Showing posts with label randomness. Show all posts
Showing posts with label randomness. Show all posts

Wednesday, October 8, 2014

From unaccustomed praise, to wind-blown leaves, to the mind-freeing scientific method.

This science bod was paid a big compliment today, probably the biggest he's been paid in his entire life. I'll spare you the details, dear reader, except to say that I try to let compliments, like brickbats, fly over my head, so as not to be too distracted by what other folk think of me and my peculiar mission in life. Which is what you may ask?

Fortunately, I had my camera and the pictures it took in a blustery Lyon this morning to demonstrate what turns me on, what gives me a buzz (recalling this site calls itself science buzz).

First, those pictures.

There I was with my dear wife, crossing a bridge that separates Old  Lyon from the Presque Ile (peninsula) that separates the Rhone and the Saone just short of their confluence. Suddenly a fierce wind  got up from downstream (the south) a short way from the confluence of the two rivers, which made the River Saone under the bridge become flecked with foam, seeming to flow against its N-S direction.


The river is flowing AWAY from me, despite those white-capped waves that seem to be coming towards the camera
Here are autumn leaves that have collected on the lee side of that wind coming from the right, with very few visible elsewhere.







Here's the same view a short while later, where a fresh gust of wind has caused a few of those leaves in the sheltered side of the footbridge to become airborne. A minute or two later they were back where they started. Why?

My wife offered  an entirely reasonable and commonsensical explanation for what you see in these pictures. "The wind has blown the leaves into that area on the right".

But how can that be, when the wind if blowing right to left in these pictures?

An entirely different explanation occurs to me, one which is apposite at the present time, given attempts by some in Shroudology to re-invent chemistry as 'stochastic physics'. (Chemistry has always featured a random element where molecular collisions, productive or unproductive are concerned, but does not generally try to track the path of individual molecules, being content merely to explain and predict the end-result in gross terms of billions of such collisions).

The leaves are not being blown into that sheltered side. The leaves are being blown into the air, and are settling all over the place. But when  leaves just happen by chance  to land in that sheltered spot, they tend to stay put, whereas leaves that land elsewhere, in less sheltered spots, tend to get lifted back into the air. Bit by bit, leaves tend to fall at random into the sheltered spot, so with time, one sees more and more leaves there, and fewer and fewer leaves elsewhere. The leaves were NOT blown preferentially into that sheltered spot.

Methinks there's a parable there, not only for chemistry, but biochemistry too (my specialist subject).

I got a buzz this morning for seeing macroscopic objects (dead leaves) behaving like atoms and molecules.

I get a buzz too from looking at the claims made by fellow 'Shroudies' and thinking to myself: "Suppose the process of image formation were not as 'intuitive' as one might think. Suppose the end-result came about by processes that are not intuitive, dare one say "commonsensical". Suppose the TS image has a thousand possible explanations, only one of which is correct? How can one set about determining which is the correct explanation?  For some of us, the buzz comes from meeting that challenge.

For scientists like myself, there is a simple answer to that question (which may or may not lead to the correct end-result, at least in one's own lifetime).  It requires no great feats of imagination or intellect - one must simply get experimenting. It matters not a jot whether the initial working models are right, or possibly right, or indeed possibly or even probably wrong.

The important thing is to keep EXPERIMENTING.  It keeps the mind (reasonably) free of preconceptions  and dogma. It makes one more amenable to ideas and possibilities that take time to incubate.

Addendum Friday October 10

 Here's a piccy I took yesterday morning at the end of the tapering spit of land midstream where the Rhone (left) and Saone (right) come together a short way south of Lyon city centre.




And here's your blogger at that same precise spot, whose troosers don't always look like that (please believe me when I say there's still that fierce wind blowing from the south that is REVERSING the current, at least on the surface, with dead leaves and other detritus being gently moved UPSTREAM).



No, I don't have ambitions to be the next Pope, practising for an appearance on the balcony at St.Peters.  The gesture is a standard one I routinely employ when standing as so often I do at the confluence of major rivers. It allows my small band of admirers (and larger army of detractors) to deduce the direction from which the two riversconverge.

Afterthought (for Hugh Farey in Comments in the first instance ): I do in fact have a short video clip, showing the behaviour of those leaves on the windy bridge, which I'm studying at the moment to try and decide how much of the shift to the wind-sheltered side is random, v. how much was assisted by Bernouilli effect, eddies, vortices etc. etc. (given the wind was in the wrong direction to blow them there directly). Sadly the video will not upload to this blog (despite there being a button that's supposed to make that possible). Here's a screen grab still image to be getting on with.




I've now learned to clip videos in MS Movie Maker, thereby cutting my windswept wife out of the picture, and still my now diminished 10 second video file fails to upload, despite an on-screen icon that suggests it's supposed to be happening (shame about the static progress bar). Don't you just hate clunky software that doesn't work, and, more to the point, keeps you hanging around for 10-15 minutes to realize it does not work.

 My loathing for all things to do with Californian software is exceeded only by the French public transport system, once the envy of Europe, and now in free fall.

Monday, June 16, 2014

Thibault Heimburger is correct - Shroud photomicrographs lend no support to the notion of a 'stochastic' imaging mechanism.




The difference between those areas within the blue and red rectangles may possibly have theoretical significance as regards the mechanism of imaging (stochastic v deterministic, if you'll pardon the jargon). Why? Read on...



I won't pretend that I understand the papers by Giovanni Fazio and his colleagues in Sicily, claiming that the Shroud image is the result of 'stochastic' processes, ones that rule out certain image-forming mechanisms, and render others more probable  (see link below).

So what are stochastic processes? One could give a textbook answer and say they are the opposite of "deterministic" ones, that they incorporate an element of randomness, as distinct from being totally predetermined. The analogy that is generally given is that of rain falling on a pavement. Gradually a pattern of drops develops, but the location of each drop is not pre-determined, but random.

 Here's another attempt to convey the difference between the situations, free from as distinct from incorporating a random element, cribbed from internet image files: chess v snakes and ladders (the latter requiring dice)





So how is that relevant to the Shroud image? Again, I don't pretend to follow the authors' line of reasoning, but gather it's related to what's been called the 'half-tone effect', a topic I was discussing not so long ago. The half-tone effect is a shorthand term for a peculiar feature that is claimed for the Shroud image, namely that regions of  differing image intensity do not have linen threads and fibres of differing colour intensity. There are either uncoloured fibres, or fully-coloured ones with no in-betweens. A region that looks dark has a higher proportion of coloured to uncoloured fibres  compared to one that looks pale. Think of it if you like as comparable to analogue versus digital stereo. The analogue audio signal can take a whole range of values across a smooth continuum, whereas the digital signal is simply a series of binary digits, either 0 or 1.

So where does Thibault Heimburger MD, Paris-based French physician and member of the Shroud Science Group enter this story? Some might be surprised to find his views being favourably received on this site, given there is so much on which we differ, notably the contact scorch hypothesis (one that TH rejects). But that does not mean he's wrong - or right- on everything, far from it, as my follow-up to a recent comment on his on shroudstory.com will now show.

TH appeared on the recent thread, the latter flagging up the presence of a Fazio et al  paper recently published in Mediterranean Archaeology and Archaeometry.   He queried the claim (or supposition?) that yellowed  fibres were randomly distributed across Shroud image-bearing regions, stating that they could appear together in bundles   (see copy/paste below). Were that correct, it would deal a devastating blow to any theory that required the coloured fibres to be randomly distributed (though occasional clumping is not totally ruled out, albeit being of low expected frequency).

I recalled that TH had recently acquired from STERA's Barrie M.Schwortz a collection of Mark Evans photomicrographs of Shroud  magnified close-ups, e.g. x64, and have just spent the last few minutes perusing them, and selecting one in particular for enhancement in my MS Office Picture Manager.

Here's one in particular that backs up TH's claim for "bundles".




That's an "as-is" image, from the TS nose region, magnified x64.

It seems fairly clear that one has two adjacent threads, one comprising mainly yellow fibres, the other largely uncoloured ones. The chances of that happening via random processes must surely be vanishingly small.

Here's the same picture after adjusting brightness and contrast that makes it easier to see the difference.





Sorry, Dr.Fazio, but I think your focus on supposed "stochastic" processes is simply unwarranted, and have TH to thank for making that point. Apologies to the latter for possibly pre-empting him, had he been intending to publish pictures from the same Evans archive.

But there's a sting in the tail for TH: juxtaposition of coloured v non-coloured threads, such as we see above, is I believe entirely consistent with contact-scorching, e.g. from an apposed heated metal template, and is difficult if not impossible to fit into any other of the proposed models that I'm aware of, especially ones that involve radiation of some sort, or diffusion of gases.

Copy/paste of TH comment. 


June 14, 2014 at 4:10 pm 
In my opinion this paper needs to be read carefully.
I do not not if the term “stochastic” is the best one.
The main problem comes from the fact that the surface distribution of the color in the Body image areas is not easy to describe.
With the help of the analogy with the effect of low doses of radiations in a large population, one can understand what the authors have in mind.
It is true that low doses of radiations have the properties described by the authors:
1) There is no minimal level of radiations (except zero, which does not exist) without effect
2) The number of tumors increases with the time/level exposition
3) it is impossible to predict who will have a tumor in this population.
This is only an analogy.
Is the TS image color distribution like a “stochastic “process ?
In my opinion, not exactly.
Why ?
It is true that the color distribution in a given image area depends only on the number of colored fibers having the same density (+/- 10%) and not on the density of the fibers (the half-tone effect).
But the colored fibers are not randomly colored.
In a colored thread, there are BUNDLES of colored fires adjacent to bundles of uncolored fibers
If there is a stochastic process, it is at thread level and not only at fiber level.
In other words, a stochastic process at fiber level does not explain the fact that the colored fibers are bulked together in bundles adjacent to uncolored bundles of fibers in a given image thread.

Update: Tuesday 17 June.  One of the difficulties I have with the proposed  (or should that be "prescribed") stochastic mechanism, apart from the problem already noted on overt non-randomness in pigment distribution, is the paucity of chemical detail. The thesis appears to depend primarily on a blend of physics and statistics, leaving the reader to fill in the chemical gaps as best he can with little more than hints re the essential chemical detail. Matters are not helped by the authors' carelessness with the chemical detail where, rarely, it does make an appearance. For example, they cite the active oxygen species-  singlet oxygen - as playing a role at some point or in some instances, but with precise details being omitted. But they refer to singlet oxygen "atoms" or to it being a "free radical". It is neither. There is no such thing as a singlet oxygen atom. It is only intact oxygen molecules, O2, in which  bonding electrons can be promoted to the excited singlet state. No, it's not a free radical either, despite the chemical reactivity of singlet oxygen, in which all the electrons are spin-paired (free radicals by definition have one or more unpaired electrons).


Despite the simplicity of its chemical formula, oxygen (O2) is a peculiar molecule. We tend to think of it as being chemically reactive, based on its role in combustion and respiration, but the oxygen around us is reluctant to enter into chemical reaction, except slowly in most instances, until it is activated in some way. In the case of combustion, that requires raised temperatures to give the reaction a kick-start. In the case of respiration it is the iron-containing cytochromes of the mitochondria that are needed as catalysts.

Curiously, it is ground state unreactive oxygen that is the free radical, more correctly a diradical (with two unpaired electrons per molecule) and with three degenerate molecular orbitals for the so-called p-type bonding electrons. Thus the description of ground-state O2 as triplet-state oxygen. In singlet oxygen the two unpaired electrons are promoted to an excited higher-energy level, but in the process become paired.

Red spots represent electrons able to flip between two energy levels. Note the two unpaired electrons in ground state (triplet) oxygen at the top. The arrows show they have "parallel spins". In singlet oxygen the corresponding electrons are paired, i.e. have opposite spins.


So how does one convert triplet to singlet oxygen? There's a very simple and pretty way of demonstrating the conversion in the laboratory. One bubbles chlorine gas through an alkaline solution of hydrogen peroxide in the dark.
The reaction mixture becomes chemiluminescent, i.e. light-emitting, giving off a red glow. That's due to production of singlet oxygen which then slowly reverts back to triplet oxygen by a complex mechanism, emitting visible red light in the process. The typical survival time of singlet oxygen as a gas can be an hour or more - in contrast to most free radicals that exists for a fraction of a second.

In everyday life, singlet oxygen can be formed more silently and surreptitiously by processes that involve sensitization to visible or uv light by the presence of dyes and/or other molecules, the classic one being methylene blue. It then attacks and degrades biomolecules (lipids, proteins etc) in the vicinity, the process being referred to as photodynamic action., Plants have to protect themselves against singlet oxygen formed "accidentally" during photosynthesis in chloroplasts. It's been suggested that accessory pigments like carotenoids are there to mop up singlet oxygen.

Singlet oxygen may be generated in the cytotoxic armoury of some killer immune cells that stop invading bacteria in their tracks. Sadly it has also been implicated in some cancer aetiology. Singlet oxygen is a big subject. Anyone proposing a role for it in Shroud studies should first take the trouble to get acquainted, if  not with its electronic configuration in molecular orbital terms, at least its broad chemical character and classification, certainly before referring to it as a "free radical", or oxygen "atoms"  or to others as "self-styled scientists".

Incidentally, anyone who has worked with singlet-oxygen mediated  reactions, as I did in Philadelphia, 1970-72, will know that far from inducing yellow coloration in organic matter, the end-result is usually to bleach it. It's not rocket science. Colour in organic compounds is often due to conjugated double bonds
( -C=C-C=C- etc). Singlet oxygen, by adding across the double bonds to form dioxetanes etc, destroys one or more double bonds that comprise a delocalised system of electrons resulting in LOSS of colour.

Conjugated diene (left side) reacts with singlet oxygen (above first arrow) to form an adduct that has only one double bond instead of two, with less absorption in visible or near uv spectrum. (In other words, the tendency is for singlet oxygen to bleach, not to make yellow).

Bleaching of fabrics, paper etc by light has been ascribed in some cases to self-sensitized production of singlet oxygen.

Here's an ad' for teeth whitener.

Note the ranking of 'active oxygen' species according to bleach power. Note which one tops the chart. Yes, it's singlet oxygen. It's especially effective the accompanying words say against the stubborn yellow stains of tetracyclin antibiotic.

PS: I see that Fazio et al are citing Mills et al* as the proponents for a role for singlet oxygen, which they describe as imparting "energy" to the fibres, resulting in yellowing  while being more circumspect themselves, They prefer to play safe, and refer to an "unknown source" of energy triggering off immediate and subsequent lag-phase reactions to produce yellowing maybe years or decades later. Would it be uncharitable to suggest that a hypothesis that depends on an "unknown" source of input energy hardly ranks as a hypothesis, or at any rate a scientific hypothesis?  Hypotheses in science have to be testable. How can one hope to test a hypothesis that is based on an unknown source of energy? 

* Mills, A.A. (1995):  Image formation on the Shroud of Turin. The reactive oxygen
intermediates hypothesis.  Interdisciplinary Science Reviews, vol. 20, 319 - 326.