Wind Map | Hacker News Reader

Wind Map

hint.fm

141 pointssethbannon23 comments

Previous discussion: http://news.ycombinator.com/item?id=3767889

Fun fact: I actually use this website in technical interviews and ask the candidate, how they would go at implementing it. (I work at a meteorological company.)

I would've gone with MADIS data, although interpolating that data over the area without sensors could be tricky, or perhaps could be worked in. Additionally it would be better if it had a more traditional slippy maps interface, was less "artsy," and like another poster on here had some labels. It would be great to also explore if highlighting gusts or or other straight-line wind type events could be shown as propagating and dispersing in areas where higher resolution sensors are located. Anyway, now that I've typed this, I guess you didn't really ask :P

I'd also like to see an underlying topographic map. It'd be interesting to see how the surface features affect the wind.

Along these lines I think it was interesting to see pretty much why "Tornado Alley" works:

http://en.wikipedia.org/wiki/Tornado_alley

This would make a beautiful visualization of the hairy ball problem:

"The hairy ball theorem of algebraic topology states that there is no nonvanishing continuous tangent vector field on even dimensional n-spheres. For the ordinary sphere, or 2‑sphere, if f is a continuous function that assigns a vector in R3 to every point p on a sphere such that f(p) is always tangent to the sphere at p, then there is at least one p such that f(p) = 0. In other words, whenever one attempts to comb a hairy ball flat, there will always be at least one tuft of hair at one point on the ball. The theorem was first stated by Henri Poincaré in the late 19th century."

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

Proof be damned, I can't help but try and think of ways to comb it flat... :)

Hairy bagels, on the other hand, can be combed without cowlicks.

What about a bagel with two holes?

The Euler characteristic of a compact orientable genus g surface X is nonzero if g is not 1 (the general formula is 2-2g), so the Euler class e(TX) of the tangent bundle TX of X is nonzero. This implies that TX cannot have a continuous nowhere-vanishing section. If it did, e(TX) would be 0. The single-holed torus is the only orientable compact surface that possesses a continuous nonvanishing vector field.

Ah, the dilemma of Wikipedia math articles, which all kick off with mouthfuls like "nonvanishing continuous tangent vector field."

If you don't have a PhD in mathematics, they're unreadable. And if you do, they're unnecessary.

[Tangent about math articles on Wikipedia]

I'd like to respectfully disagree. 'Nonvanishing continuous tangent vector field' is a phrase that can be understood after freshman level multivariable calculus.

More generally, Wikipedia articles on math do tend to rely on the accepted terminology of the field for the opening sentence. It's generally better, however, to skim that bit, pick out the words you know to make sure you're in generally the right place, and then skip around until you find something you can latch onto. If you're really lost, you might need to click around a bit.

It's actually pretty amazing what you can pick up just from looking at the pictures and trying to get the general gist of what's going on rather than getting caught up in the details. But the details have to be there -- without the precision and proper terminology, these articles would not only be wrong; they would also be useless to anyone trying to actually apply them.

This uses computer model predictions, not observations. While the forecast may be more accurate towards the center of the hurricane, I can say that it isn't very accurate in many other places. Particle simulations are great though! And the general aesthetic is very pleasing, but I fear that such things trumping good data understanding are going to be the norm going forward.

Data presentations should always be both aesthetically considered and correct, the right aesthetic can aid massively to help interpret the data. See Envisioning information by Edward Tufte http://www.edwardtufte.com/

I don't understand why news organizations haven't adopted this kind of visualization yet for hurricanes. This has been around for awhile, and I always pull it up when a hurricane hits the US. Its a far better way to display wind than the standard precipitation-oriented maps.

This is pretty cool. You can actually see the impact of Sandy on the East Coast live as the winds speed up... hopefully that one turns out to be less bad than the forecasts are saying.

It's hard to judge speeds from this, and they're often more important than direction. It's beautiful, but it doesn't tell me if the next high tide will have an even higher surge.

This is beautiful, although it would be useful to mark a few more major cities on the map.

More cities show up if you click to zoom in. Some cities may take a few zoom levels to show up, though, like my native Amarillo.

> ... it would be useful to mark a few more major cities on the map.

It would. And some of the ones that are marked are kinda pointless. Anyone with any knowledge of U.S. geography can find New York, San Jose, and Seattle by the shape of the coastline. But what about Salt Lake City, Kansas City, Memphis, Omaha, Louisville, Indianapolis, St. Louis, etc.

Regardless, very nice visualization.

Why is he wind going backwards on the East Coast?

There's a hurricane centered below New York right now that is drastically altering the wind patterns.

Unless you're seeing something I'm not...

The actual satellite view of Sandy shows most of the wind flow directly onshore from New Jersey towards Philly. This app shows it flowing the opposite direction, unless my eyes are playing tricks. The white areas are flowing towards the East.

Absolutely beautiful.

Reply on news.ycombinator.com