Allow manual correction of locations in Map Editor
While making a map, I encountered a number of locations that I would've loved to include, if only they weren't so horribly misplaced as to make 5k-ing impossible. So I'd like to suggest a feature in handpicked mode for placing a secondary pin, tied to the primary photo sphere selection pin but placeable freely on the map, to set the real location on the map.
For context, misplacement is obviously a well-known problem for user-uploaded photospheres, but it can plague official coverage as well. One set of examples is some of the Dubai metro stations, where they got the spatial directions completely inverted (https://goo.gl/maps/sw84RYuZcJo7MbMWA should actually be in the southwest corner of the station, about 100m away, looking southeast because you can see the Burj Khalifa in front). I'm not optimistic about the chances of getting through to Google or uploaders to correct locations, hence the suggestion on the Geoguessr side.
One concern with this feature is that map makers might intentionally or unintentionally set incorrect locations themselves. A solution might be to consider both the original and the map maker-set location to be correct, and use the shorter of the two distances to compute the score. Perhaps there could be an upper limit on how far you can change the location (1km?) as well.
For context, misplacement is obviously a well-known problem for user-uploaded photospheres, but it can plague official coverage as well. One set of examples is some of the Dubai metro stations, where they got the spatial directions completely inverted (https://goo.gl/maps/sw84RYuZcJo7MbMWA should actually be in the southwest corner of the station, about 100m away, looking southeast because you can see the Burj Khalifa in front). I'm not optimistic about the chances of getting through to Google or uploaders to correct locations, hence the suggestion on the Geoguessr side.
One concern with this feature is that map makers might intentionally or unintentionally set incorrect locations themselves. A solution might be to consider both the original and the map maker-set location to be correct, and use the shorter of the two distances to compute the score. Perhaps there could be an upper limit on how far you can change the location (1km?) as well.
Follow this post
2
followers
Thanks for the suggestion!
Just so I understand it completely, a secondary pin would be the equivalent of a zone within the location due to some locations being off on the Map correct?
I.E being able to customize the 5k ratio more as a map maker.
Its a good idea, and I have added it to our Map-Maker feedback that we have.
KR
Not quite. What I was thinking of wasn't to change the scoring curve while still calculating distance to the location in Google's database, but rather to add a marker to set a new 0m point on the map, from which the "correct" distance would be calculated. Something like this:
I'm not sure how this would feel to players of the maps, particularly if you now have two 0m points to guard against mistakes/abuse as I mentioned. Your interpretation could very well turn out to work better (though one downside to relaxing the distance-score curve would be to increase the chances of rewarding imprecise guesses with higher-than-deserved score). I'm not going to be too dogmatic on how it could be implemented 🙂
Have added this to the ticket as well and we shall see if it is something we can add in the future. 🙂
Thanks!
This would finally alow for China maps, because it seems that everything in China is misplaced due to a discrepancy between the map and the GPS data. The only points that are correct are the ones bound to a specific Google location in China (i.e. shops etc.).
If this ever gets implemented, I'd also love something to correct the orientation (N/W/S/E) of a photosphere, because I often also found cool photospheres to include in my maps, but the orientation was wrong even though it was placed correctly.