National Library
Aerial photo
National Library
Aerial photo
National Library
Aerial photo
National Library
Aerial photo
National Library
Aerial photo
Role
Role
Product Designer
Product Designer
Product Designer
Product Designer
Timeline
Timeline
4 weeks (MVP)
4 weeks (MVP)
4 weeks (MVP)
4 weeks (MVP)
Platform
Platform
Web, Mobile
Web, Mobile
Web, Mobile
Web, Mobile
Constraints
Constraints
Strict MVP scope for the pilot
Strict MVP scope for the pilot
Strict MVP scope for the pilot
Strict MVP scope for the pilot
The product
The product
A map-based digital platform for the National Library of Norway that turns historical aerial photography into an interactive, crowdsourced archive.
A map-based digital platform for the National Library of Norway that turns historical aerial photography into an interactive, crowdsourced archive.
A map-based digital platform for the National Library of Norway that turns historical aerial photography into an interactive, crowdsourced archive.
A map-based digital platform for the National Library of Norway that turns historical aerial photography into an interactive, crowdsourced archive.
Business goal
Business goal
The library had millions of historical photos with bad or missing metadata. Machine learning could only guess rough locations, and hiring experts to sort them manually was too expensive. They needed a way to invite the public in to help them organize this massive piece of history.
The library had millions of historical photos with bad or missing metadata. Machine learning could only guess rough locations, and hiring experts to sort them manually was too expensive. They needed a way to invite the public in to help them organize this massive piece of history.
The library had millions of historical photos with bad or missing metadata. Machine learning could only guess rough locations, and hiring experts to sort them manually was too expensive. They needed a way to invite the public in to help them organize this massive piece of history.
The library had millions of historical photos with bad or missing metadata. Machine learning could only guess rough locations, and hiring experts to sort them manually was too expensive. They needed a way to invite the public in to help them organize this massive piece of history.
The approach
The approach
Instead of building a boring data-entry tool, I designed it around local pride and curiosity. Users get to test their local history knowledge by finding and pinning old photos of their hometowns.
Instead of building a boring data-entry tool, I designed it around local pride and curiosity. Users get to test their local history knowledge by finding and pinning old photos of their hometowns.
Instead of building a boring data-entry tool, I designed it around local pride and curiosity. Users get to test their local history knowledge by finding and pinning old photos of their hometowns.
Instead of building a boring data-entry tool, I designed it around local pride and curiosity. Users get to test their local history knowledge by finding and pinning old photos of their hometowns.
In return, the library gets its massive archive accurately tagged. I used lightweight progression and ranking mechanics so people would actually enjoy the process and keep coming back to do more.
In return, the library gets its massive archive accurately tagged. I used lightweight progression and ranking mechanics so people would actually enjoy the process and keep coming back to do more.
In return, the library gets its massive archive accurately tagged. I used lightweight progression and ranking mechanics so people would actually enjoy the process and keep coming back to do more.
In return, the library gets its massive archive accurately tagged. I used lightweight progression and ranking mechanics so people would actually enjoy the process and keep coming back to do more.
Setting clear expectations before users touch the map
Setting clear expectations before users touch the map
Filtering unplaced photos dynamically by the current map area
Filtering unplaced photos dynamically by the current map area
Adapting a heavy map and image feed for small screens
Adapting a heavy map and image feed for small screens
Splitting the task into two clear steps to prevent accidental drops
Splitting the task into two clear steps to prevent accidental drops
Forcing users to isolate and tag a specific detail in the photo
Forcing users to isolate and tag a specific detail in the photo
Login and fast predictive search to jump regions
Login and fast predictive search to jump regions
Using rankings to make repetitive data entry addictive
Using rankings to make repetitive data entry addictive
9
mil
Photos in the National Archive
9
mil
Photos in the National Archive
9
mil
Photos in the National Archive
9
mil
Photos in the National Archive
9
mil
Photos in the National Archive
64.76
64.76
64.76
%
%
%
Of pilot images geolocated in just 6 months
Of pilot images geolocated in just 6 months
Of pilot images geolocated in just 6 months
64.76
%
Of pilot images geolocated in just 6 months
64.76
%
Of pilot images geolocated in just 6 months
Tıme to hit
pause,
see you soon
Tıme to hit
pause,
see you soon
Tıme to hit
pause,
see you soon
Tıme to hit
pause,
see you soon
Tıme to hit
pause,
see you soon