Hacker News new | comments | ask | show | jobs | submit login
Google Maps is no longer a flat map (techcrunch.com)
339 points by kylesellas 6 months ago | hide | past | web | favorite | 161 comments



Google Maps must be one of the modern wonders of the world, it's still nothing short of breathtaking to zoom out of streetview in crowded London and then land somewhere on a country crossroads in remote Vietnam.


I count with one hand all the technologies that have blown my mind: discovering the electronic parts of my toys, playing a NES game in my friend's PC, my first contact with the internet, the first iPhone. But discovering Google Earth was a really special one, suddenly I was able to explore every street in the world, I was able to look at my house (in a small city in a developing country). Then I started imagining all the possibilities, also my conspiracy side kicked in, I mean, I as a teenager with a Celeron PC was able to look at any place I wanted in the world, then what kind of secret technologies could a big government have?? Good old days, I hope technology will keep blowing my mind in the near future.


There was a guy on a TV documentary once that had worked for the government in some capacity. He was explaining that the satellite maps provided to the public are extremely zoomed out compared to what the government is able to see.

He said that the government had photography so detailed, that they could find a golf ball on a golf course, and even tell you what brand of golf ball it is.


There's always some dude on TV bragging, but I'd be skeptical.

https://www.quora.com/How-powerful-are-the-cameras-on-modern...

TL;DR is that satellites have a 9 cm resolution at best. That's not enough to see a golf ball (4 cm diameter), and far away from discovering a producer's imprint on it.

With drones you can get an order of magnitude better, but that's still only barely enough to find said golf ball, and not enough to read fine print.


Probably right. But my guess is that the actual ability of current satellite photography is highly classified.


These pictures are from 3D flyovers, not satellite. They usually flyover every year, mostly municipial, not by country.


Who said it has to a be a satellite?


jjeaff did.

> He was explaining that the satellite maps provided to the public


> compared to what the government is able to see.

That does not imply satellite maps.


I guess you can read it that way if you try. But neither I or the other commenter got that meaning.


> if you try

Please refrain from baseless accusations. Thank you.


the government's spy satellites have very high resolution, but that claim was just bogus.


And once you land on those country crossroads, you can continue exploring the area at the street level. It's really an incredible tool for sating wanderlust - just pop over to some place you've wanted to visit and wander the streets via streetview, all from the comfort of your seat!

Speaking of which, GeoGuessr [1] is a great example of how to enjoy Google Maps. You start at some random streetview and try to figure out where in the world you are by walking around (and hoping to see a sign or two)! A lot of times you end up in dirt roads in small countries, and it's really cool to see what it looks like there.

[1] https://geoguessr.com/


As far as I know google street view costs money from this month, and it's not exactly cheap, $14 for a thousand streetviews:

https://cloud.google.com/maps-platform/pricing/sheet/

I wonder if this tool stays free, lots of site owners are complaining about paying much more with the new pricing.


It'll be an interesting thing. They definitely understand the gaming angle due to Pokemon Go causing them to specially create tools and customize their API for gaming: https://cloud.google.com/maps-platform/gaming/

But that seems to be Enterprise level only.

They seem to have already updated for the new level. They also have advertisements and are a profitable business.

My guess is that they "Contacted Sales" like the Gaming API asks you to do and have a very agreeable discounted rate.


I've never seen a third party site using streetview so not much will be lost. But considering the effort and volume of data they have meticulously collected you can't blame them for monetizing it directly. Google is going to learn their lesson with the insane pricing on regular maps, though.


They could monetize it by putting ads on the maps/streetviews. They are an ad company after all, so they should have an option to show a map with ads for free, or without ads, but then it costs money.


This is fantastic!


It is, i remember being captivated by this film in the (mid? late?) 90s.. (from the late 70s, pieced together by hand). Never imagined it would be (partly) possible in realtime.. some 10 years later.

(eagerly hanging out for orbit view milky way, but probably not in my lifetime)

https://www.youtube.com/watch?v=0fKBhvDjuy0


Thanks for posting that - I was thinking about it the other day when I saw this[0] posted in another thread here. Great for sense of scale, but no "street view" in this one.

[0]: https://www.youtube.com/watch?v=q1mkjkTqg0Y


Google Earth VR is mindblowing.


Holy Shit! I just youtube'd some videos about that and it looks amazing! This is the first thing that has really made me want to get VR.

Example: https://www.youtube.com/watch?v=SCrkZOx5Q1M


Big picture I think it's truly one of the high water marks of human achievement


Was going to make the same comment. I was pretty stunned by it.


Thanks to this comment I just spent the last hour exploring Moscow, Tokyo, Kyoto, and the Dominican Republic. I feel like I've always taken street view for granted. But it is a truly incredible experience to drop into a city you've never been to and see things you wouldn't see even if you visited.

Granted, nothing compares to actually experiencing the culture and places in person, but street view makes for a completely different, incredible experience.


I tried something similar to Google Earth in the late 90's at SIGGRAPH, running on an SGI Origin or similar. Was gobsmacked as I "parachuted" from orbit into Dodger Stadium. :D


Google Earth originated from a team that spun off from SGI (twice, kind of), so you probably saw the demos by Michael Jones and co.

Nowadays, you can build or buy a Liquid Galaxy to get a better experience than the Origin's, at a fraction of the price: http://liquidgalaxy.org/


By the way, for anyone who prefers Google Maps to load fast and show actual satellite/aerial imagery (rather than trying to render each tree as a weird crumpled blob) there's a flag you can set to get the old, fast 2D mode:

https://maps.google.com/?force=canvas


Thank you!

I'm actually quite annoyed that Google took away the visible option to disabled the 3D aerial view. It's such a resource hog, and my work's internet is often quite slow, and 3D has far more data to retrieve.


The option is there. It's in the hamburger menu, labeled "Globe".


[flagged]


No personal attacks on HN, please.


I think it's worth replying to this to reinforce the point that the new mode has technical disadvantages compared to the old mode. Some of my projects involve a great deal of time spent looking at aerial imagery, sometimes for hours a day, and while Google Maps is not the panacea of aerial images it is one of the best single sources in terms of usability and quality of images.

That said, prior to knowing this trick I was stuck with the effectively unmaintained Google Earth instead of the web view primarily because of the web maps insistence on the 3D rendering mode. The attempt to conform buildings and, most annoyingly, trees to the elevation data makes the imagery almost useless at close zoom levels, as all the actual detail disappears into clutter artificially introduced by the edges on the elevation data. It significantly reduces the amount of spatial resolution you can get out of areas where they have very good images.

Earth still has tools that web view doesn't have, but often I'm not even that interested in measurements and just want to be able to carefully inspect the image. This doesn't seem to be something Google wants to support any more!


It's not elevation data, it's reconstructed geometry.


My assumption had been that Google built these models using high-resolution LIDAR data, which is why I called it elevation... on a little more research though it looks like I was wrong and the models are perhaps constructed based on street view images and model data for other similar objects.

Honestly, I feel like this makes the situation even more indefensible, because I get that collecting high-quality DEM is very expensive and so trying to use it comes with a lot of compromises. But if they're just substituting in models of "similar trees" like one person on Quora suggests it seems almost embarrassing for them that they're making the user experience more frustrating in order to deliver a 3D street-level experience that's still so bad as to be useless.


It's more complicated than that. Among other things, it led to the open sourcing of Ceres: http://ceres-solver.org/users.html (a few clues on that page)

You definitely can't use Street View data alone for objects not at ground level. Actually, sometimes there is no Street View data for a place, at all. E.g. check the planes on the runways at airports like SFO, AUS and JFK. Incidentally, the varying quality at different airports is also a hint of how easy it is to fly over each of them at the ideal altitude(s) and how often one can get the necessary permits to refresh imagery.


People wanting an option for efficient systems is not the same as using an archaic browser.


You can also just use the "Globe" button in the side panel to toggle flat satellite imagery, while still keeping the smooth zooming.


For those like me who had trouble finding it, it's in the hamburger menu.


I particularly like the transition when zoomed out and toggling the Globe button.


Thanks, I delayed the release to squeeze that in :)


A breath of fresh air, being able to visualise our place in the world as it truly is.

I think it's all fantastic, extremely fast, smooth, love the inertia on spin, the camera movement or simulation of is great, occlusion of objects nicely done.

Would like to see an example with clustering.

Is it possible to add elevation data ?

Is it possible to access this mode in the API to use the webgl version ?


The alternative to switch back is: https://maps.google.com/?force=webgl


Thanks a lot. I was afraid that I was going to be stuck on the canvas mode. Sometimes, my curiosity gets the best of me :(


Crumpled weird distorted trees and houses have to be one of the worst features of the newest google maps.


I still don't understand how could anyone find this defacing useful.

Just give me the satellite or the Street View photos.

TBH Bing's "Bird's Eye" view was far more useful that the shitty resolution (mandated by law BTW) satellite photos.

Of course, Microsoft killed it :(


This is actually slower in many cases. If you want to view 2D imagery and/or Mercator maps you can simply enable/disable Globe mode from the hamburger menu on the left.


That does seem to show older images though.


Agreed, the images do seem lower quality than they used to be. I was curious, though, and chose buildings at random around the Google office in NYC, just to do a side-by-side, and even the old pictures look better:

https://i.imgur.com/0xEG4pb.jpg

https://i.imgur.com/bxwr3Ez.jpg


I'm assuming left is old and right is new? This looks like aerial photography shot from a plane rather than satellite imagery, which seems like it would depend more on the quality(?) of the plane than the quality of the cameras.


"On flat maps, it’s impossible to represent land mass size on a relative scale."

Um, no it's not. TechCrunch needs better editors. You just need to use a non-Mercator projection.

https://en.wikipedia.org/wiki/Gall%E2%80%93Peters_projection


You can represent size or shape correctly, not both.

Anyone who uses the Gall-Peters projection is a cop.


> You can represent size or shape correctly, not both.

You can if you don't mind having a non-rectangular map border.

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


Dymaxion maps are still deformed, just not as much as most other projections.


You can get pretty close though. Like a Dymaxion map (shown in sibling's XKCD link) or one of those orange-peel maps if you have enough little slices.

Edit: better Dymaxion link https://en.wikipedia.org/wiki/Dymaxion_map It's made of triangles that you can rearrange, which isn't really shown in the xkcd.


Obligatory XKCD

https://www.xkcd.com/977/


Google decided to go with "A Globe".


At low zoom levels.

Given the behavior of the poles (they're not viewable; the south is also a black hole…), I suspect that there's a very smoothly done transition to web mercator for the higher zoom levels.


Correct. At high zooms, the trig required in the spherical math becomes inaccurate, so we fade back to simple web mercator.


Cartography wouldn't be able to use all these cheap tricks if the poles were habitable territory.

Can you imagine the distortion on a map of a small town located at the pole if Mercator was used?


That's still not an accurate map. 3D maps don't have projection issues, they can have near perfect representation of the world.


The eyes see a 2-dimensional picture. Computer displays produce a 2-dimensional picture. There is no such thing as a “3D map” on a standard computer display. Only an orthographic (or maybe perspective) projection which animates in response to user input.

In general an orthographic projection is a poor choice. The thing going for it is that it is easy to understand by analogy to a physical globe, and human spatial understanding can partially compensate for the distortion by imagining a sphere.

But there are also disadvantages, for example on an orthographic projection north is not always vertical, latitude and longitude lines are neither perpendicular nor straight, shapes and areas can be very dramatically distorted near the edge of the map, there are dramatic distance errors between far away points or points not in the center of view, it is impossible to view more than half of the planet’s surface in one view, etc.

There are many other choices of map projections, and for any particular purpose, some other projection is usually better.


Regardless of what the eye sees, the brain perceives three dimensions. This projection happens to be the one that the brain can automatically unpack into an accurate 3D representation. Technically, things near the edges are squished and distorted, but that gets undone in the wetware.


Let me give you a concrete example.

Here is what North America looks like on this new map:

https://i.imgur.com/lCI8DjY.jpg

This is not a good projection to use if you want to see all of North America. There is dramatically more distortion than necessary, and shapes near the edges get very squished (the west coast of the US is okay), but it is not at all obvious to a naïve viewer what kind of distortions are involved, and past experience looking at soccer balls or whatever isn’t going to fix it.

(Note: a Mercator projection is also not a good choice for this. On a Mercator projection sizes are dramatically bloated as you go north.)


I'm not convinced it isn't still the best option. The naive viewer may not properly perceive the distortions, but I bet they'll get closer than with anything else, for general purposes.

What projection would you say would be good for that view?


For North America per se you could e.g. pick the conformal projection with roughly constant scale around the boundary of the continent (one world atlas does this), or you could pick a projection which minimized distance errors (there are several possible ways of quantifying those, which will yield different maps), or you could pick among several equal-area projections (especially useful if you want to make a choropleth map), or for something dynamic you could use an azimuthal equidistant projection or some conic projection (e.g. Lambert conformal conic, Albers equal-area conic), or .... https://pubs.er.usgs.gov/publication/pp1395

There are many choices. An orthographic projection is generally one of the worst except insofar as it gives the impression of a physical sphere when you zoom all the way out.


You really can’t get an accurate sense of angles/shapes, areas, distances, or directions with an orthographic projection except by moving the view around. Even when you do, it is very difficult to accurately compare attributes of regions which are far away on the globe, and easy to mislead yourself. (This is a problem with physical globes as well.)

For many types of mapping use cases, orthographic projections are substantially worse than other possible choices.

Sometimes physical spheres (or virtual representations of spheres as dynamic orthographic or perspective projections) are nice. That’s why I spent a whole bunch of time building myself a spherical chalkboard, for example. But they aren’t the best choice for any purpose. Personally I don’t think they make a very good choice for a general-purpose world map. They are better as a supplement to flat maps.


I'm amazed this has been downvoted so much... As the OP stated, "globe" view (a.k.a. orothographic or perspective, depending) is not fundamentally better than other projections. It has tradeoffs. You should always choose the projection that matches your use case.

A map is a visualization tool. The goal is not to reflect reality as accurately as possible. The goal is to communicate data and spatial relationships. Projections are part of that. There is not a "best" projection. Orthographic with some animations is a nice visual tool for some purposes and terrible for others.

How could pointing this out be controversial? A bar graph is not a perfect representation of a group of numbers, but it's a very useful visualization. A map is no different than any other visualization.


You can, of course, use VR to get a stereoscopic 3D display as well, like what humans see in the real world. Except for depth of field (which some prototype VR systems can still accomplish), that's as 3D as real life. And Google Maps/Earth VR can indeed do that.


what about focus?


Some VR companies are working on that.


it's a theorem of differential geometry that we cannot have a (perfectly) accurate 2d map of the earth.


The author of TFA didn't write "perfectly accurate" but "land mass size on a relative scale". Bud is correct.


the author of TFA probably thought the desire for correct shape in a map is implied and wasn’t trying to defend against map pedantry. of what use is a map if the shapes are wrong?


One of the revolutions in the first Underground/Subway/Metro maps was that by throwing away the actual shapes, it is able to represent the stations and the relationships between them far clearer than otherwise.


a subway map has little to do with a geographic map. only the nodes and time traveled between them matter. the shape of the actual routes is immaterial.


It's much cheaper to make an incorrect statement and let people correct you for free, then edit the article! ;)


The visualization is really cool but for daily use seems more practical being able to plot long trip routes without them disappearing on the back of the globe than being able to assess the true size of landmasses: https://www.google.com/maps/dir/Madrid/Sydney,+New+South+Wal...


I get what you're saying, but I'm not sure if planning trips to the other side of the globe qualifies as "daily use" either :-)

I think it's neat to see that the route is along a great circle -- that wasn't clear at all before this change.


Unfortunately, the line drawing doesn't make this as cool as it could be. The line representing the flight from BOS to PEK is drawn due west, as would look natural on a 2D map. The real flight (and shortest route) heads north and passes near the north pole.

Hope they can update this, which would make international flights seem even more impressive than they already do.



Yeah the flights in particular need to be updated for this projection.


Millions of people do it everyday, so it seems right to call it daily use: https://twitter.com/flightradar24/status/1018260113814061057...


>On flat maps, it’s impossible to represent land mass size on a relative scale.

This isn't right. It is impossible to represent both relative position and relative size accurately. Gall-Peters and other equal-area projections accurately represent relative sizes.


There are cartographically reasonable equal-area or nearly-equal-area projections. Why are we still talking about Gall-Peters?

Peters was a charlatan who knew very little geography (he didn't even want to give credit to the geographer who discovered his map), and the Gall-Peters projection is Mercator for old-school hippies who think they can promote equality by making Africa taller.


Expand?


Yes, but only vertically.


I was trying to ask the parent to expand on the many things he was claiming lol


Here's a couple views on Peters and his map projection.

An article from "The Map Room" [1] on the argument about Boston public schools having adopted that map, which really dominates Google results right now I guess, but even within the lens of that topic it hits the major points.

And here's a "New Internationalist" article about Arno Peters [2] -- it sounds like he was fine as a historian, and I even agree with his motivation for not wanting to plot historical events on a Mercator map, but it sure would have been great if he had learned about the equal-area alternatives to Mercator that Gall and Mollweide and Eckert and other geographers had already made.

It's not that I'm against making a map projection to "question Western imperalism". I take more issue with Peters' questioning of basic geometry -- he claimed his map preserves angles, which is clearly false (imagine a NW-SE road meeting a NE-SW road in Ecuador and what that right angle would look like in Gall-Peters), and he claimed that it was the "only" equal-area projection.

And I don't think it actually advances the conversation about equality to make a map that looks right in Europe and stretches Africa and South America like taffy.

EDIT: left out the links

[1] https://www.maproomblog.com/2017/03/the-peters-map-is-fighti...

[2] https://newint.org/features/2003/01/05/arno/


I wonder what the long term geopolitical consequences will be when we all stop looking at projected maps. That small peninsula to the west of Asia just got a little smaller. A certain island kingdom is now more clearly the size of Uganda. Maybe Google can also flip the globe "upside-down" for all users "below" the equator.


I think it accomplishes the opposite of the intended political goal. Larger regions aren't more loved. Larger regions look more threatening. It is now far more obvious that people near the equator can't all migrate to the EU and US. The US and EU look much more vulnerable and precious.


Honestly, I doubt there will be any long-term geopolitical consequences. It's comforting to imagine that we can change the world just by more correctly representing a map, but I suspect the reality is that we can't.


Do we really think very much is really affected by map projections?

Korea looms large in people's imaginations for a lot of reasons that have nothing to do with its occupied land mass size, and a lot to do with their history and cultural and economic power.


If Korea is not occupied by the U.S. and allowed to unite then it would take a decade or so before their economy would be number two in the world with them kind of needing to be on the U.N. Security Council.

The same can't be said about if Sudan was united again, Sudan would not be magically punching above India, Russia and Germany in international commerce.

The world where we see the world as it is, a sphere that we can interact with and with that sphere being the default view has been a long time coming. We will look back and wonder why we got the perspective wrong for so long. There are no other objects we display in Mercator Projection et al.

If we could go back in time with what the fine people at Google have achieved and give it to navigators of the past, in the days before aeroplanes got popular, then all the hacks that 'demand Mercator' would never have arisen. Maybe we would have to teleport GPS back to this fictional past too.

The none-sphere maps make as much sense as those paintings done before perspective was worked out, we have just ended up believing them out of familiarity and due to the navigation and other hacks devised for the atlas makers of the past.

As recently as twenty years ago the idea of just showing the earth as a sphere for things like weather broadcasting would have seemed a radical and controversial idea. Thanks to Google and de-facto Maps we have made a small but important evolutionary leap in understanding our world.


>Maybe Google can also flip the globe "upside-down" for all users "below" the equator.

I must be the only weirdo who wants the opposite: equator up. South-up in the Northern hemisphere and north-up in the Southern hemisphere.

Why? Because I want the Sun at the top of the screen (or more accurately, the sunny side of buildings/hills/anything).


Then you'd have to special-case maps that include the equator itself.


Indeed, but this is no different from "north up in Northern hemisphere / south-up in Southern hemisphere" proposed above. Since we're talking interactive maps, you can probably just go by the center point (barycenter) of the viewport.

W/r/t Sun angles, things are pretty weird anywhere between the Tropic of Cancer and the Tropic of Capricorn.


A generation growing up on this will probably be less likely to believe in a flat earth, since the intuition will develop from the time they first start looking at maps.


Globes (the good old physical item) exists since ~150 BC


Australian maps are shown with Oz in the middle, and the pacific / indian oceas each side, with South East Asia and Antarctia surrounding, exactly the same way US maps put the US in the center.


Do you mean whole-world maps have Australia in the center? (I assume so, since obviously a map of only Australia would have Australia as the center...)

That's fascinating! I grew up in the US and I'm not sure I've ever used a world map with the US in the center. Curious where you got this impression. I can find some examples if I specifically search online, but they look very weird to me.

Every world map I grew up with has the US on the far left, presumably due to the convenience of splitting the map on the Pacific Ocean. The first example that came to my mind: https://i.imgur.com/FjBKD6k.jpg


Maps in Japan have Japan and the Pacific ocean in the center and the Americas off to the right. Which make the phrases "the west" and "the east" kind of interesting choices.

Example of a map poster for kids: http://happylilac.net/sekaitizu-mihon.gif


There is something weird going on near the poles still, so I'd not be surprised if its technically expensive to make an "upside down" map.

Upside down and mirrored should be simple though.


Very cool. However it's not using great circle routes for airline directions, for example ask to go from Calgary to London. It should go well over Greenland, but instead just follows the same path you'd get from drawing a line on a Mercator map.


Eastbound flights often take non-great circle routes in order to make better use of the jetstream - depending on the forecast, a straight eastward line may be more accurate than a great circle.


Weird you're right, however if you use the measure distance tool it does follow the curve correctly.


>the path you'd get from drawing a line on a Mercator map

Fun fact: this is called a rhumb line, ie a line of constant true bearing.

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


It even allows you to see satelite pictures of other planets! I'm not sure how old is that feature, but it's awesome.

You just have to unzoom far away from earth and it'll show you a menu where you can choose where you want to go.

https://www.google.com/maps/space/mercury/@47.5337211,-157.4...


And interior of the ISS :). Thanks, that was a fantastic journey!


While I was working on the PowerBI globe map visualization https://powerbi.microsoft.com/en-us/blog/visual-awesomeness-... (video in link)

I had to do some funky magic to stitch square tiles in a sphere from bing maps and do reverse mercerator projection. I had to also ensure GPS coordinates would exactly align to the right pixel otherwise the 3D chart would all look weird. So there was a lot of fun math involved over a bunch of evenings and weekends.

The hard part though, was to do an animation that would unravel a spherical earth into a flat earth and vice versa.

Naive me though it would be a simple linear interpolation that would take a couple of minutes but it ended up being many many nights of crazy mathematics to get it right.

Map projections are really interesting. Much respect to their inventors.


I'm interested in learning more! I imagined linear interpolation could work too. What went wrong?


This does a hemisphere but doesn’t map lng and lat. Gives an example of the kind of maths involved. https://math.stackexchange.com/questions/1334241/interpolate...

It’s not a simple linear x0, y0, z0 -> x1, y1, z1 because you kind of want an animation like a square blanket is wrapping around a ball, stretching the right amount in different places such that it becomes a sphere.

I tried to find a single magical equation that would give a x0, y0, 0 -> x1, y1, z1 based on param t that goes from 0 - 1. I’m sure there exists one but I tried all sorts of fancy math I knew for a day or two and still had major trouble writing it down as a single equation function.

Then I thought about it in a different way after I ate a fruit and realized I need an orange analogy. What if we cut slices of orange. The ends are crescent shapes. Now it’s a matter of animating the crescents into rectangular strips where the pointy ends stretches but the middle part remains the same.

Since in computer graphics, a sphere isn’t really a true sphere, but an approximation with triangles the strip analogy works well. We just need to animate the ends of triangles.

The gotcha is the pointy end of sphere expands from 0 to length of mercerator projection (flat square) so they cut a tiny bit of the top and bottom otherwise you have infinite stretching at the end

So you have three little functions. For an x,y in a square you figure out what slice they go into and what side of the slice. The further to the poles, the more you squeeze towards the center, so you get something like this https://blenderartists.org/t/producing-a-12-petal-gores-from...

Then you have a blanket wrapping function I.e morphing a line into circle and that’s a line -> arc -> circle animation. That’s not too hard. The arc of a very large circle is almost flat. So you divide the line into two and animate the left and right segments into two hemicircle arcs that then form a proper circle. Kind of like curving the lines from the end to bend it into a circle.

Now longitudes and latitudes both go through this line to circle animation. To get the z, we take the x arc position, the y arc position, the point on crescent strip and we do matrix rotate transforms to get final x,y,z

So rather than it being one single equation (which i’m guessing is possible using super crazy maths), we take a piecemeal approach and animate lots of little segments that add up as a plane to sphere animation using simpler maths and matrix transforms.


Very cool! It's always seemed a shame to me that we stick with highly-inaccurate projections when we have the tech to do much better, simply out of inertia. Double-cheers as well for ignoring the weird convention that you shouldn't care about accurate sizing when looking a political rather than satalite map.


Looks like this isn't rolled out to everyone. I'm not seeing the 3D version (logged in & incognito).


on my work computer, which is tunneled through san francisco, so i show as in USA, i can see the 3D map.

checking maps on mobile which is on Canadian ISP, in chrome and the google maps android app, i can only see the "flat" map.


Thanks. I thought it was a "browser thing". My desktop firefox shows flat maps only.

-ss


i can confirm that chrome, firefox and IE browsers show the new view with my location as USA only.


For rollouts like this, a small holdback experiment is typical, to assess long term trends. Or maybe your browser doesn't have hardware acceleration enabled? Or you could be in canvas mode? Try ?force=webgl on the end of the maps url in case it's the latter. Try shadertoy.com to see if webgl is available.


Webgl works on my browser. Shaders work. Google maps is still flat. force=webgl does nothing. I'd say it's a partial roll out.


?force=webgl fixed it for me, thanks


I don't see the map either (and I'm in US).


I see it in Europe. Probably affected by your browser and/or extensions.


It works on Chrome. Did they disable it for FF?


If there's a job where I just get to explore the wonder of Google Maps all day, I want it.


Maybe the Google Maps QA team?


Taxi driver?


Finally we are rid of the biggest source of Mercator! Finally we can look at the south pole!


Actually no, McMurdo remains a white sport still :(


This is an unfinished demo from a while back, but I made vector and raster map tiles in a WGS84 projection, so McMurdo is included — though you have to find it yourself [1]

I'm unlikely to work on the globe, it's cool[2] but we don't really need it. Maybe when OpenLayers Cesium[3] gets support for vector tiles, so we don't have to see fuzzy, distorted raster tiles.

[1] https://tile.gbif.org/ui/Globe/temp-debug.html (click "Toggle" in the top right)

[2] https://tile.gbif.org/ui/Globe/

[3] https://openlayers.org/ol-cesium/


I can see a lot of individual buildings there, and I can see facets on the dome. There's even Streetview!


Unfortunately not. The behaviour at the poles suggests that the whole thing is merely a hack.


Openlayers Cesium provides a globe view for map tiles (or other data), it's it not to much work to transition to a flat projection at a high zoom.

https://openlayers.org/ol-cesium/examples/sidebyside.html


I find it curious that it loads the old 2D representation first -- made me think it wasn't working for a moment or two.

Also, is it just me or is zooming a bit broken at low zooms now? The globe drifts as I zoom.


So it's no longer a map application, because it's producing weird non-standard projections, tiles that can't be stitched/georeferenced properly etc.

Will Google support and maintain an actual maps application? Or will they merge google earth and google maps together in an even less capable tool purely for consumers to wow at? Castrating Google Earth with the webgl version and locking it into Chrome was bad enough. I wish this stops here.


Perspective projection is not weird or non-standard. In some sense you could consider it the least weird projection, since it's what your eyes actually see when looking at a globe. But if you just can't get enough of Mercator, the option is still there in the hamburger menu (labeled "Globe").


Google Earth desktop edition is still available. And it's still approximately ten thousand times better than the browser edition.

https://www.google.com/earth/download/gep/agree.html


It's gradually phased out. Right now it's essentially abandonware. At some point in the future they'll just kill it out of the blue. Things that provide value to advanced users don't matter to most of these "data-based" companies.


While this is a great improvement, this is most needed on mobile devices.

This was especially a problem with Ingress and Pokemon Go games where you need the map to be correct where it isn't. Even more difficult is using those apps in places like the Arctic or Antarctic, there's no way to accurately place GPS coordinates, deal with game portals or have directions other than the compass.


You can see the difference between Africa, Congo and Greenland as mentioned in the article: https://imgur.com/a/g7uIYGN

Greenland is the size of Congo, around 2,2 million km2 while entire Africa is around 30 million km2. Europe being around 10 and North America 24 million km2.


I believe this makes Google Maps the only mainstream SaaS map providing orthographic projection, and I wonder if that (partially) explains the recent increase in Google Maps pricing (previously: https://news.ycombinator.com/item?id=17570029).


That's the technology behind: s2geometry https://opensource.googleblog.com/2017/12/announcing-s2-libr...

Annoying that techcrunch could not find that out.


What's really interesting that in 2D maps, every country had their own variation of it.

If you were to take a 2D printed world map from America and compared it to a 2D printed world map, the sizes of continents wouldn't be the same, nor the centerpiece of the map itself. It would always center around that respective country first.


It never was a flat map. That was just a kludge to make it work in the browser. The original application, from Keyhole, was 3D from 2003 or so. NVidia used to offer it as a promotion, before Google bought it.

Now it's a Chrome browser promotion. Google Earth requires Google's browser.


Google Maps.


Just a note, they also changed how navigating works. It's almost as if the map curves differently at different zooms. The physics is very cool but it is a little nauseating. It is very realistic, almost like spinning a globe, but I'm not sure if its more preferable.


There is a really good short youtube video about map projections and to understand the tradeoffs that are made to create them. https://www.youtube.com/watch?v=kIID5FDi2JQ


A couple of suggestions for improvements: 1) Use quaternions instead of Euler angles for rotation 2) Show latitude and longitude grid so it's easier to identify equator and polar regions, etc...


Langitude longitude grid would definitely make it easy to map it to the flag mercaretor projection.

One can easily also correlate why things on the pole look so big when you have evenly spaced latitudes in a sphere.


This was going to launch years ago with the vector rewrite, but it was only enabled for the globe mode (Earth/GL, complete with a simulation of the objects in the sky). Interesting timing.


> as east and west in all flat mappes -- and I am one -- are one, so death touches the resurrection

-- sermons of J Donne (as quoted by oppenheimer)


As neat as this is, it's kinda a crummy experience in a few expectation-breaking ways :\ Most notably perhaps: if you put your cursor on a spot on the globe and scroll, you don't zoom there. You zoom somewhere in between.

Mostly things that can be fixed in the future (except perhaps "requires multiple click-drags to go to the opposite side", where before you could see most/all of the world at once). But it still seems a bit more shiny than useful.


I am always amazed to see how ridiculously huge the african continent is. There are really huge countries over there.


Did the compass disappear? I'd like to still be able to rotate the map.


Apple Maps used to do this too. They took it away a few versions ago for reasons I don’t quite understand. I really liked the “Pale Blue Dot”-like overview of the Earth when zoomed out all the way.


Apple Maps (desktop version) still does this, but only in Satellite mode. Which feels very intuitive, to be honest.


Of course, the definitive pale blue dot image was not taken by a satellite, but rather a spacecraft on a hyperbolic trajectory.... ;).


But time is still a flat circle.


Question is, will it destroy my battery like satellite view does?


No, surprisingly it's not much more expensive to render than Mercator.


Yeah pretty surprising. Did a test, switched over to sat view and my fans ramped right up.


I am honestly surprised how Google has not released a GPS navigator to compete with TomTom or Garmin. It would absolutely demolish them in seconds.


They already have: the smart phone in everyone's pocket.




Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: