Crowdsourcing, the idea of soliciting information from citizens and relying on their participation to achieve certain goals is almost 10 years old. We ourselves, have been building products that allow users to collect, curate and visualize information since 2008, relying solely on the power of the masses. In the last few years, internet enabled sensing devices have permeated every day life, through mobile devices and through very cheap and accessible electronic sensors.
We've wondered for some time now how we might combine the perceptive abilities of the crowd with the long term sensing capabilities of machines to help us deal with issues of verification and continuous monitoring. Machine sensing is precise and incontrovertible for the most part, which is why we feel it plays a role in the crowdsourcing conversation.
I'll try to and illustrate potential uses of human-machine sensing using two Ushahidi deployments. It's probably worth noting too at this point that machine sensing is inconsequential in situations where environmental variables play little or no role. The examples below deal for the most part, with air and water pollution and its effect on communities. Hopefully we can see sensors start to play a role in other areas.
Deepwater Horizon oil spill (2010) - This was considered to be the largest oil spill in the history of the petroleum industry. The
Louisiana Bucket Brigade crowdsourced the after effects of the spill on Louisiana's gulf coast communities. Many of the spill and leak sightings, along with odor and noise reports could have been well corroborated in a single platform with continuous air and water sensory data. Some of it was, with some data from Louisiana's Department of Environmental Quality. Imagine however, if this data was collected in real time within the Ushahidi platform and further, imagine if they didn't have to rely on a government agency entirely for this data.
Japan Earthquake and Tsunami (2011) - Launched by the Japanese OpenStreetMap community, this Ushahidi deployment crowdsourced disaster relief efforts, status updates on transportation, shelter and more. Additionally they used external data to map the radiation fallout. Radiation fallout is a critical piece of sensing that needs to be precise, monitored well into the future and necessary to keep this deployment up to date -- some of which can be done with cheap sensors.
These are just two use-cases that come to mind. Imagine combining citizen generated media with:
- Noise sensors in urban environments
- Rubbish level sensors
- Air pollution sensors
- Water quality sensors
- Water level sensors
- Human presence sensors
- Power outage sensors
In version 3 of the Ushahidi platform we've been working on allowing you to send data from the sensor web into the Ushahidi platform in real time and present it in a useful way alongside the crowdsourced data that you're familiar with. Look out for more updates on this in coming weeks.
Photo courtesy of AdaFruit Industries.
Crowdsourcing, the idea of soliciting information from citizens and relying on their participation to achieve certain goals is almost 10 years old. We ourselves, have been building products that allow users to collect, curate and visualize information since 2008, relying solely on the power of the masses. In the last few years, internet enabled sensing devices have permeated every day life, through mobile devices and through very cheap and accessible electronic sensors.
We've wondered for some time now how we might combine the perceptive abilities of the crowd with the long term sensing capabilities of machines to help us deal with issues of verification and continuous monitoring. Machine sensing is precise and incontrovertible for the most part, which is why we feel it plays a role in the crowdsourcing conversation.
I'll try to and illustrate potential uses of human-machine sensing using two Ushahidi deployments. It's probably worth noting too at this point that machine sensing is inconsequential in situations where environmental variables play little or no role. The examples below deal for the most part, with air and water pollution and its effect on communities. Hopefully we can see sensors start to play a role in other areas.
Deepwater Horizon oil spill (2010) - This was considered to be the largest oil spill in the history of the petroleum industry. The Louisiana Bucket Brigade crowdsourced the after effects of the spill on Louisiana's gulf coast communities. Many of the spill and leak sightings, along with odor and noise reports could have been well corroborated in a single platform with continuous air and water sensory data. Some of it was, with some data from Louisiana's Department of Environmental Quality. Imagine however, if this data was collected in real time within the Ushahidi platform and further, imagine if they didn't have to rely on a government agency entirely for this data.
Japan Earthquake and Tsunami (2011) - Launched by the Japanese OpenStreetMap community, this Ushahidi deployment crowdsourced disaster relief efforts, status updates on transportation, shelter and more. Additionally they used external data to map the radiation fallout. Radiation fallout is a critical piece of sensing that needs to be precise, monitored well into the future and necessary to keep this deployment up to date -- some of which can be done with cheap sensors.
These are just two use-cases that come to mind. Imagine combining citizen generated media with: