Once again for those who may have forgotten, the Moore Method, as practiced by R.L. Moore and endorsed by Paul Halmos, et al, is far from the "reform math" as pushed by the education industry and Keith Devlin's idea of "the best math teacher ever" ignores the real one by an unbiased observer, Jay Mathews of the Washington Post:
Back to what Halmos actually saw, there was no "guide on the side" valid interpretation.  This "guide" dominated the room - scared the hell out of many with good reason - and precision mattered absolutely.  No fuzzy thinking was involved and the approach is only possible under the leadership of a very knowledgeable and talented practitioner, rare even at the university level let alone K-6 in US schools that I consider most important let alone US middle and high schools:
I am reminded of Paul Halmos's great "automathography", I Want to Be a Mathematician on the Moore Method, often equated with the math avoidance of so-called "reform math".  Nothing could be further from the truth.  Attached (sorry for the crude scan) are a couple of pages that summarize his approach to mathematics instruction clear down to first-year Calculus where he was, at least for some students, highly effective.  A personal friend now deceased, eventually became a well-known seminary professor (Berkeley Union and then a permanent Visiting Distinguished Professor here at Fuller after his retirement) who started his life with a very strong bachelor's degree in physics from UT Austin with calculus from R. L. Moore.  He was so impressed that he credited Moore with his start as a top-notch academic researcher and maintained a lifelong correspondence with Moore.
As you can see, he was nothing close to the "not a sage on the stage but a guide on the side" wishful thinking but, in Halmos's words, "his personality dominated the room".  Only a very special kind of person can carry it off and don't try it without the "security blanket" of tenure.  The one tenure-track assistant professor who tried it at our campus was very lucky (only on special appeal to the dean) to not have had the following year be deemed his terminal year rather than another year toward eventual tenure.  En masse complaints to the chair during the quarter followed by record-low student evaluations at the end of the course.  Note the familiar cliché, "I hear, I forget; I see, I remember; I do, I understand."  It's all in what you mean by "doing".  Moore meant the real thing; not some phony imitation.  I have heard that in advanced topology classes he went even further than Halmos describes; he gave the words different names so that students couldn't "cheat" even if they wanted to.  They would have to understand the subject well enough to know how to read conventional books for proofs of his list of statements to be proved by students independently.

I never had the privilege of watching R.L. Moore at work, as did Halmos, but I did see U Chicago's Paul Sally run a demo class of his exceptional precollegiate class for Chicago-area students.  A wonder to watch and no doubt who was fully in charge every minute.  Amazingly, Everyday Math evolved from that but only after the ed-types took over and chased Sally, the original NSF grant PI (UCSMP), out entirely.

At 02:38 PM 4/15/2013, Richard Hake wrote:
Some subscribers to Math-Teach might be interested in a recent post "R.L. Moore - Pioneer of Math Education Reform" [Hake (2013)].  The abstract reads:

ABSTRACT: Contrary to the misrepresentation of the "Moore Method" < http://bit.ly/LElQzB> by direct instructionist Wayne Bishop at < http://bit.ly/qvnOIa>, I excerpt ten commentaries demonstrating that the Moore Method is, in fact, (a) an example of "math education reform ," and (b) taught by a  "guide on the side." The commentators are:

1. Keith Devlin 1999) in "The Greatest Math Teacher Ever" part 1 at  < http://bit.ly/12GYCSR> and part 2 at < http://bit.ly/17pBWdu>.
2. Educational Achievement Foundation's (2006) "A Quick-Start Guide to the Moore Method" at < http://bit.ly/ZvQZly>
3. Paul Halmos in "The Problem of Learning to Teach" (Halmos, Moise, & Piranian, 1975) at < http://bit.ly/12BgyOP>.
4. F. Burton Jones (1977) in "The Moore method" at < http://bit.ly/17nyIaB>.
5. Albert C. Lewis (1999) in "Reform and Tradition in Mathematics Education: The Example of R.L. Moore" at < http://bit.ly/YbjUoy>.
6. G. Edgar Parker (1992) "Getting More from Moore" at < http://bit.ly/YbmaMI>.

7. The MAA review of "The Moore Method: A Pathway to Learner-Centered Instruction" [Coppin, Mahavier, May, & G.E. Parker (2009)] at < http://bit.ly/LEdug3>.
8. "Discovery Learning Project" at the University of Texas (2013) at < http://bit.ly/12FqZEW>.
9. Lucille S. Whyburn (1970) "Student oriented teaching-The Moore Method" at < http://bit.ly/YNS5X4>.
10. David Zitarelli (2004) in "The origin and early impact of the Moore Method" at < http://bit.ly/149JYIJ>.

To access the complete 54 kB post please click on < http://yhoo.it/132baYU>.

Richard Hake, Emeritus Professor of Physics, Indiana University
Links to Articles: < http://bit.ly/a6M5y0>
Links to Socratic Dialogue Inducing (SDI) Labs: < http://bit.ly/9nGd3M>
Academia: < http://bit.ly/a8ixxm>
Blog: < http://bit.ly/9yGsXh>
GooglePlus: < http://bit.ly/KwZ6mE>
Google Scholar  < http://bit.ly/Wz2FP3>
Twitter: < http://bit.ly/juvd52>
Facebook: < http://on.fb.me/XI7EKm>

REFERENCES [URL shortened by <http://bit.ly/> and accessed on 14 April 2013.]

Hake, R.R. 2013. "R.L. Moore - Pioneer of Math Education Reform," online on the OPEN Net-Gold archives at  < http://yhoo.it/132baYU>.. Post of 14 Apr 2013 15:57:26 -0700 to AERA-L and Net-Gold. The abstract and link to the complete post were transmitted to several discussion lists and are on my blog "Hake'sEdStuff" at < http://bit.ly/XLLYE2> with a provision for comments.