Michigan State University students had an unconventional math class this week after professor John McCarthy, 57, reportedly stripped naked in Calculus 1 class and began screaming that “there is no f—ing God” and “It’s all an act and none of it’s real.” In the age of laptop computers, the most surprising thing is that the students noticed and looked up from their computers. Presumably one student asked the standard question of “will this be on the test?”
Witnesses said that McCarthy was going through a routine derivative equation when he started to talk about his colleagues and how “they’re all actors.” He added that “It’s all an act and none of it’s real.”
McCarthy was arrested but not actually charged with a crime. He can certainly be charged with indecent exposure but I hope that he is not and that the police show a modicum of discretion. He obviously had a psychological meltdown. Brilliance sometimes comes with such mental issues as vividly shown in movies like “A Beautiful Mind.”
I would also hope that the faculty treats this matter as a mental illness and allows McCarthy to seek treatment rather than simply fire him. I realize it will be difficult for him to return to the classroom, but we need to treat mental illness like other forms of illness. Thus far the school has merely reassigned his classes. I expect that he is unlikely to return to teaching but in my view the school should treat this matter as an illness rather than misconduct. What do you think?
Source: NY Daily News
This is a topic that’s close to my heart… Many thanks! Exactly where are your contact details though?
Tony,
First off, you are conflating the discussion about two separate things, neither of which you have any particular expertise in, and then claiming your expertise in a third area makes your comments relevant. The first is business calculus—which is designed to teach the basics of calculus using examples drawn from business rather than physics (another example of this is drawing examples from biology—I have taught all three types of calc I, by the way). The purpose of these classes are not to teach people to solve applied math problems in the various fields (although they mesh well with such classes), but to use familiar subject matter to aid in the teaching of calculus and critical thinking. The other is the application of math to economics—specifically my belief about what will happen in the future. I tried to make this clear when I said:
“…in my opinion business/economics is going to be the next biology in terms of math—which is to say that it will become the new frontier of applied math, but that’s not what business calc is about…” [I went on to describe sample types of problems drawn from business to illustrate several standard calc I concepts]
But I guess you didn’t get it. While there is nothing wrong with robust statistics and it is certainly applied to the type of modeling used in business for the last 25 years—it’s just not at all what I was talking about. In general, there are two kinds of modeling: top-down modeling and bottom-up modeling.
The statistical models you are familiar with are top-down models where historical data is used in statistical methods (like linear regression) to build a model which makes predictions by determining (in some fashion) how the current data correlates with the historical data. This is fine as far as it goes, but all you can get out of this kind of model is what you put into it—i.e. the only relationships between the inputs and the outputs are the ones you have explicitly included. Also, these tend to use linear techniques—as one of my professors once said, “separating functions into linear and non-linear is like separating food into bananas and non-bananas”. Now, the reason people use linear methods is because they are easier—both to implement and to understand—, but they are also, well, linear. Which severely limits the sort of dynamics they can model.
In bottom-up modeling, on the other hand, the components of a system and their basic interactions are modeled—generally either by differential equations (deterministic models) or by probabilities (stochastic models). The individual interactions are all very simple to model—the sort of things that you would be able to understand after a basic calculus or probability course, but there are generally a lot of them. The advantage (or one of them) of this type of model is that they are capable of emergent behavior—they end up being able to do things that they weren’t specifically designed to do. In other words, if you make a model of, say, how you think that a bunch of proteins interact, you can then do simulated experiments and see if the behavior that results from the model matches experimental results. If they don’t match, then either you are wrong about some of the interactions or there are interactions that you don’t know about—and the model can be used to test new hypotheses to see if they result in a better match with the data (without doing any additional bench testing). Once the model accurately exhibits the behavior seen experimentally (say that you’ve modeled the response to a dozen different drugs), then you can easily use simulations to determine the effect of combinations of all of those drugs at varying dosages. In other words, if you can model a checkpoint in the cell cycle and it’s response to various chemotheraputic agents, then, by taking protein expression data from a patient’s tumor as well as normal cells, the cocktail of drugs and radiation which is most toxic to the tumor and least toxic to the normal cells could be found. Now, you might think this is just speculation on my part, but I’ve described the outline of a grant proposal that the head of the lab I was part of for my post-doc suggested we submit (which I hope to do in the coming months).
The problem with these sorts of models is that they become very complex and difficult to manage. In my post-doc, it took me about 3 years to build a working model of the DNA damage G2 checkpoint and another 9 months to design the software to create the model files, run them, and then collect, process, and collate the data for a perturbation study to show that the results of the model were robust with respect to the choices of the model parameters. Since this process is far too long and requires too much expertise for most bench labs to be able to afford it, I’ve been designing a tool to speed up this process. When I get done with the stable alpha version of this software (my goal is Halloween), I estimate that I will be able to construct a model (and experiment) of the same complexity roughly two orders of magnitude more efficiently—that’s right, about 100 times faster. Eventually, my software will allow someone with an understanding of the biology and minimal training on the software to build, maintain, extend, and run these types of models in a fraction of the time it currently takes someone with a very high level of expertise to do.
This is the type of modeling that I believe could revolutionize cellular biology and, in my opinion, the next big area that will be pioneered by this kind of math will be economics. As I think you can see, I have some relevant expertise in this sort of modeling and, in my professional opinion, applying it to economics isn’t fundamentally more difficult that applying it to biology. My guess is that you don’t have any experience with this type of methodology nor the ability to implement it—even though there isn’t any math here that would be beyond your understanding (in the biological models I described—different problems require different mathematical techniques… for instance, anything involving human decision making, I would probably use a healthy amount of game theory to model).
So next time you read something I wrote and you think I’m saying something stupid and obviously wrong, I suggest you take a minute to consider that you might not be understanding what I’m talking about and that I may not be the idiot you so blithely assume.
Note to Malisha: I avoided using the “P”-word just for you (a friend told me it just makes people’s eyes glaze over, but sometimes there isn’t another word that will work… 😛 ).
Tony C. 1, October 6, 2012 at 7:03 pm
Back at you.
And it is my opinion that you do not know what I’m talking about. Maybe you should have paid more attention to the old saw about it better to not say anything and be thought a fool…
It’s just my opinion, but I think you failed—badly.
Is there a reason that your sensitivity analysis links to the top of this thread? I cut this out as I didn’t really disagree with much of anything you said… and where’s the fun in that?
No, that’s fine. Much of the time I have trouble putting my ideas in layman’s terms—I’m not the teacher Feynman was.
Semantic difference—I’ll use the term “economics” from here on out to mean a much wider view of the subject as anything having to do with money (or commerce of any sort) and use the term “business” how you’re defining it. Okay? I still believe that this sort of modeling can be useful in business—because I have actually done it professionally.
To understand robust statistics? Maybe. To use tools to build their own statistical models—I think not.
For this sort of modeling to be successful, the clients are not the consumers of the methods or the models, but of a service which doesn’t require them to understand anything outside of the paradigm (sorry Malisha!) with which they are comfortable. In other words, it’s about people like me doing the math so business-type people don’t have to.
Whatever—I can model that, too (and I would bet that I could come up with a profitable trading algorithm, given the chance).
I’m currently in the process of constructing a model of my company which will be in the business of creating modeling methodologies, engineering modeling tools, and producing custom modeling services for clients. That good enough? (If not, I’ve also created a prototype model of the logistics and finances of a manufacturing company…)
While the biological model I spoke of required the management of over 60,000 simulations of a model consisting of over 700 differential equations representing almost 2,000 one-way reactions, the most complex of those reactions was of the form:
[raff—remember your happy place!]
da/dt = k*a*b
where k is a constant and a and b are variables. I would think nothing of making my students solve an equation of this sort by hand with Euler’s method on a calc I test (actually solving large systems of this sort requires more sophisticated methods, but they are the same as Euler’s method conceptually). Even simple math can sometimes yield powerful results.
I would guess that most models don’t require very sophisticated consumer behavior, but I think that, as I said above, with game theory (and some other tricks) fairly complex behavior could be modeled if necessary. Also, you can’t predict news—by definition. The value add of modeling is to determine how news affects the system (in order to, say, identify and take advantage of arbitrage opportunities). Nate Silver’s model at 538 predicted a 25% decline in John McCain’s chances less than 48 hours after the collapse of Lehman Brothers, as an example.
That’s for others to judge, but this comment was premature at best…
I’m sure that all of that has given you a deep, nuanced understanding of bottom-up modeling, the management of complexity, and non-linear dynamics (i.e. chaos math) that puts what I’ve learned over the course of the last two decades of studying and using mathematics to shame.
Managers need to understand enough to know what their engineers are saying—Feynman discovered a disconnect between the managers (100,000:1), and the engineers (300:1), and an independent analysis (as high as 50:1) regarding the odds of a mission failure on the space shuttle due to the engines. I have reason to believe this sort of disconnect is endemic in business and has at least contributed to things like the mortgage-backed security debacle. An understanding of critical thinking which can come from calculus is useful in combating this serious problem.
Which doesn’t mean that predictive models can’t be constructed using the mathematics learned in calc classes that provide significant insight to these people—without requiring them to understand the math.
Quite often business problems include physics or engineering problems, but, again, that’s not the point of “business calculus”—which just means using problems from business to illustrate and drill basic calc concepts. This allows the student to use the intuition which they have (hopefully) developed or are developing regarding business to inform their understanding of mathematics—and vice versa.
Gene et al.,
You may now consider what part of who’s ass I did or did not kick. 😉
OS,
Sorry about your friend.
Matt, that is comment enough. Thanks.
Otteray Scribe 1, October 7, 2012 at 6:33 pm
Matt. Yes I know what courage is. My friend CJ Campbell was the very face of courage. He died last Sunday. Our mutual friend Kelley just published a tribute to him.
Courage? Its name is CJ Campbell:
============================
I’m truly sorry for your loss. I looked at the tribute. Looks fantastic, but I can’t comment.
OS,
Sorry to hear about the passing of your friend.
Today is Niels Henrik David Bohr’s birth day…
Re: Dr. Mallett’s time machine: This article helped me just now http://phys.org/news63371210.html I just couldn’t initially understand where HIS brain was around this. He believes in parallel universes it turns out. Also I was wrong it wasn’t a HIGGS boson it was a gauge Boson. (Gauge? Another awful semantic-word).
@Otteray Scribe – I kick myself for passing up a wi-fi controlled lithium-cell helo for $25 bucks the other day. You’re not allowed “second thoughts” in this world of instant opportunities. Gives meaning to “strike while the iron is hot” doesn’t it?
Yeah R/C’s are much safer at least for the pilot (LOL) – not the bystanders and home owners… 🙂
SoTB,
She has not gotten a pilot’s license. Not with insurance rates through the roof and 100 octane low-lead avgas at $6.25/gallon. She has taken up flying radio control airplanes.
@Otteray Scribe – Sorry to hear your personal pain. We all gain strength from the sharing. He sounded like a interesting man.
That’s right! I forgot you are a puddle-jumper pilot too. I guess your daughter has her certs now?
@All – In the venue of “cool gadgets”: Can someone please explain to me “twisted lasers”? Evidently there is a new technology that allows massive amounts (i.e. several Terrabytes) of data to be transmitted via “twisted lasers” for only a few feet (through open space). OK FYI… I get the basic concept but trying to find ANY detailed data (like a white paper) on the gadget is like pulling hen’s teeth.
The same goes for Dr. Ronald Mallet’s (Univ of CT) frame dragging ring laser system that purportedly can send a Higgs Boson BACK one second in time. Again hen’s teeth! All you get is laymen’s terms on coffee stirring and his late father’s smoking habit. All I want to see is a photo of the actual time machine in his Physics Lab in Storrs CT and some rational explanation of “frame dragging” and the “slowing down time”. BTW this is NOT a hoax it is quite real and perfectly sane science by Dr. Mallett. But the US government refuses to subsidize it (in the open that is).
Please do not offer Wikipedia articles as they make no sense to me. (BTDT!).
I think the problem I having is scientist’s use of common words like “twisted” and “dragging” to explain their inventions but it doesn’t seem to fit the scenario being described. I think they need better semantic-wordsmiths.
SoTB:
Thank you. I had a really hard time reading the poems because he mentions me (not by name) in a couple of them. He was a true Renaissance man. I had to laugh when I first heard the story of him having to make a dead stick emergency landing in a lettuce field, while holding on to his cameraman by the belt. The door had been removed from the Cessna, and the cameraman was hanging outside.
He was the face of courage right up to the end. He continued to coach kid’s soccer up to a few days before he died, despite white-hot pain from the cancer.
@Otteray Scribe – Condolences for your loss my friend.
@Otteray Scribe – Regarding Albert Ellis’ adolescence and being “shy” around women and his later affiliation with Alfred Kinsey; do you feel that his adolescence shyness toward females could be attributed to… (fill in the blank)?
FYI – I’m not being serious here… I’m trying to inject cyber-levity to our off-topic discussions. 🙂
Matt. Yes I know what courage is. My friend CJ Campbell was the very face of courage. He died last Sunday. Our mutual friend Kelley just published a tribute to him.
Courage? Its name is CJ Campbell:
http://www.dailykos.com/story/2012/09/30/1138046/-Nurse-Kelley-Sez-ulookarmless-has-passed-away
@Tony C – Yes I agree with the Wikipedia article on AO (adaptive optics). However, I was referring to not deforming a GLASS lens. I don’t feel the Perkin-Elmer patent you, albeit inadvertently referred to, dealt with that. Actually I am in agreement with you that the CIA did use such a technique to “fix” spy satellite photos AFTER they returned to earth to enhance resolution and mitigated blur. That would mean that P.E. scanned the Carl Zeiss glass lens with the laser beam just as you suggested. Whether or not CIA/DS&T (our QBranch) used the technique on common photographs seems like a logical evolution of the technology. So I agree with you a very “cool spy gadget”.
Next: Am I highly influenced by “doctorates”? As my Turley-friends already know I am autodidactic. So the answer is an emphatic “YES”! However, I don’t always agree with everything the they say. So if you titans want to clash – so be it. Go for it (LOL). However, I’m delighted that your topics have morphed slightly. If I don’t have to continue to read this cat & mouse-esque* exercise about Business Calculus I will be a happy camper (sigh).
*Example: http://tinyurl.com/8eeup3n
At least we got Dr. Otteray Scribe to wake up! That’s a good thing. (LOL) Hi O.S.
@Matt Johnson – No I don’t think you are displaying ODD characteristics. Tony C, you, and me seem to share a healthy opposition to authority (or reasonable facsimiles of same). It’s the key reason why I could never be in the military or any other related community. I presently work for people in the private sector (ex-everything) who display characteristics of psychopathy (See Hare Psychopathy Checklist). Even though they display apparent improper emotions for the corporate workplace does not mean they in fact are clinical psychopaths. This probably why I am so sensitive to the cat & mouse maneuver mentioned above (I’ve uncomfortably have been the mouse all too often).
Mike Spindell 1, October 7, 2012 at 5:06 pm
“So you’re a psychotherapist, Mike? What’s the 90/10 rule?”
Matt,
What OS said and though I’m a fan of Ellis, I don’t think you can quantify the percentages.
================
I know what the percentages are, but I don’t know who Ellis is.
Otteray Scribe 1, October 7, 2012 at 4:42 pm
Sometimes stuff happens and there is not a damn thing you can do about it except perhaps die.
==================
My Dear Friend, OS. Everybody gets one day older until you don’t. It isn’t your choice. Do you know what courage is?
Matt, I know of a 90/10 rule, but it applies to business and economics. Never heard of it applied to psychotherapy.
Are you referring to Stephen Covey’s notion that 10% of life is made up of what happens to you and the other 90% is how you react? If so, that is the basis, more or less, of Rational Emotive Behavioral Therapy, which was developed by the late Dr. Albert Ellis. In the real world, I do not think it can be quantified down to a 9:1 ratio. Sometimes stuff happens and there is not a damn thing you can do about it except perhaps die.