Thoughts on Google's "Terrain" maps | Hacker News Reader

Thoughts on Google's "Terrain" maps

41latitude.com

45 pointsj05300318 comments

Richland Chambers Reservoir shows up as "Corsicana River" on terrain data. I asked around and was told that 20+ years ago that was the name of the river before they broke a dam and extended the area. Perhaps the data being used to map terrain areas is quite old?

I know that the Montana topographic quads are based off of surveys done in the 60s and 70s. A lot of them have been updated, but the names don't always get changed.

Google hasn't spent the time to go through and prioritize the labels. I imagine they could do this fairly automatically with their web index, but it's still a fairly large chunk of work for a relatively small benefit. I'm sure most people hit the maps from a search, not just to drag the map around and critique the labeled markers.

I'm sure most people hit the maps from a search, not just to drag the map around...

I know people who spend hours dragging the map around in Google Earth. And while I'm sure that a lot of people are going to the terrain maps via a search, there's most certainly another large chunk of people on there just dragging the maps around (I do it myself in a number of mashups/web apps that I visit regularly).

Given the post's examples, I think the critique is more than warranted, and I also think you're underestimating the benefit of prioritizing the labels.

Consider this: if paper maps disappear (along with the newspapers), shouldn't there be something marking the seas, mountains, and deserts?

I don't like the idea that paper maps will disappear though. But I also don't like the idea that 'always on' Internet everywhere means that everything should reside in the 'cloud' (for as much as the term even means any more). If paper maps are going to disappear, I would want there to be some sort of an off-line replacement b/c maps are probably one thing that is used more often in areas outside of cell service.

That's one reason (among many) I like OpenStreetMaps, since it has an API that lets you download subsets of data for offline usage, along with bulk dumps of the whole thing. Allows for all sorts of offline uses, not to mention data mining, that are impossible with Google Maps.

OpenStreetMaps is great for offline usage, I just wish the online page was faster.

I constantly use Maps to explore areas where I haven’t been before which involves a lot of dragging around, not searching. Maps is tremendously helpful when your goal is to get a feel for the area and more consistent labels could certainly help making Maps much better.

I’m puzzled by Google’s labeling decisions on their "Terrain" maps.

Replace "Terrain" with anything else and that's pretty much the subject of all of this blog's entries. I don't know why the naming on the terrain maps would be any better than the labeling on any of the other map modes.

If Google can distinguish references to different features with the same name on web pages, then Google should rank features by number of references on the web and label accordingly.

You can report mistakes (http://maps.google.com/support/bin/answer.py?hl=en&answe...). I reported one once... about 6 months later received an email from google saying it was fixed... it was. Of course a wikipedia-API would be awesome. Google usually likes things like that: http://images.google.com/imagelabeler/

It has actually improved, I think. They used to label weird little water features even at fairly zoomed out levels. You'd see a map of some state and see "Level Ditch #4" in the middle of Oregon or Idaho or something.

Incidentally, I think Google Maps is one of the coolest things ever... I used to love to look at USGS quad maps as a kid, and now I get something that is 80% as good for pretty much the entire world. How cool is that...

I don't think the terrain tiles get much attention. For a while it was really excited about irrigation ditches, then a few months ago lots of river rapids showed up. Hopefully they'll make the relief shading layer available in styled maps at some point.

Could it be that these terrain maps are compiled from hiking maps, which tend to be very local and don't label obvious or general features like "the Nile" or "the Alps"?

I can't think of a single hiking map I've ever seen that didn't label rivers, no matter how "obvious" they were.

To me, "terrain map" means "physical map". If the "terrain" maps are intended for hikers, Google should instead label them as "hiking" maps. But I doubt that they are intended purely as "hiking" maps... none of the trails near my house are on them.

"Terrain" to me means "topographic". That's not the same as hiking maps, it mostly means that it has elevation contours.

Compiled from hiking maps? In places like Egypt, his first example? My guess is that there aren't a great deal of hiking maps for the Sahara desert...

They're compiled, I think, from whatever they can get their hands on. It's interesting to see borders, for instance, where there is clearly a difference in the data available. Switzerland, for instance, seems to have really good data:

http://maps.google.com/maps?f=q&source=s_q&hl=en&...

Austria... not quite to the same level of detail.

I'm pretty sure it's Shuttle Radar data from NASA.

http://www2.jpl.nasa.gov/srtm/

Reply on news.ycombinator.com