Tobler's First Law of Geography | Hacker News Reader

Tobler's First Law of Geography

en.wikipedia.org

63 pointssebwi17 comments

His name is Tobler? His first law, you say? Would it be fair, then, to call it "Tobler one"?

You won't do any better than that. I think you've peaked.

Yeah, I never thought I would see a pun thread at the top of a HN comment section. This deserves some sort of special award. :P

Agreed. In fact I haven't checked HN since I posted that and fully expected to be downvoted to all hell for that comment when I logged back in just now. I'm really impressed with myself. Made my week.

Yeah, and the best part is that nobody can downvote your toplevel comment now. :-)

Something i find really fun is that every specific field of study has developed its own completely general statistical tools. There's no reason they couldn't be used in other fields. They just aren't.

Geography apparently has Kriging:

https://en.wikipedia.org/wiki/Kriging

Economists have LOESS (okay, used beyond economics, i admit it):

https://en.wikipedia.org/wiki/Local_regression

Maybe geographers refuse to use LOESS because to them, that's a boring rock?

Astronomers have sophisticated deconvolution algorithms that are completely unrelated to the ones microscopists use, etc.

Kringing is the same thing as gaussian process regression, and astronomers use GPs a fair bit. I'm not sure whether they are used more widely.

My favourite forgotten/isolated statistical method is MCMC. These were first used by nuclear physicists at Los Alamos in the 40s/50s, but weren't really recognized more widely until the 80s. This is probably partly because only people working on bombs had access to the computing power before then, but still.

Astronomy is also an interesting case because this law doesn't necessarily hold true. The power spectrum, or Fourier transform of the autocorrelation function, isn't monotonic for matter on large scales. For example, baryonic acoustic oscillations at early times in the Universe get "imprinted" into the cosmic structure, such that galaxies have a preferential separation between each other.

MCMC is definitely not forgotten, it's still taught and used extensively by a bunch of folks, including those writing graphics engines!

I'm not suggesting that at all! I'm just saying it took 30 odd years to take off/break out of the nuclear physics world.

Though rereading my first comment that wasn't entirely clear...

How is this not just common sense? Not trying to sound cynical... I don't have background in this space. Honestly trying to understand why "things that are close together and more related than things farther apart" is considered so profound.

Try to read Newton's Laws of Motion and ask the question.

You would be surprised how much of today's "common sense" was cutting edge scientific thinking at some point in time in the past.

Why write a law explains that apples fall down from trees?

My favourite "application" of this law is Kriging:

https://en.wikipedia.org/wiki/Kriging

Has allowed me to make some pretty awesome spatial interpolations of water surveying and wifi quality data using unmanned robots, especially given the massive isotropic bias of the collection method that makes many other methods nonviable.

Also known in a number of other fields like psychology or sociology, under some catchier rubrics like "everything is correlated": https://www.gwern.net/Correlation

I wish inverse distance scaling worked better in other domains - e.g. weighted nearest neighbor estimators seem to not generally perform better than unweighted ones.

Computer types would say: "the law of spatial locality"

Reply on news.ycombinator.com