[alancalverd]

I would class a Feynman diagram as a mathematical model, as, I think, would Feynman.

Experimental evidence is that some electromagnetic radiation is generated or absorbed in discrete amounts, as evidenced by line spectra and photon-counting receptors, but there are no inherent discontinuities in the radiofrequency spectrum, and the photon model doesn't predict diffraction.

Some folk still talk about "duality" but that's clearly nonsense. Only a deity can change its form ad lib (in Greek mythology) or exist in three distinct physical forms at once (Christian mythology). Real phenomena sometimes need more than one descriptive and predictive model.

[Origin]

The veracity of information doesn't depend on the form of its source. It can be statement of a teacher, or colleague, textbooks, online encyclopedia, news, TV, website, blogs, social media, or AI models.

Perhaps that's the issue.  Without critical thinking skills, all sources no matter how unreliable seem to be on equal footing?

[varsigma]

I would class a Feynman diagram as a mathematical model, as, I think, would Feynman.

I would say, if I was asked, that a Feynman diagram is a representation of a(n element of a) symmetry group. Namely U(1) symmetry. The fermions are there for closure (that's over the algebra).

With the caveat that the symmetries are not exact according to modern theories.

[hamdani yusuf (OP)]

Huygen's construction doesn't purport to explain anything.

It would make it useless.

[hamdani yusuf (OP)]

The veracity of information doesn't depend on the form of its source. It can be statement of a teacher, or colleague, textbooks, online encyclopedia, news, TV, website, blogs, social media, or AI models.

But it has to be correct, and if it is in response to a question, it has to be relevant and informative. Better still if it is actually helpful.

One line of code:

IF CHAR=<?> THEN PRINT "the answer may be found in textbooks". 

How's that for efficient programming, in genuine 1960s Algol!

Gemini's answer is better than most sources found online which I posted on the first page. Some of them are physics courses for students. It's even better than Feynman's answer, which was a leading scientist.

[hamdani yusuf (OP)]

I would class a Feynman diagram as a mathematical model, as, I think, would Feynman.

I would say, if I was asked, that a Feynman diagram is a representation of a(n element of a) symmetry group. Namely U(1) symmetry. The fermions are there for closure (that's over the algebra).

With the caveat that the symmetries are not exact according to modern theories.

How can Feynman's diagram be used to explain diffraction and interference?

[hamdani yusuf (OP)]

The veracity of information doesn't depend on the form of its source. It can be statement of a teacher, or colleague, textbooks, online encyclopedia, news, TV, website, blogs, social media, or AI models.

Perhaps that's the issue.  Without critical thinking skills, all sources no matter how unreliable seem to be on equal footing?

That's why Jordan Peterson said that people who are not that smart are better to be conservative, and follow the mainstream.

[alancalverd]

Gemini's answer is better than most sources found online which I posted on the first page.

Which is a great shame, because it is complete garbage.

[alancalverd]

I would say, if I was asked, that a Feynman diagram is a representation of a(n element of a) symmetry group. Namely U(1) symmetry.

Or to quote Wikipedia

In theoretical physics, a Feynman diagram is a pictorial representation of the mathematical expressions describing the behavior and interaction of subatomic particles.

which is pretty much what I said.

How can Feynman's diagram be used to explain diffraction and interference?

It can't, because it is a representation of particle interactions, not waves.

[alancalverd]

Huygen's construction doesn't purport to explain anything.

It would make it useless.

We have very accurate models for predicting sunrise and high tides, but don't explain either. Try telling a mariner that they are useless.

[varsigma]

which is pretty much what I said.

There it is. We are saying basically the same thing. At least I think we are.

Feynman diagrams are composable, There are sums over all the ways a photon can be emitted and absorbed. How many do you need to glue together to represent this? That was I think, what Feynman was using his "new" notation for.

[hamdani yusuf (OP)]

Huygen's construction doesn't purport to explain anything.

It would make it useless.

We have very accurate models for predicting sunrise and high tides, but don't explain either. Try telling a mariner that they are useless.

What is predicted by Huygens construction, when a laser beam hit the edge of a totally reflective surface?

[alancalverd]

Edge (dimension → 0) or surface (dimension → ∞)? One produces diffraction, the other produces reflection. Real objects fit somewhere between.

[hamdani yusuf (OP)]

Edge (dimension → 0) or surface (dimension → ∞)? One produces diffraction, the other produces reflection. Real objects fit somewhere between.

Point, 0 dimension.
Edge, 1 dimension.
Surface, 2 dimensions.
Volume, 3 dimensions.

[alancalverd]

OK,I wasn't being pedantic enough!

The length of an edge is immaterial to the Huygens construction of diffraction. All that matters is that, perpendicular to the wavefront, it can be mapped to a point without losing important information.

You can't usefully map a surface to a point.

[hamdani yusuf (OP)]

I searched for an image of knife edge diffraction.
Here's what I found.

http://www.mike-willis.com/Tutorial/PF7.htm

The question is, how much of the wave is diffracted? How to determine the angle?

[alancalverd]

A rather blunt knife, but at least commensurate with the wavelength, so Huygens is a good approximation.

[hamdani yusuf (OP)]

A rather blunt knife, but at least commensurate with the wavelength, so Huygens is a good approximation.

Do the characteristics of the barrier have any effect?
What does Huygens' principle say about this? Is there any quantitative statement?
What's your threshold for being good?

[alancalverd]

Huygens is a mathematical construction involving points, lines, and absolute absorption and reflection. Anything that isn't a mathematical point (zero dimension) or line (1 dimension), or is a real material, is an approximation.

If the "edge" is actually a surface, and/or the source has finite width, you need to consider the contribution of all the points on the surface and the source. Intuitively (though I guess ES probably has an appropriate integral) this means that the diffraction pattern becomes more diffuse if either is wider than λ.   

[hamdani yusuf (OP)]

A rather blunt knife, but at least commensurate with the wavelength, so Huygens is a good approximation.

Do the characteristics of the barrier have any effect?
What does Huygens' principle say about this? Is there any quantitative statement?
What's your threshold for being good?

Don't forget that I also have made the same experiment in microwave frequency.

I have uploaded new video showing diffraction in microwave frequency.

Basically, the experiment result leads us to conclude that diffraction comes from the material blocking the microwave path. When the obstruction is opaque enough, we find no diffraction. It's similar to my experiment using laser showing non-diffractive obstruction.

This result is not widely known yet.