Comments on: Mistakes

By: Luigi DiLella

Luigi DiLella — Fri, 09 Nov 2007 14:11:56 +0000

Ellipsis: Sorry, but your story about me discovering a huge amount of background in my experiment, and modifying the electronics to remove it, thus missing a discovery, is totally wrong. The experiment was designed to search for new particles, but none was discovered because of a huge background from an unexpected phenomenon, namely the collision between two proton constituents (quarks and gluons), which occurred at such a high rate that we had to crank up an electronic threshold to reduce the data acquisition rate. This unexpected background was itself a discovery (we published it and reported it to various international conferences), but it prevented us from discovering the 4th quark. So, it was a new, unexpected discovery preventing us from making a more important discovery. A little like Christopher Columbus, who planned to reach India “from the other side”, but found something unexpected on his way (the Americas). In his case, of course, the unexpected discovery was more important than the planned one. In the case of my experiment in 1972 (and not 1974), it was the contrary.

By: Doug van Orsow

Doug van Orsow — Mon, 29 Oct 2007 06:09:39 +0000

Fellow Teaching Company addicts can now view Yahoo groups and phpbb forums:

A forum on each individual lecture in all recent courses:
http://teachingcompany.12.forumer.com/index.php

My posts in Robert Hazen’s “Origins of Life” forum:
http://teachingcompany.12.forumer.com/viewforum.php?f=17

Some of my new Yahoo groups:
http://groups.yahoo.com/group/Teaching_Company_Users/?yguid=317656331
http://groups.yahoo.com/group/Teaching_Company_Users_Professor/?yguid=317656331
http://groups.yahoo.com/group/Teaching_Company_Users_Subject/?yguid=317656331

Doug van Orsow
moderator

By: Belizean

Belizean — Wed, 12 Sep 2007 18:42:00 +0000

I like what John Wheeler would say about mistakes, “Our task in physics is to make our mistakes at the fastest possible rate.”

By: Sean

Sean — Tue, 11 Sep 2007 19:03:11 +0000

Yes, it was Jack Steinberger. He's pretty accomplished.

By: Mike

Mike — Tue, 11 Sep 2007 18:30:17 +0000

Sean,

Talk to David Politzer about comment #15. I think there’s a funny story behind this (if I remember correctly, the above mentioned anomaly was discovered by some a graduate student who figured he had made a mistake… in the aftermath of events frustration over this caused this particular physicist to switch from theory to experiment!… I think the experimentalist is quite accomplished, but I’ll withhold the name because my memory is probably wrong).

By: citrine

citrine — Mon, 10 Sep 2007 21:22:13 +0000

What about Messier’s cataloging work done to *prevent* mistaking fuzzy Astronomical objects (star clusters and galaxies) for comets?

By: Brooklyn Is Not Expanding | Cosmic Variance

Brooklyn Is Not Expanding | Cosmic Variance — Mon, 10 Sep 2007 16:56:46 +0000

[...] referred to this scene from Annie Hall in my talk yesterday. A [...]

By: Haelfix

Haelfix — Mon, 10 Sep 2007 09:19:05 +0000

I think theres a statistic somewhere that claims that over 50% of all published research is wrong in some way. Which sounds right, theres nearly always a sign error or some trivial constant ommission.

There are some very famous results too where factors of ‘i’ were misplaced and which screwed up a calculation completely leading the authors to make a wrong claim.

Of course, thats just the easy mistakes, there are many examples of much more complicated and serious mistakes. Einstein himself was guilty of one or two such papers -in fact the ‘greatest blunder’ quote might actually not have to do with the CC addition, but rather a divide by zero mistake he made in the paper.

By: amused

amused — Sun, 09 Sep 2007 06:42:03 +0000

One famous QFT mistake was the Sutherland-Veltman “no-go theorem” which claimed that neutral pions can’t decay into photons via the EM interactions. The proof assumed that the axial vector current is conserved…which is wrong! It was later shown by Adler, Bell and Jackiw that this current isn’t conserved – there’s an “axial anomaly”. The important lesson from this was that symmetries of the classical field theory don’t always carry over to symmetries of the quantum theory – they can get broken by quantum effects (“anomalies”).
Well, it’s good to know that even Nobel Prize winners stuff up sometimes…

By: Kaleberg

Kaleberg — Sun, 09 Sep 2007 01:07:55 +0000

For an example of how science progresses by changing our point of view, consider the humble History of the Dumplings.

By: Schwa

Schwa — Sat, 08 Sep 2007 20:48:48 +0000

I think my favorite ‘mistake’ is the Michelson-Morley experiment to prove the existence of the luminiferous ether.

By: MJ

MJ — Sat, 08 Sep 2007 16:35:18 +0000

Michael Gelb proposes 7 attributes of genius/creativity in his simple yet revelatory How To Think Like Leonardo da Vinci. Check out the chapter on Dimostrazione -- the openness to experiential learning, including mistake-making.

By: Ian B Gibson

Ian B Gibson — Sat, 08 Sep 2007 14:00:20 +0000

The topic is Mistakes! I think we’re all familiar with them.

How dare you! I’ve neveer maid a mistaek in my life.

By: Tim

Tim — Sat, 08 Sep 2007 03:28:06 +0000

One I like is how Hubble screwed up his calibration of the Cepheids, and had the temerity to draw a straight line through that mess he got (getting a slope an order of magnitude too high in the process) and declare the universe to be expanding.

I remember reading something from Allan Sandage talking about how one of Hubble’s greatest strengths was his ability to see through the data (in general, not just this case) to what was really physcially important.

By: Elliot

Elliot — Sat, 08 Sep 2007 02:23:13 +0000

Jason,

Yes it only works in SI units therefore is an insignificant result.

Elliot

By: Jason Dick

Jason Dick — Sat, 08 Sep 2007 01:59:24 +0000

Elliot,

I’m pretty sure the answer is it’s a fluke of the units you are using: the units in that formula don’t add up to mass. So you haven’t calculated the mass of a neutron at all: you’ve simply calculated a number that has some units other than mass, and happens to match up to the mass of the neutron in one choice of units. Given the large number of possible combinations of constants, of possible parameters for which you are attempting to “calculate” values, and the large number of possible choices of units, it is unsurprising that this sort of formula appears from time to time.

To see why the units don’t match up, notice that Boltzmann’s constant includes as part of its units temperature in Kelvin. It is the only constant in that expression to include temperature, and thus the units cannot come out to mass.

By: klien4g

klien4g — Sat, 08 Sep 2007 01:36:11 +0000

Somewhat related:
Here’s something that cheers me up on bad days/weeks (as i work toward my thesis)
” An expert is a man who has made all the mistakes, which can be made, in a very narrow field.” — Niels Bohr

By: Jim Kakalios

Jim Kakalios — Sat, 08 Sep 2007 01:12:21 +0000

As Franck Wilczek has said: If you’re not making mistakes, you’re not working on hard enough problems!

By: Ellipsis

Ellipsis — Sat, 08 Sep 2007 00:54:06 +0000

Anne: In 1974 Luigi di Lella of CERN discovered a huge amount of background in his experiment and modified the detector electronics to remove it. It was — as later realized — from the undiscovered 4th quark, the charm quark, 1976 Nobel prize for Sam Ting and Burt Richter, who simultaneously found it 2 months later.
Yes, there are a bunch of these cases!

By: Anne

Anne — Fri, 07 Sep 2007 23:58:43 +0000

There’s a story, unfortunately with no names, about the discovery of pulsars (which helped win Anthony Hewish the Nobel Prize). His student, Jocelyn Bell, was looking at chart recorder paper, and noticed some odd-looking noise. She noted it down, saw it again at the same spot in the sky a few times, and persuaded Hewish to let her follow it up. The story is that around that time another researcher saw the same funny noise, gave his instrument a thwack, and it went away. No discovery.

This is the mistake of looking only for the expected, which is very easy to make…