Hacker News Reader

Comment by lisper | Hacker News Reader

This is pure gold. I wonder why there are so few modern equivalents to this sort of thing. All of the modern physics pedagogy that I've seen describes this experiment in idealized terms, and it turns out that this idealization is really a lie: the split beam is not two cleanly separated beams, it's smeared out because of the range of speeds of the thermalized atoms. I think we do students a serious disservice by sweeping details like this under the rug. Worse, the trustworthiness of the whole scientific enterprise is undermined when we simply describe idealized results and ask people to take it on faith that we're telling the truth, especially when it turns out that we're not.

View on HN

I totally agree! I also very much appreciate the amount of time they dedicate to explaining the details of the apparatus. I suspect that very few physics students get any reasonable amount of instruction on experimental techniques these days. I barely did fifteen years ago, and I think things have only gotten worse since then. Without that, the learning curve of working in a research lab can be pretty brutal.

A second benefit I get from spending all that time on the apparatus is kind of paradoxical: those extra details seem like they would distract you from learning and understanding the core concept, but somehow they make things much more approachable. The idealized experiment is slightly too abstract to easily digest.

I have never seen this kind of physics taught as THE TRUTH. Almost everything is an approximation, and there are almost always higher order terms smearing things around in a lab setting. This is part of why physics is taught with a lab component, to help build intuition for handling uncertainty and its sources.

The SG experiment is a great intro topic for an undergrad quantum class because the essence of it, that you measure two clusters, cannot be explained with classical mechanics. Complicating that headline further is pedagogicaly counterproductive. There is no grand conspiracy to deceive students.

Well, it depends on your goals. All the physical exigencies of the SG experiment require a great deal more theoretical machinery to describe and I think all the mix-states/pure-states stuff can prolong the period during which the student clings to the fantasy that some simple classical physical system is simply obscured by its size and the physical limitations of classical measurement. I do think a full accounting of the SG experiment is useful for the student, but if the intent is to describe the machinery of quantum mechanics per se, then details obscure, rather than clarify.

You would run into the same problem with teaching CS and the underlying mechanics of the abstractions that underlie libraries/frameworks to apathetic programmers: "When am I gonna use this in real life? I just wanna make money so all I need to know are the abstraction."

Reply on news.ycombinator.com