all the time.

The prosecution cheating as usual looks so bad that you have to let R & A out of this lop-sided looking crime that gathers a neatness and believability to it when it's just pinned on the burglar Rudy who supplied all the real evidence.

When you do the case appears sane....

It does seem strange that the "crack" forensics team involved in this case seemingly don't know how to properly store a piece of textile/metal material with DNA evidence on it, without it deteriorating to the point of uselessness within four years. Surely this shouldn't be beyond the basic competence levels of "world-class" forensic scientists........

 

[RoseMontague]

I think it's worth repeating at this point that there are two separate issues worth of discussion in regard to this case: a) a discussion of what really happened on the night of November 1st/2nd 2007 in Perugia (aka "the truth"); and b) a discussion about the criminal trials of Knox and Sollecito (and Guede).

I'm primarily interested in the second of these discussions. And that's because I think not only that it's far more relevant to current events, but also because a search for "the truth" is esoteric, unprovable and semantic.

Remember that a criminal trial is not a search for the truth per se. It's an examination of whether certain defendants can be proven beyond a reasonable doubt* to be guilty of the crimes with which they are charged. And it has to be this way. It's the only way of holding people accountable in a fair criminal justice system. The inquisitorial system of criminal justice has been shown to be a failure. Like communism, it's one of those ideas which may sound good on paper, but which is completely unworkable (and indeed actively destructive) in application.

The only conceivable way in which "the truth" can even be attempted to be discovered in such cases is by way of separate judicial inquiries, or coroners' inquiries. Such inquiries do not concern themselves with assigning individual (or group) blame to a criminal standard. Coroners' inquests are solely concerned with establishing how and why someone died, and judicial inquiries are usually set up to establish what really happened so that lessons might be learned.


* Again, for clarity, this phrase means that guilt must be proven beyond all doubt that a reasonable person (i.e. a person of sound mind and normal world experiences) could reasonably hold. It does not mean anything along the lines of "guilt must be proven beyond a reasonable level of doubt" - which incorrectly implies that jurors are allowed to have a "reasonable" level of doubt (i.e. some doubt) and still vote to convict. In fact, if they have any doubt that is based on reason, they must acquit.

Not sure I follow you LJ but I am interested in the truth of what happened as well as the court reaching a judgment that reflects the truth.

 

[komponisto]

I think it's worth repeating at this point that there are two separate issues worth of discussion in regard to this case: a) a discussion of what really happened on the night of November 1st/2nd 2007 in Perugia (aka "the truth"); and b) a discussion about the criminal trials of Knox and Sollecito (and Guede).

I'm primarily interested in the second of these discussions. And that's because I think not only that it's far more relevant to current events, but also because a search for "the truth" is esoteric, unprovable and semantic.

Aieee, beware the fallacy of gray! The fact that we may not be able to be 100% certain about what happened at the house on Via Della Pergola doesn't mean we can't be 99% certain (let alone 95%, 90% etc.). And if we can be 99% sure of something, I certainly wouldn't call it "esoteric, unprovable, and semantic".

And similarly, the fact that we can't know every detail of what happened on Via Della Pergola doesn't mean we can't know the important facts, i.e. whether Amanda Knox and Raffaele Sollecito were there on the night Meredith Kercher was killed, or not.

* Again, for clarity, this phrase means that guilt must be proven beyond all doubt that a reasonable person (i.e. a person of sound mind and normal world experiences) could reasonably hold. It does not mean anything along the lines of "guilt must be proven beyond a reasonable level of doubt" - which incorrectly implies that jurors are allowed to have a "reasonable" level of doubt (i.e. some doubt) and still vote to convict. In fact, if they have any doubt that is based on reason, they must acquit.

Actually, what it means is "guilt must be proven beyond x% certainty (for some specific x such as 99), but we're not allowed to say that because people can't handle the fact that this means we tolerate 100-x% of convictions being wrongful. So we just pretend it isn't true and avoid thinking about it." Cf. this comment by thoughtful on PMF (Channel 2):

...reasonable doubt cannot be quantified, and above all it should not be quantified, because doing so goes against the values of our justice system. For instance, stating that if guilt appears 95% certain then one should vote for conviction can be equated with saying that our system is expected to convict 5% of innocent people, which is unacceptable.

I find this kind of thinking astounding and truly alien: does anybody really think that by not acknowledging the error rate in a justice system, we can make it go away?

(Also, "convict 5% of innocent people" is a misstatement. "Allow 5% of convicts to be innocent" is what is meant.)

 

[komponisto]

Skeptical Bystander™ is not interested in finding the truth, in my opinion.

No frickin' kidding.

(Love the ™; I think I'm going to start using it. I don't understand what their point is with Bruce Fisher. Are they trying to suggest that Bruce Fisher isn't a real person, but instead some kind of collective pseudonym, like Nicolas Bourbaki? If so, so what? What would that imply, exactly? It certainly makes a lot more sense here, since "Skeptical Bystander", being in no sense an accurate descriptor, can only be a brand name!)

 

[Katody Matrass]

Skeptical Bystander™ is not interested in finding the truth, in my opinion.

It is in fact quite astonishing to revisit that board and find how much it deteriorated into an outlet of cult-like mentality.

Fine simply pointed out a factual error in their beliefs - Massei mixed up trace numbers and there is contradiction in his motivation - it's quickly sorted out with the quotations from the trial testimony and the technical report.

They were baffled only for a moment, then they transcended that contradiction using faith and their belief is now that defense lied and falsified the quotations in an official appeal document.


Other amazing developments of the cult:

There is a conspiracy of mafia, free masons and scientologists to destroy Mignini - the campaign of dark forces to free Amanda is just a byproduct, a means to achieve that goal.

Famous FBI profiler John Douglas didn't simply conclude that Knox is innocent on his own. He has been hired by the all powerful family and PR juggernaut (and the mafia, of course).

Hellmann's independent experts have been taken control of by the forces of FOA. Their report have been dictated by phone straight from the USA.

It's most disappointing to see that also former JREF sceptics succumb to that conspiracy theories, crazy rationalizations and cultist behavior.

 

[LondonJohn]

Rose/Komponisto

I didn't mean my earlier post to imply either that a) I wasn't at all interested in getting as close as possible to the truth of what happened that night, or that b) a search for the truth is either impossible or unrealistic.

My point is that I think that "the truth" is a far, far harder (and more abstract) concept to work with than whether Knox/Sollecito should be convicted or acquitted. For the latter case, the courts have to weigh the evidence placed in front of them (and only the evidence placed in front of them), and determine whether the defendants can be shown beyond a reasonable doubt to have committed the crimes with which they are charged. Conversely, a search for "the truth" usually involves a fair degree of hypothesis and deduction - both of which are highly susceptible to error.

Take, for example, the case of Strauss-Kahn. It may forever be impossible to determine "the truth" of whether there was consensual sexual interaction between him and the maid, or whether a crime was committed. The only two people who know "the truth" are the suspect and the alleged victim. Nobody else will EVER know the truth, unless both these people agree to a single version of events (which obviously will probably never happen).

To me, therefore, the most important issue for debate in this case is whether Knox and/or Sollecito should be convicted or acquitted of the murder of Meredith Kercher. My view is that they should definitely be acquitted. And that leads to a second-order finding on the likely "truth" behind this crime - which is that Guede alone was responsible. But nobody will ever be able to say that they have definitively found "the truth" of this crime. Whereas I think it's possible to say with a pretty high degree of certainty that Knox/Sollecito should be acquitted (and to say with a lesser degree of certainty that they were not involved to any degree in the crime).

PS: On "beyond a reasonable doubt", let's be clear that this really does mean - to all intents and purposes - "beyond all doubt". The caveat is that the doubt has to be based on reason. And "reason" is broadly defined as "being feasible and possible in the real world, in the context of the case being discussed, in the view of an average person of sound mind and normal world experience".

 

[LondonJohn]

By the way, I personally think that some of the DNA evidence against Guede might be susceptible to a successfully challenge by his defence lawyers. Specifically, they might make some decent headway on the DNA on the handbag (purse) and possibly also on the DNA inside Meredith's body.

But - as I've also said before - I don't believe that these go any way towards constituting decent grounds for appeal for Guede. And the reason why I think this is as follows: 1) I believe that his hand print in Meredith's blood next to her body is completely unimpeachable evidence of his presence at the murder scene; 2) I believe that his behaviour after the murder (going dancing - seemingly without a care in the world - within a couple of hours, then fleeing Italy within two days) is totally incompatible with his version of events (that he was there but merely found Meredith bleeding and made an effort to save her). I therefore think that these two pieces of evidence ON THEIR OWN would convict Guede of the murder of Meredith Kercher.

So, in my opinion, anyone who "argues" that it's inconsistent or dishonest for people to claim that the DNA bungling means far more to Knox/Sollecito than to Guede is either ignorant or disingenuous in their "argument" (hi stilicho!)

 

[RoseMontague]

Does the "simultaneously" deposited blood/DNA on the sink contain a profile of a third female subject?

4. Occorre, infine, riportare un ulteriore dato scientifico che contrasta con la motivazione della sentenza.
Dalla lettura degli elettroferogrammi relativi alle tracce miste Knox – Kercher sin qui analizzate non è possibile escludere la presenza di una ulteriore traccia biologica riconducibile ad un terzo soggetto di sesso femminile.
La dott.ssa Stefanoni ha dichiarato:
<<non poteva escludere una terza persona perché si trattava di profili molto bilanciati>> (trascrizioni udienza 22 maggio 2009, pag. 222).
<<In quel caso potrebbe essere presente anche una terza persona sempre di sesso femminile che però le stesse caratteristiche presenti in questo misto>> (trascrizioni udienza 22 maggio 2009, pag. 229).
<<E lei nei profili genetici misti che riguardano la Knox esclude che ci fosse una terza persona? R – È proprio questo che stavo tentando di dire. Io non lo posso escludere proprio>> (trascrizioni udienza dibattimentale 22 maggio 2009, pag. 222).
<<Questi accoppiamenti diversi fanno sì che io posso includere altrepersone rispetto a quelle già presenti. Questo è un po’ il diciamo… ed in questo caso ovviamente si parla di compatibilità perché ci sono sicuramente gli alleli della vittima e della Knox ma, appunto, con questa combinazione possibile non si può escludere che ci siano altre…>> (trascrizioni udienza dibattimentale 22 maggio 2009, pag. 225).
La presenza di una terza persona non individuabile veniva rappresentata anche dal consulente di parte civile, dott.ssa Torricelli.

 

[komponisto]

I was going to make some pointed remarks on the evaporative cooling going on in idiotland, and the utter fatuity of Ganong's rationalizations for banning every single poster who doubts Knox and Sollecito's guilt, but I came across this piece of work from erstwhile JREF poster "zilcho" (if he's going to misspell my pseudonym, I can misspell his), which is probably of greater interest:

As far as I can understand it, Komposto [sic] and Dr Library Card have entirely misapplied Bayesianism in this case. Without citing, I recall that the story on Less Wrong woven by Komposto and picked up the middle school teacher is that the prior probability is set in stone by the rarity of women committing murder, the rarity of women committing sexual violence, and the rarity of three newly-found friends committing murder. That begins the entire exercise with a low probability and it simply gets worse from there.

I don't know who this "middle school teacher" associate of mine is, but I myself have never said anything about "the prior probability [being] set in stone" by anything. Quite the opposite indeed, because I've had to argue against people who are under the misapprehension that "the" prior probability "should" or "must" take into account this or that, which is not the case at all.

Your prior probability (which is what you should really say as a Bayesian, not "the prior") is determined by whatever information you happen to take into account as your background assumptions. In other words, what you decide to call a "prior" is strictly an ad-hoc convention for a particular discussion. One person's prior is another's posterior. You could equally well start with different assumptions and a different prior, and end up having the exact same discussion, with the exact same numbers at the end -- in fact, you should, because you should eventually end up taking all of the relevant information into account, regardless of what subset of that information you decided to start with.

They never include things such as the high probability of a cover-up at a murder scene, the high probability of that cover-up including a staged burglary, or high probability of a murderer guiding others to "discover" the victim, each of which happened in this case as though it were a textbook example.

Well, if zilcho thinks I haven't taken this information into account, why doesn't he come out and say how much he thinks it should raise my probability of guilt, then? As opposed to complaining about "misapplying Bayesianism" or making a silly argument that I'm somehow obliged to include this in the prior rather than updating on it to obtain a posterior.

Suppose I start with my prior of 0.001 (probably an overestimate) based on the rarity of the hypothesized crime; and then zilcho informs me that most murders involve a cover-up. How far upward should I update? How much evidence that Knox and Sollecito are guilty does that fact constitute?

The answer is zero. If the murder rate is small, then the murder-followed-by-cover-up rate is even smaller -- regardless of the proportion of the latter within the former. As in the old feminist bank teller story, zilcho has fallen victim to the conjunction fallacy. What zilcho needs is information that raises the probability of guilt, not information that raises the probability of something else given guilt.

(EDIT: How ironic that in the course of criticizing the explicit use of probability theory, zilcho commits exactly the kind of reasoning error that shows how important it is.)

Moreover, some elements are included and others are arbitrarily discarded. Guede left traces "everywhere" inside Meredith's room but left absolutely no trace on the lamp brought in from Knox's room. The five loci of mixed DNA outside the locked room are simply forgotten. Only elements inside the locked room are figured into the calculation except for the staged burglary (which they, of course, interpret as a real burglary). There is a foggy calculation of how far from the body a piece of evidence has to be found to be considered.

Again, zilcho has it wrong. Nobody has "forgotten" the mixed DNA in the bathroom; I just don't think it counts for anything. It's not substantially more likely to be found under the assumption of guilt than under the assumption of innocence, in my view. Sure, it's "figured into the calculation"; it just doesn't change the result much.

And yes, closer evidence is more important than more distant evidence. That's common sense -- or at least it would be, if not for the human obsession with the psychology of other humans, at the expense of "boring" biological traces left at the scene.

Practically the only use for this kind of statistical analysis in this case is the probability of five different loci including Meredith's and Knox's DNA without there being some kind of contact between the two when at least one of them was bleeding. The prior probability starts at roughly zero for Knox and Sollecito and one for Guede. Then the calculation for the five loci is figured in and the probability for Knox increases to something around a millionth of a per cent less than 100.

I don't know if zilcho is intentionally trying to be confusing by using the word "loci" here, but he's apparently not talking about the genetic kind, despite the proximity of the term "DNA". Actually, this whole paragraph makes no sense as written, and I have had to resort to context clues to guess that it is an attempt on the part of zilcho to express the fact that he thinks the bathroom DNA is strong evidence of guilt. Well, he's wrong.

I would hate to have these people doing a market or product analysis for the company I work for. Behaviour is completely ignored. In real life, consumers reach predictably for the product stored at eye-level, even if it is not the best value. In real life, criminals lie or try to distract the police.

Yep -- more confirmation of the guilters' obsession with behavior.

Of course criminals lie or try to distract the police. The point is that not everyone who behaves differently than you expect is "lying" or "trying to distract the police" -- let alone a criminal.

(You know who else "lies" and "tries to distract the police"? Witches.)

 

[Chris_Halkides]

false convictions and appeals

Aieee, beware the fallacy of gray! The fact that we may not be able to be 100% certain about what happened at the house on Via Della Pergola doesn't mean we can't be 99% certain (let alone 95%, 90% etc.). And if we can be 99% sure of something, I certainly wouldn't call it "esoteric, unprovable, and semantic".

And similarly, the fact that we can't know every detail of what happened on Via Della Pergola doesn't mean we can't know the important facts, i.e. whether Amanda Knox and Raffaele Sollecito were there on the night Meredith Kercher was killed, or not.




Actually, what it means is "guilt must be proven beyond x% certainty (for some specific x such as 99), but we're not allowed to say that because people can't handle the fact that this means we tolerate 100-x% of convictions being wrongful. So we just pretend it isn't true and avoid thinking about it." Cf. this comment by thoughtful on PMF (Channel 2):



I find this kind of thinking astounding and truly alien: does anybody really think that by not acknowledging the error rate in a justice system, we can make it go away?

(Also, "convict 5% of innocent people" is a misstatement. "Allow 5% of convicts to be innocent" is what is meant.)

komponisto,

One person here put reasonable doubt as 75%, which strikes me as very low. However, the advantage of putting a number on one's certainty is that it forces one to think about the level of false convictions, and this in turn forces one to rethink the appeals system in one's country. In the U.S. the burden of proof reverses, whereas in Italy it does not, as we see in the present case. Cameron Todd Willingham had many reviews of his case, but almost no reexamination of the inexcusably bad arson forensics that were done.

 

[Chris_Halkides]

extra alleles

Does the "simultaneously" deposited blood/DNA on the sink contain a profile of a third female subject?

RoseMontague,

I am not sure about the sink, but I have seen one of the electropherograms for Filomena's room, and there are extra alleles in it. DNA has no time stamp. Massei's attempt to say that Amanda's DNA was deposited when she washed off blood makes no sense. I will have to curtail my discussion here for a little while.

 

[pilot padron]

Arguing with one's self....and losing ??

By the way, I personally think that some of the DNA evidence against Guede might be susceptible to a successfully challenge by his defence lawyers. Specifically, they might make some decent headway on the DNA on the handbag (purse) and possibly also on the DNA inside Meredith's body.

But - as I've also said before - I don't believe that these go any way towards constituting decent grounds for appeal for Guede. And the reason why I think this is as follows: 1) I believe that his hand print in Meredith's blood next to her body is completely unimpeachable evidence of his presence at the murder scene; 2) I believe that his behaviour after the murder (going dancing - seemingly without a care in the world - within a couple of hours, then fleeing Italy within two days) is totally incompatible with his version of events (that he was there but merely found Meredith bleeding and made an effort to save her). I therefore think that these two pieces of evidence ON THEIR OWN would convict Guede of the murder of Meredith Kercher.

So, in my opinion, anyone who "argues" that it's inconsistent or dishonest for people to claim that the DNA bungling means far more to Knox/Sollecito than to Guede is either ignorant or disingenuous in their "argument" (hi stilicho!)

I was going to make some pointed remarks on the evaporative cooling going on in idiotland, and the utter fatuity of Ganong's rationalizations for banning every single poster who doubts Knox and Sollecito's guilt, but I came across this piece of work from erstwhile JREF poster "zilcho" (if he's going to misspell my pseudonym, I can misspell his), which is probably of greater interest:



I don't know who this "middle school teacher" associate of mine is, but I myself have never said anything about "the prior probability [being] set in stone" by anything. Quite the opposite indeed, because I've had to argue against people who are under the misapprehension that "the" prior probability "should" or "must" take into account this or that, which is not the case at all.

Your prior probability (which is what you should really say as a Bayesian, not "the prior") is determined by whatever information you happen to take into account as your background assumptions. In other words, what you decide to call a "prior" is strictly an ad-hoc convention for a particular discussion. One person's prior is another's posterior. You could equally well start with different assumptions and a different prior, and end up having the exact same discussion, with the exact same numbers at the end -- in fact, you should, because you should eventually end up taking all of the relevant information into account, regardless of what subset of that information you decided to start with.



Well, if zilcho thinks I haven't taken this information into account, why doesn't he come out and say how much he thinks it should raise my probability of guilt, then? As opposed to complaining about "misapplying Bayesianism" or making a silly argument that I'm somehow obliged to include this in the prior rather than updating on it to obtain a posterior.

Suppose I start with my prior of 0.001 (probably an overestimate) based on the rarity of the hypothesized crime; and then zilcho informs me that most murders involve a cover-up. How far upward should I update? How much evidence that Knox and Sollecito are guilty does that fact constitute?

The answer is zero. If the murder rate is small, then the murder-followed-by-cover-up rate is even smaller -- regardless of the proportion of the latter within the former. As in the old feminist bank teller story, zilcho has fallen victim to the conjunction fallacy. What zilcho needs is information that raises the probability of guilt, not information that raises the probability of something else given guilt.



Again, zilcho has it wrong. Nobody has "forgotten" the mixed DNA in the bathroom; I just don't think it counts for anything. It's not substantially more likely to be found under the assumption of guilt than under the assumption of innocence, in my view. Sure, it's "figured into the calculation"; it just doesn't change the result much.

And yes, closer evidence is more important than more distant evidence. That's common sense -- or at least it would be, if not for the human obsession with the psychology of other humans, at the expense of "boring" biological traces left at the scene.



I don't know if zilcho is intentionally trying to be confusing by using the word "loci" here, but he's apparently not talking about the genetic kind, despite the proximity of the term "DNA". Actually, this whole paragraph makes no sense as written, and I have had to resort to context clues to guess that it is an attempt on the part of zilcho to express the fact that he thinks the bathroom DNA is strong evidence of guilt. Well, he's wrong.



Yep -- more confirmation of the guilters' obsession with behavior.

Of course criminals lie or try to distract the police. The point is that not everyone who behaves differently than you expect is "lying" or "trying to distract the police" -- let alone a criminal.

(You know who else "lies" and "tries to distract the police"? Witches.)

Just cannot help but ask:

When you are reduced to arguing as above with an absent poster from another (apparently admittedly *superior*) Board....

1) Does it help give strength to your argument if you do so in front of a mirror ???
Or per haps with a picture of the absent poster taped to your mirror ??

2) Do you always *win* when you have to argue (with yourself) in this very strange manner?

3) Since the 'style'/content of arguing that you continue alone now is sooo strange;

3A) is this the *lively friendly skeptical* arguments that Mr Randi envisioned...

3B)... for 55,000+, and often 50 a day from one individual ????

(Bolding in above lengthy quotes mine)

 

[komponisto]

Just cannot help but ask:

When you are reduced to arguing as above with an absent poster from another apparently superior) Board....

1) Does it help give strength to your argument if you do so in front of a mirror ??? Or with a picture of the absent poster taped to your mirror ??

2) Do you always *win* when you have to argue (with yourself) in this very strange manner?

3) Since the 'style' of arguing that you continue alone now is sooo strange;

3A) is this the *lively friendly skeptical* stuff that Mr Randi envisioned

3B) for 55,000+, and often 50 a day from one individual ????

Arguing with an absent poster about the subject at hand beats arguing with a present poster about who the present poster should argue with.

 

[Chris_Halkides]

google based DNA forensics

BTW
The also 'piling on style' repeatedly regurgitated requests for guilter timelines with accompanying associations of 'inability' also seem senseless to me and little more than tag team echoing.
This because if one does not accept the Google based ToD, the very complete and very well debated previous PMF easily accessed timelines are in fact still very valid.
Furthermore even if one does accept Google's Gook, the to date silence on the matter from the principals in the case who are the only ones who matter is also underwhelming and grounds to ignore the questionably motivated requests.
This particularly if we need to conserve electrons here as earlier suggested by others

pilot padron,

Given your derision of "Google's Gook" research, what do you think of the "Google based" discussion of DNA forensics at PMF or TJfMK and why do you believe it?