How A Vermont Farmer Proved No Snowflakes Are Alike

cnn.com

11 points·andrewl··5 comments

Open articleView on HN

In lab situations, it is possible to grow "identical twins".

Some various coverage on Professor Libbrecht and his snow flakes...

KEQD: https://www.kqed.org/science/1536955/identical-snowflakes-sc...

Quartz: https://youtu.be/N2oJaW7Xiek

Veritasium: https://youtu.be/ao2Jfm35XeE

And if you want to dig deep: http://snowcrystals.com

The article's title would be better if it was something like "the Vermont farmer who was the first to photograph snowflakes". The article isn't about the chemistry / physics, it's about how he collected lots of pictures. It's a cool achievement, just not reflected in the title

What bothers me is this: Popper once argued that to prove something, you need to falsify the claim. I am reminded of the story about "no such thing as a black swan", which held right up until someone brought a black swan into town from afar.

When this came up yesterday, at https://news.ycombinator.com/item?id=34130864 , I pointed to an article by Nancy C. Knight, "No Two Alike?" in Bulletin of the American Meteorological Society, 1988. https://journals.ametsoc.org/downloadpdf/journals/bams/69/5/... .

> Figure 1, a photograph of actual snow crystals, not replicas, illustrates a striking example of two snow crystals which, if not identical, are certainly very much alike. ...

The two very similar snowflakes are in the bottom center of figure 1, side-by-side.

Instead of extrapolating "there are no black swans in town" to "the world", it would be better to calculate the probability of no black swans being in town (given that there are N swans in town), based on hypothetical populations of swans and varying uniformity of distribution.

A zero-knowledge proof[0] uses similar reasoning.

[0] https://en.wikipedia.org/wiki/Zero-knowledge_proof

Reply on news.ycombinator.com