posted 11-20-2006 03:32 PM            

I'll tell you this up front: Our PD used a DOS based Stoelting several years ago, and we had problems with it you couldn't seem to fix (e.g., a top pneumo tracing that appeared when it felt like it). I wasn't an examiner back then, so I can't give you much more, but when I purchased my Lafayette (or rather the PD did), we didn't even consider Stoelting because of the poor service reports (substantiated by our experience). What has changed that might sway the check writers in the future?

posted 11-20-2006 04:17 PM            

Why should examiners choose a Stoelting?

1. The hardware is the most technologically advanced on the market. We use 24 Bit Analog to Digital Conversion on all 8 channels. All sensors and plug-ins are all medical grade, which are approved by the FDA. We also utilize a true USB interface, not an imitation USB (an internal RS-232 converter) for faster communication with your computer.

2. Our software is the only software that quantifies reactions for each channel (including the PLE) for the examiner when they numerically score their charts. We even show you where the software gets the information, and the user has the option to apply different rules (i.e. changing pheumo RLL measurement from 10 secs to 15 secs).

3. Users of the CPS II also get the Exclusive Utah Probability algorithm. Our algorithm has been shown in repeated studies to be the most fair and balanced, especially regarding verified truthful subjects.

I have many other reasons, but I will not put all of my cards on the table yet...

posted 11-20-2006 06:15 PM            

quote:

---

The hardware is the most technologically advanced on the market.

---

Everybody says that. What makes it true?

quote:

---

All sensors and plug-ins are all medical grade, which are approved by the FDA.

---

Why should I care?

quote:

---

We also utilize a true USB interface, not an imitation USB (an internal RS-232 converter) for faster communication with your computer.

---

Why do I need more speed? What am I missing now?

quote:

---

2. Our software is the only software that quantifies reactions for each channel (including the PLE) for the examiner when they numerically score their charts. We even show you where the software gets the information, and the user has the option to apply different rules (i.e. changing pheumo RLL measurement from 10 secs to 15 secs).

---

What do you mean exactly? How does it ""quantify" reactions? Lafayette measures line length as you describe.

quote:

---

3. Users of the CPS II also get the Exclusive Utah Probability algorithm. Our algorithm has been shown in repeated studies to be the most fair and balanced, especially regarding verified truthful subjects.

---

posted 11-20-2006 07:50 PM         

Barry: Talk it up! The forum needs a diversion while DoDPI's shredders work overtime! (But it's way too late.)

posted 11-20-2006 09:48 PM            

Shawn may even sell a few systems if he plays his cards right.

posted 11-20-2006 10:48 PM         

[This message has been edited by stat (edited 11-20-2006).]

posted 11-21-2006 03:03 AM            

I don't mind folks from the instrument manufacturers reading and monitoring this forum, but I think I'd rather they remain in lurk mode. Responding to direct questions seems OK, but I would really hate to ever feel as if I were being "marketed" or "advertised" in this forum.

We work in a profession that is sometimes hampered with proprietary and parochial loyalties, not to mention financial, and that gets us not always thinking completely objectively.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 11-21-2006 09:04 AM            

I liked DOS. I miss it. It was a lot more logical than what we now have (where everything works in the background, hidden from view of those of us with a need to know the hows and the whys of what the system is doing). There is no excuse for not being able to make the CPS system work as it was supposed to have.

Ray,

Close your eyes. This is going to turn into a marketing "event" given my questions.

All (you hypersensitive folks),

I asked pointed questions because I want serious answers. I expect puffing from salesmen, but I hope we won't see that here. The CPS system boasts that it is science based, as is its scoring system and criteria, etc. We've had discussions about those topics here, some of which left unanswered questions. This is a great opportunity for somebody to give us the "whys" they do what they do - and more importantly - why that should be important to us. I'm not conceding Stoelting has it all right, but I expect a decent discussion could result.

posted 11-21-2006 10:30 AM            

Barry I would be happy to answer all of your questions, but I will not do it in this format. Our toll free number is 800.860.9775.

posted 11-21-2006 11:52 AM            

Without trying to be too rigid or hypersensitive. I think its fine to respond directly to some of Barry's direct questions in this forum, though I understand your wish to do some of that through back-channel email. I'd like to hear some of the answers myself. The current CPS II system has an impressive feature set and impressive pedigree with the folks in Utah. There is a balance to achieve, in forums such as this, between providing information and proffering hype. As a general rule, any information based on data or research, and empirically supported principles should always be welcome. Its just that I'm always a little annoyed when I sit in trainings that are actually marketing sessions. Keeping this forum free of proprietary objectives, preserves the space for discussions about important things like professional practice, science, and mounting guns on motorcycles.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 11-21-2006 12:00 PM            

Do you want me to answer my own questions and you fix the errors - if I make any? I didn't ask those questions because I don't know the answers. I asked to get you to talk about what Ray mentioned above.

posted 11-21-2006 01:44 PM            

That is one of the reasons I was so blunt. In my conversations with the manufacturers, some are quick to dismiss others as just wanting to sell instruments. In other words, don't believe what X has to say since he has a dog in this fight.

Dr. Raskin isn't shy about saying CPS is the best system out there (he and Dr. Kircher created it), but he takes it a step beyond that and will explain why that it true (in his humble opinion) - using a science-based argument.

posted 11-21-2006 02:32 PM            

i really like this unit and have no complain with stoelting service.

i also gets great resoults with this great unit.

posted 11-21-2006 03:09 PM            

The problem is, what are "great results," and what makes the unit "great"? How do you know? Would you have better results on a different system because that company records, for example, the EDA channel better.

"Great" is a conclusion. What is the evidence? The guys with CVSA boast great results.

Do you get plunging EDAs on a Stoelting? Axciton boasts they never do. Another CPS guy said that's because Axciton filters the data - a no-no in his book. Which is better? Why?

posted 11-21-2006 03:20 PM            

Everybody says that. What makes it true?

To the best of my knowledge we are the only company that uses 24 Bit A/D conversion on all of our channels. Lafayette and Axciton do not make there technical specifications public (I looked everywhere). Limestone touts their 16 Bit A/D in all of their advertising, and claims other companies use 12 Bit. You need a minimum of 22 Bit A/D to capture the entire range of EDA. We also sample all of our data at 60Hz (60 times per second), again no other company publishes this information. If you were to open our sensor box and compare it other instruments, you would also be able to see why I made the infamous claim of being the most technologically advanced.

quote:
--------------------------------------------
All sensors and plug-ins are all medical grade, which are approved by the FDA.
--------------------------------------------

Why should I care?

For all practical purposes you probably should not, BUT we use the same quality of sensors and plugs-ins that you would find in your local ER. (which cost considerably more than analog polygraph type sensors) to ensure that users are getting the highest possible quality sensors (most accurate tracings) for their investment. Our company also introduced the silver-silver chloride eletrodes; which are superior in collecting EDA, when compared to the old style finger plates. Stoelting could have used the AgAgC1 exclusivley, but we elected to allow other companies to offer this innovation to their customers.

quote:
--------------------------------------------
We also utilize a true USB interface, not an imitation USB (an internal RS-232 converter) for faster communication with your computer.
--------------------------------------------

Why do I need more speed? What am I missing now?

Right now probably nothing. But again, Serial Bus Ports and RS-232 converters(imitation USBs) are outdated and inferior technology, and are more prone to crashes. Most users would not see a differece, but some users want the most advanced hardware for their money.

quote:
--------------------------------------------
2. Our software is the only software that quantifies reactions for each channel (including the PLE) for the examiner when they numerically score their charts. We even show you where the software gets the information, and the user has the option to apply different rules (i.e. changing pheumo RLL measurement from 10 secs to 15 secs).
--------------------------------------------

What do you mean exactly? How does it ""quantify" reactions? Lafayette measures line length as you describe.

I sent Barry an email that contained a screen shot of our numerical scoring screen. By quantifying reactions I mean that the CPS II will give you a value (range is 0-99) listed at the top of each question region for each channel you are scoring. The Utah folks invented the computerized polygraph and wanted to offer an aid for examiners to improve the efficiency, reliability, and accuracy when examiners score their charts. This is exclusive to the CPS II, and again supports the claim of the most technologically advanced instrument.

quote:
--------------------------------------------
3. Users of the CPS II also get the Exclusive Utah Probability algorithm. Our algorithm has been shown in repeated studies to be the most fair and balanced, especially regarding verified truthful subjects.
--------------------------------------------

I don't understand what you mean by "fair and balanced." How can you be "especially" fair and balanced with the truthful? Do you mean, in short, symmetrical cut-offs?

Referring to (Dollins et al., 1999) where they used verified cases to compare decision accuracy of CPS, Polyscore, Axciton, and Identify. Our algorithm correctly identified 91 percent of the deceptive cases and 90 percent of the truthful cases. Other algorithms correctly identified 98% deceptives and only identified between 79-73% percent of the truthful cases. Based on this study, we have far fewer false positives, and are "more balanced" and objective than other algorithms on the market. Simply put the other algorithms are more biased against the truthful. It is also important to note that in this study, the software that APL created for us to allow our algorithm to read Axciton charts was 1 second late on EDA, meaning that our algorithm missed 1 second of data; which Dr. Raskin and Dr. Kircher contend would have allowed our algorithm to perform even better.

posted 11-21-2006 03:32 PM            

quote:

---

You need a minimum of 22 Bit A/D to capture the entire range of EDA. We also sample all of our data at 60Hz (60 times per second), again no other company publishes this information.

---

posted 11-21-2006 05:19 PM            

quote:

---

To the best of my knowledge we are the only company that uses 24 Bit A/D conversion on all of our channels. Lafayette and Axciton do not make there technical specifications public (I looked everywhere). Limestone touts their 16 Bit A/D in all of their advertising, and claims other companies use 12 Bit. You need a minimum of 22 Bit A/D to capture the entire range of EDA. We also sample all of our data at 60Hz (60 times per second), again no other company publishes this information. If you were to open our sensor box and compare it other instruments, you would also be able to see why I made the infamous claim of being the most technologically advanced.

---

This is exactly why I'd rather see the instrument manufacturers lurk. Its advertising, or campaigning, not science to devote this time to comparing your box to the others.

Lets, be clear about the fact that as a Stoelting rep you do not have the objectivity or freedom to say anything other than the above.

Whether they make specifications public (which would be nice) or not, Lafayette and Axciton are marketing 16 bit systems.

I would rather hear you describe your equipment more accurately, and spend less time talking about the others. If you have well-referenced commentary on the underlying science as employed by others, then that is fine with me.

But you have not done that.

Now about this,

quote:

---

You need a minimum of 22 Bit A/D to capture the entire range of EDA.

---

That is nonsense. Try to prove it.

What is the entire range of EDA? Are you referring to the range of possible resistance or conductance values?

An 8 bit system would scale input data along a Y axis compose of 256 gradations or units. If you had a possible EDA range of hypothetically 20 thousand to 9 million ohms, the max value is charted at the high end and min value is charted at the low end, the rest in between.

A 12 bit system could could chart the same input data values as an 8 bit system, only it would spread those values across a Y-axis scale of 4096 gradations.

Of course, a 16 bit system charts data across of Y-axis of 65,000 some-odd units. But the units are simply fractions of the total range. In computer scoring systems, the measurement units themselves are algebraically canceled - leaving only dimensionless values. I believe, the CPS manual describes how they compute standard or z-scores, and Kircher and Raskin (2002) described their within-subject standardization in vague detail.

In the same way would could call a linear mile a mile, or chop up that mile into 1760 yards, or 5280 feet. The linear distance is the same – you are simply chopping it up into smaller units.

The number of bits, is simply an issue of granularity or smoothness. It has primarily to do with visual impressions on the examiner – and the kind of gee-whiz shiney-object marketing hype that gets potential customers bobbing their heads up and down.

And this,

quote:

---

We also sample all of our data at 60Hz (60 times per second), again no other company publishes this information.

---

Where is that published? Kircher and Raskin (2002) describe the CPS system as sampling at 384 Hz – that's 384 cycles per second – across the X-axis (time-scale). But wait, they also describe that they compute and record the average of ever 48 samples for Pnuemo, EDA and movement channels – meaning they effectively sample at 8 cycles per second. They reported they compute and record the average of every 8 samples for BP (necessary due to the inherently higher frequency data) – meaning they effectively sample BP/cardio at 48 cycles per second.

Kircher and Raskin (2002) further described how they extract cardio features by creating a trend line of the sums of all samples for each one-second interval, and interpolating between those one-second summed values. This is probably similar to the second-by-second averaging procedure, for BP and Fingertip pulse amplitude, which they described in 1988. In 1988, they did indicate they sampled at 10ms (100 times per second) for cardio, and 100ms (10 times per second) for respiration and EDA. They also described using a step-wise averaging procedure to smooth data values in EDA, and interpolating data values for artifacted data.

And more,

quote:

---

I sent Barry an email that contained a screen shot of our numerical scoring screen. By quantifying reactions I mean that the CPS II will give you a value (range is 0-99) listed at the top of each question region for each channel you are scoring. The Utah folks invented the computerized polygraph and wanted to offer an aid for examiners to improve the efficiency, reliability, and accuracy when examiners score their charts. This is exclusive to the CPS II, and again supports the claim of the most technologically advanced instrument.

---

I think you are referring to what CPS calls relative response magnitude, which appears to be nothing more than a quantile score or cumulative distribution function of the standardized values. There is nothing complicated or unique about that, and I'm unsure why Stoelting would cause confusion by providing a new label for a common mathematical procedure. Moreover, it should be emphasized that the quantile and RRM score is impossible to interpret according to any well understood or well described decision thresholds or other mathematical procedures. So, while its and interesting set of numbers, decisions based on them are again arbitrary.

And,

quote:

---

It is also important to note that in this study, the software that APL created for us to allow our algorithm to read Axciton charts was 1 second late on EDA, meaning that our algorithm missed 1 second of data; which Dr. Raskin and Dr. Kircher contend would have allowed our algorithm to perform even better.

---

I read about this, and it does seem to appear that there were anomalies in the data provided, but I don't recall that people ever succinctly identified or define the error. It would seem that if we were sure the data were off by one second, it would be fairly expedient to correct that before proceeding.

Please understand, I have the utmost respect for the work of Kircher and Raskin, and others, and greatly enjoy reading and studying there published works. Its just that its fairly easy to get into marketing hype and drift from science.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 11-21-2006 05:52 PM            

What I meant by entire range of EDA was the ability of an instrument (A/D converter) to pick up any range of signal from a subject. A 24 Bit converter covers the entire spectrum of possible values that each person's unique physiology might bring to the table. Bottom line: 24 Bit A/D provides higher levels of precision.

In regards to your comment about my blathering about other instrument companies: I wished I could get my hands on any underlying science / published research that validates the whats, whys and hows of what they do. It would make my job alot easier!

I apologize, but I do not have the time right now to address your other comments.

Until next time...

Shawn

posted 11-21-2006 08:06 PM            

Also, I don't know what strabismus is.

I do enjoy a good argument.

And I thinks its obvious that more bits = greater precision - but that's a different measurement issue than the range of the data to be measured. The range of data is really a feature of individual variability, the nature of the data to be measured, and the transducer that produces the signal. The ATD converter simply maps that signal onto the datastream, across an X-axis (which is reflected in the sampling rate or time scale) and Y-axis (in which the minimum possible transducer value is assigned to the lowest possible scale value and maximum possible transducer value assigned to the highest possible scale value). All the values between the maximum and minimum are essentially chopped up into however many units of precision are possible based on how many bits are employed by the ATD converter.

While more bits are generally better, it is an oversimplified mistake (or misrepresentation) to assume that more bits is like a longer measuring tape. More bits is like more little lines on the same sized measuring tape - just closer together - that's why its more precise. If their were more lines, but spaced equally, then the tape would be longer - but that's where we cross the gap from physical metaphors to conceptual metaphors.

Additionally, the designers of CPS seems have realized that more data doesn't always mean more accuracy. So they don't use the full capability of the converter they use, and they average their data to second-by-second cardio trends anyway (that's a defacto rate of one sample per second) - only they have effectively stabilized random measurement error through the use of multiple averaged measurements. This places their assumptions on much more solid ground when they assert that the values they are feeding to the scoring algorithm are representative of the true activity within the subject. This is not any different than the procedure employed by any of the other manufacturers, and it is arrogant, inaccurate and pretentious to suggest otherwise.

However, there are still potential artifact problems, for which the Stoelting software offers a good set of features for managing/editing/removing.

I love the fact that Kircher, Raskin and others have published descriptions of their work. I think its a mistake to surrender to technology that we don't understand.

I have some technical questions about the sum of squared deviations procudure for pneumo measurements - described in Kircher and Raskin (2002), but I can't articulate them right now 'cause my head hurts.

I'd love to have more nuts and bolts discussions. I'm not sure that UBB does subforums.

Stat,

As with other technologies, much web technology is full of hype. A quick look around the Internet reveals that many commercial websites run secure web forums using phpBB - price $0.00 - and $48.00/mo for webspace is outrageous unless you need a dedicated server (doubtful). I suspect you were being marketed.

I'm with Barry - DOS was great.

However, I like Macs too, and I'm currently working on a Linux box. Someday soon I'll work on getting the Limestone software running in an emulator on the Linux. It will be interesting to see the outcome of the recent M$oft-Novell agreements and subsequent reaction from the GPL community.

Alright, I'm off to rewind my propeller.

Have a safe holiday everyone.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 11-21-2006 08:21 PM         

Stat,

I'm with you. But what I want to know is this: When those guys iron their socks, how much starch do they use?

Honest to God, they make a great argument for analog instruments!

posted 11-22-2006 09:33 AM            

A 22-bit A/D varies 4 units per .001 microSiemens of conductance with a system designed to record from 0 - 100 microSiemens. Most people have tonic levels of SC that range from 1-10 microSiemens, but conductance can approach 100 microSiemens in some subjects.

We didn’t raise the issue of precision of measurement – Limestone did.

A 16-bit A/D will vary less than 1 unit (.065) per .001 microSiemens of conductance. Given the inherent noise in A/Ds, a change of perhaps 6.5 units from an A/D may be reliable. Therefore, a 16-bit system is capable of detecting a change of about 0.1 microSiemens. The standard in the field of psychophysiology is to have sufficient precision to reliably measure changes as small as 0.02 microSiemens. Changes less than .02 microSiemens are considered noise. With CPS-I we used a DAC to provide a bucking voltage that ‘centers’ the signal from a subject, which then was amplified and fed into the 16-bit A/D to provide more precision of measurement. With the 24-bit A/D, we don’t need DACs to bring the signal from the subject into the range of the A/D. We can reliably register small changes in conductance regardless of the subject’s tonic level of conductance.

In the Dollins et al. “bakeoff” study, the ASCII data we received indicated that every test question began exactly on a 1-second boundary. The probability that 30-40 questions per subject for each of 100 or so subjects would occur only on exactly 1-second boundaries when samples are stored at 60 Hz is zero. Visually, event marks often occurred in the rising portion of the electrodermal response, so it appeared that the event marks often were late by one or two seconds. Since CPS ignores electrodermal responses that begin within 0.5 seconds of question onset, proper placement of event marks is critical for CPS. There was no way for us to know where the event marks should have been placed because subjects sometimes do show spontaneous electrodermal responses just prior to question onset or within 0.5 seconds and those responses should be discounted.

posted 11-22-2006 10:58 AM            

I don't think I could ever be more busy than I am now, but believe it or not - this board is a great stress reliever for me. It keeps my mind on a topic I enjoy, and everything else goes away for a few minutes throughout the day.

posted 11-22-2006 07:51 PM            

some of you need to get a hobby. I ain't stupid but all the discussion here left me feeling a little "less than..."

If I have this correct, you are basically arguing the difference between a Jaguar X12 and a Hyundai Accent. They both get you there, not not quite with all the smells and bells. I remember when Limestone came out with, oh, I dunno, 57 auxillary channels (slightly exagerated) or something, I thought to myself, WHY!?

Polygraph is Polygraph. Analog, v. computerized, or arguing manufacturers against each, it doesn't matter. Three channels recording simutaneously (with movement sensors), and as long as there are no changes or modifications between questions on the same chart; anything else is extra "stuff" we can't even use.

Thanks for letting me post here. I'll go back to my beer and Baywatch...

LOL

posted 11-22-2006 08:53 PM            

That's my question. Why does it matter? What am I missing? Why does one study say one manufacturer is using 20 or 40 year old technology? What's the difference? Why does ASTM require that EDA record either SC or SR, yet one of the manufacturers doesn't? Does all this stuff matter. Does it all get us to the same place? That can be answered with scientific or anecdotal evidence, which means there's really only one way to know....

Don't forget the 4th channel: the PLE.

Beer and Baywatch? Are you bedridden? Why would anybody do that to himself?

posted 11-22-2006 10:13 PM            

Polygraph depends on both art and science, but we can't afford to emphasize the art over the science. Imagine an artist trying to provide information (artwork?) to a court... or a police investigator - “would you be willing to discuss this with our polygraph artist...” or Paula Woodward (local investigative news-person) “the opinion of the polygraph artist was...”

Sorry guys, but polygraph, like psychology, is also half science. In this country we've emphasized a scientific form of psychology, just as in polygraph. Science means measurement.
Shawn,

Thank you. That was much more informative, but you've illustrated my point about the inaccuracy of your earlier statement...

quote:

---

You need a minimum of 22 Bit A/D to capture the entire range of EDA.

---

So its really not the “range” but the granularity or precision – just as I illustrated. 22 or 24 bit converters simply chop the “range” into smaller pieces – making them potentially more sensitive to very subtle changes. But are those subtle changes really useful or informative.

quote:

---

The standard in the field of psychophysiology is to have sufficient precision to reliably measure changes as small as 0.02 microSiemens.

---

Who sets these standards?

And for the Math check...

quote:

---

A 16-bit A/D will vary less than 1 unit (.065) per .001 microSiemens of conductance. Given the inherent noise in A/Ds, a change of perhaps 6.5 units from an A/D may be reliable. Therefore, a 16-bit system is capable of detecting a change of about 0.1 microSiemens.

---

So if we want range of 100 uS to a precision of 3 decimals, then we have 100,000 data points in our range. Now divide 65536 (which is the number of different values available with a 16 bit converter) by 100,000. I get .65536, not .065

Also, if we assume a noise level of 6.5 units – meaning that a 6.5 unit change is reliable - and our response level is .65 for every three-decimal division of our 100 uS range, that means 6.5 divided by ~.65 = 10

Now 10 x .001 = .01 not .1 as you stated.

Meaning that the 16 bit ATD converter should be capable of detecting changes at the standard you referenced.

So correct me if I am misunderstanding.

I don't dispute that a 22 or 24 bit converter would offer more precision.

The 24 bit example of the above would have us divide 16,777,216 by 100,000 = 167.77216

Meaning a ~167.8 unit change for every .001 uS change in response level.

Now 6.5 divided by 167.8 = 0.0387......

and .0387 X .001 = 0.0000387

meaning the 24 bit converter is really measuring a lot of NOISE...

its wasted precision.

Wasted because now you have to store all that noise data, and now you have to be sure you ignore it. Which is certainly possible to do, with excellent reliability, because we have all these underutilized computing cycles.

But, it might make more sense to use a converter that is less sensitive to noise.

All this precision is like my 7 megapixel camera – which produces images that are that are about 3000 pixels wide. And all I do is look at them on my computer screen that is 1280 pixels wide – and send them to my family who also look at them on a computer screen. All the extra precision is wasted. Excess data/precision slows down the system and clogs up space. If you want to “appreciate” the extra precision, then you can scroll around the computer screen as if were a peep-hole or periscope – but that's not fun.

And besides, with my strabismus I can't see worth beans anyway.

To illustrate,
http://www.raymondnelson.us/rn/humbolt.html

note the un-ironed socks.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 11-22-2006 10:15 PM         

>>>Beer and Baywatch? Are you bedridden? Why would anybody do that to himself?

I wholeheartedly agree. Why would anyone prefer that over basking in the afterglow of pondering "microSiemens of conductance?"

posted 11-22-2006 10:18 PM         

note the un-ironed socks.

r
------------

posted 11-23-2006 02:07 PM            

Happy Turkey Day to all!

Jim

P.S. I do not decry anyone's scientific and intellectual pursuit of knowledge pertaining to the "micro-details" of how the "box" works.