Folding Paper Globes | Hacker News Reader

Folding Paper Globes

foldingglobes.com

193 pointsdango250641 comments

They could get 10% bigger globes out of the same sheet of paper by rotating the icosahedron a bit (14º)

one of my favourite hacker news quotes. what makes this forum great.

Why not 30°? Is 14° as far as you can turn before the ends of icosahedron hit the top and bottom of the landscape sheet of paper?

Neat! There's a notable omission though--none of the earth globes look like earth does from space, with true color* AND clouds. A while ago I got really into making paper globes because I wanted one like that and I couldn't find one I liked. I asked Claude to write a script to generate a net out of this: https://web.archive.org/web/20250316112538/https://visibleea... Instead of an isocahedron, I used gores like a real globe. This approximates a sphere using intersecting cylindrical surfaces (not sure what the term for that solid is!) I made a bunch but usually they came out pretty lumpy, here's the best one: https://imgur.com/a/eikGrTW

*Edit: I also couldn't find a good point of reference for what "true color Earth" even looks like. Different photos/satellites seem to have notably different color calibration, just google "blue marble" and you'll see. And this isn't even getting into the weeds about atmospheric effects, which appear strongest near the limb... obviously there's no way to replicate that on a globe!

Flexify (http://www.flamingpear.com/flexify-2.html) is an incredible decades-old Photoshop plugin for making these. I made a mobile for a grade school math project using Flexify and NASA's equirectangular planetary maps, turning Mercury into a tetrahedron, Venus a pentahedron, etc.

The gallery of output projections is neat: http://www.flamingpear.com/flexify-output-modes.html

The author of Flexify is

https://news.ycombinator.com/threads?id=lloydb

That's slick. I use papermodel+Blender for my models, but I haven't bit the UV mapping bullet yet!

Neat. For folks who are more comfortable with using the terminal than paying for things, I have used this python libraries to make isospheres from 360 photographs: https://github.com/Teque5/pydymax

That sounds really cool, can you expand a bit on the workflow you used: what format were the 360 pictures in and how did you convert that into data that this Python library uses ?

Try interrupted sinusoidal, ot will be less edgy.

https://www.jasondavies.com/maps/interrupted-sinusoidal/

This is a beautiful site. Don't miss it if you haven't been there before

TFA has the option to do the same thing. Default TFA page shows the free option. The less edgy options are for not free but available.

You are indeed correct. The options are to be found here

https://foldingglobes.com/pricing

For those who want some more, try making a globe by hand. It is instructive and a lot of fun.

https://observablehq.com/@mxfh/a-papercraft-friendly-world-m...

I made a custom projection once, that's a bit friendlier to fold and comes in one tile.

Has no rendered flaps, but you should get the idea where to place them.

Still somewhat stiff, but managed to actually build two of them, would recommend printing on heavy A3 sheets or bigger.

Anyone is welcome to iterate. There even is a todo list with obvious things to improve upon.

Very interesting. Thanks for sharing. I haven't looked at the pointed site yet but will do once I get some time on my hands. Maps end up nerd-sniping me.

The offset to prevent tears is a nice touch.

The offset also forces the gore to be bent in two directions, so it’s also bit tricky, and makes the gore stiffer then it needs to be, dampening the paper there certainly helps.

Maybe I'm overlooking, but would love if this had an accurate political globe (e.g., countries and capitals) for my young kids.

Borders are a nightmare because many countries have contested areas and ask everyone inside the country to use the official one, specialy at schools.

Too late to edit: As a sibling comment says, in the top of the site they offer an option to make a version with only the borders. I don't see an option to combine satellite photos and borders anyway.

I do find it interesting that while modern map apps do a great job of replacing road atlases, and we can get satellite and photogrammetry overlays that give us immense insight into how maps relate to the context on the ground, and we do have pretty good topographic mapping apps (though sadly not built in to the big map apps like Google or Apple), but one kind of map that we used to get in paper atlases but can’t explore so easily online are political maps. Google Maps is also not great as a physical map in that it is bad at labelling and naming geographic features - Apple Maps does a better job at this. But neither offers a good way to hide all the roads and just look at borders.

It's great for learning historical geography. At my school in the late 90s, our atlases still had the USSR.

"or with your own pinned places via the custom globe builder."

Right there at the top of the page with a link even: https://foldingglobes.com/

In fact, it looks like the default landing page for the site. TFA looks to be a quick link to the free options

I hear you, but I'm suggesting a preconfigured one like the others that are linked here.

by pre-configured, I'm reading as free since all of them are pre-configured and are just a button click away.

The site doesn't give you printouts you can fold into a globe, but it is very informative. You highlight countries, drag them around, and their size adjusts to whatever degree of distortion you'd get for any land mass at the given longitude. So you can, say, highlight Greenland and drag it on top of Brazil and see how they compare. You can highlight and drag countries, US states, Canadian provinces, and Australian states. Hovering over an entity shows you area and population.

Another thing I miss in common digital maps is the use of typeface sizes and weight to approximate city populations. There are many other ones, like toll road, multi-lane, two lane, paved, unpaved, under construction/washed out, as well as better communication of biome characteristics like forest, marsh, swamp, desert, farmland, lava flow, (plus the usual parks, preserves, wilderness, etc.)

Paper maps communicated a lot of different information, not only things about roadways. It's like roadmaps are the only maps the common online digital maps cared to build digitally.

Their keys and icons were so much better.

fil's AirOcean projection on the original ObservableHQ is a fun starting point for rendering your own such map in Javascript from scratch:

https://observablehq.com/@fil/airocean-projection

The source data is down for my vector tile remix of it but here is a screenshot: https://static.observableusercontent.com/thumbnail/0fac8a928... and the code: https://observablehq.com/@thadk/reprojecting-vector-tiles

I did it as a little MapTime project in 2019 or so.

Found some imagery:

https://mapscaping.com/diy-folding-paper-globe/

Not same source but related:

https://goodwoodglobes.com/guide/folding-globe

https://lyncean.education/projects/astronomy/polyhedral-glob...

I would like to see some pictures of the final product, seems very cool

Interesting idea, but are there any photos of the assembled globes anywhere? I’m on mobile but I can’t seem to find an example on the site.

very cool. as someone not familiar with this I would love to see how they look when folded

That they don't show how they look when folded is either bizarre, or that the end results are disappointing.

Great!

> Meteorite Landings

There seem to be significantly more of those on the northern hemisphere. I wonder what might be the reason for that?

Meteorites landing in the middle of a large sea, or in unpopulated areas like Antarctic, are less observed.

it's a good idea, but it uses the distorted Mercator projection, and then cuts it so that you can glue it back into a somewhat globe-like shape.

it's distorted, twice.

Are you sure? Greenland looks quite small and in https://foldingglobes.com/ there is an option to print the borders with parallels and meridians and they are split and curved.

i didn't spend much time with it beyond noticing that africa was tiny.

Africa is fine in Mercator, the problem is that Europe is too big in Mercator!

Some photo from a satellite. https://www.sciencephoto.com/media/83403/view/africa-satelli... Note that Africa is quite similar to the image in a Mercator projection.

There is a lot of fuzz about https://en.wikipedia.org/wiki/Equal_Earth_projection but if you see the image in "Formulation" it shows that Africa (and all places near the Equator) are stretched.

> Africa is fine in Mercator

here's a reality check:

https://licensing.visualcapitalist.com/product/how-many-coun...

or just find a globe, or zoom out in google maps.

also, what you write doesn't make much sense. the problem is with the relative sizes, with the non-linear distortion. regardless of which zero point you pick.

Have you looked at https://foldingglobes.com/how-it-works It compares a globe and a rotating version of the icosahedrum. Also read the description there, the distortion is small.

If you look at https://en.wikipedia.org/wiki/Extreme_points_of_Africa considering only the mainland that is easy to see in the map:

* from West to East covers 69°, i.e. 1/5.2 of 360°. In the maps of foldingglobes the width of Africa is close to 1/5 of the map. The same applies to the other maps, I don't see too much disagreement on the width.

* From South to North, coves 72°, i.e. 1/2.5 of 180°. In the maps of foldingglobes the width of Africa is close to 1/4 of the map. The same applies to the other maps, I don't see too much disagreement on the width.

Here I pasted a 4x2 grid of Africas over the map https://imgur.com/a/huMJgbG I hope it's clear there is room for one more copy on the side (like half on one side and half on the other) and half a copy at the bottom, so I consider it a 5x2.5 grid. So the size on the map is pretty close to the globe.

>> Africa is fine in Mercator

> here's a reality check:

I don't understand what you mean. Africa is near the Equator and is fine in the Mercator projection. A circle is mapped to almost a circle. The problem is that Europe and North America have a bigger scale. (I'd try to print the maps and make a collage later.)

An interesting comparison in the map you linked is Norway, that is very North and has a huge scale difference in the Mercator projection.

In the map you linked, if you move slowly Norway from the center of the Sahara to the North, when it reach the cost it goes approximately form the southest part of the gulf in Libya (near Rā's Lānūf) to the frontier between Morocco and Algleia (near Nador).

Using a plastic rule on the screen on the screen on my computer in whatever resolution I have now in Google Maps (Mercator), I get 5.5cm from Rā's Lānūf to Nador. But when I measure Norway with the same method I get 9.5cm form Sandnes to Berlevåg. That shows that the scale of Norway is like x2 the scale of the north cost of Africa, but I think we both strongly agree in that Mercator use different escales.

If I repeat the same measurement in the map in foldingglobes, I get 3cm for Norway. The cost of North Africa is hard to measure because it is split, but it looks like 4cm. Can I claim a 1 to 1 scale? I think the north cost of Africa that I'm measuring is slightly bigger in the map. I guess the cost is closer to the edges and then they have a slightly bigger escale.

Anyway, the conclusion is that the huge x2 difference of escale disappears.

For completeness:

If I use the distance measurement in Google Maps I get 1700km for Norway and 2000 for the part of the north cost of Africa. So close enough for me, not a huge x2 mistake in the comparison.

If I use a plastic ruler in the map you pasted I get I get 6cm for Norway and 6cm for the part of the north cost of Africa.

Earth without water seems a bit deceptive.

this is awesome !

Reply on news.ycombinator.com