search

stamen / terrain-classic

World-wide CartoCSS port of Stamen's classic terrain style
ISC License
144 stars 35 forks source link

Language labeling inconsistencies in China (and perhaps beyond) #75

Closed geoglrb closed 8 years ago

geoglrb commented 8 years ago

Hi, perhaps an OSM thing as much as anything, but in case not:

Here in Chengdu, Chinese labels are largely absent at zoom levels <= 11, then from there on in, increasing zoom, it's mainly Chinese-characters-only up to ~17, where romanized labels start reappearing.

http://maps.stamen.com/terrain/beta/#11/30.6586/104.0567

geoglrb commented 8 years ago

Here may be some more useful clues:

Neither the city of Chengdu (lower right in first image below) nor the city of Dujiangyan (upper left in first image below) are labeled in either of these two zoom levels below. In the second, you do get something in the proximity of Dujiangyan, but labeled by the name of Guankou. There are a few oblique references to this toponym buried in the records I'm looking at, but this is not the name of the city nor a good label. See: https://nominatim.openstreetmap.org/details.php?place_id=630162 https://en.wikipedia.org/wiki/Dujiangyan_City

screen shot 2016-07-11 at 2 25 28 pm screen shot 2016-07-11 at 2 30 02 pm
geoglrb commented 8 years ago

Also a lot of less-than-ideal landform shading... at certain zoom levels, gets pretty discrete in the pixelated steps that become visible...

geoglrb commented 8 years ago

All that said, the relations of labels to administrative hierarchies in Seattle area is also non-obvious to me...so maybe it's just me!

screen shot 2016-07-11 at 2 36 17 pm
clhenrick commented 8 years ago

@geoglrb thanks for pointing out these various issues. In the future when creating Github issues please only open one issue per specific subject.

Closing to break this out into multiple issues.