Skip to content

May 9, 2009

A Common Crowdsourcing Platform?

by brainoids

Here’s the executive summary:   A common smartphone app interface and underlying open API could unlock crowdsourced data collection across a number of different fields.  The trick … how to get critical mass.

Amazing how a 3-minute TED talk can stimulate alot of thinking. The following snippet describes a cell-phone-based solution for citizen reporting of political oppression in Africa:

Note that Peter Gabriel was already working on lower-tech human rights crowdsourcing back in 2006:

Hersman’s comments about extending the range of the platform are most intriguing. The Ushahidi site already features additional deployments of crowdsourcing reports of swine flu and Indian election monitoring.   As inspiring as this effort is, it also begs the question of the eventual proliferation of niche deployments, and whether a common platform or clearinghouse is possible (how many “drops” of interest must a crowdsourcing participants remember and bookmark?).   (I’m reminded of my very first post here about iPhone severe weather spotting crowdsourcing.   How many citizen spotter apps would I eventually load up?)

One approach Hersman discusses is leveraging a common platform such as Twitter and somehow mining the information out of the tweetstream.   This would require significant back-end filter development (as noted) as well as some scheme to assess veracity or credibility.   As optimistic as I am about the eventual maturation of semantic technologies, this puts the payoff pretty far in the future, I think.

Another approach could be a niche for the ‘next big social networking site’, a clearinghouse for crowdsourcing reports on a multitude of topics – a “big dream” version of Ushahidi.   One site, one set of APIs, one place to report (again leveraging cell phones and smartphones), but with a variety of modular plug-in interfaces for topics of interest.   A quick-and-dirty brainstorm might yield ideas like those below, spanning the gamut from developing world to developed world priorities and concerns; from the mundane to the severe:

  1. Political oppression / violence (c.f. Ushahidi)
  2. Election fraud (c.f. Ushahidi)
  3. “Garden variety” crime
  4. Disease outbreaks (c.f. avian or swine flu)
  5. Air quality
  6. Traffic / accidents
  7. Severe weather
  8. Mundane weather
  9. Climate phenology

Now, granted, that’s a pretty grim and “engaged activist citizen” centric list.   People love to contribute, but I think it would require a particular and, sadly, niche frame of mind to attract widespread participation.   Assuming a lot of transparent geotagging, the pot could be sweetened by adding modules which “accentuate the positive” (heads-up, there’s a beautiful sunset going on right now; rainbow on the horizon! ; whatever).   Reports could easily be converted in parallel to tweets, facebook updates, whatever, but a key element is that the data collection itself could be centralized at the “hip, go-to place”.

Why would centralization be a good idea?   I think, because of the critical issue of credibility and validation.   eBay showed us how critical participant credibility is to highly-democratized commerce, and really, the concept above is just a highly democratized marketplace of information.    With a central crowdsourcing clearinghouse platform, participants could incrementally earn credibility with each “valid” report.   eBay establishes credibility by, effectively, peer review.   Crowdspotters could earn credibility by a couple of mechanisms (I’m envisioning some point scheme):

  1. Peer validation (low points earned incrementally when the same event is reported by multiple spotters in the same time / location)
  2. Independent corroboration (some back-end algorithms validate user reports against “official” databases after the fact, and then assign out higher point awards)

Credibility scores then become a badge of pride / status – an incentive for participating in improving the global, connected citizen database.   There would have to be some weighting (corroborated political oppression reports get way higher points than “it’s warm and sunny in San Diego”), as well as serious consideration to how not to inadvertently disenfranchise the developing world (the villager in Africa who can’t afford unlimited text messages or an iPhone), but those are really all implementation and calibration issues.   The key “thing” is leveraging common and mundane ways for people to contribute to establish credibility and keep people engaged for when the “rare big important” reporting opportunities come up.

Technically, there’s no really Twitter itself couldn’t spin this up as riding in parallel with its current activity.  If you want to yammer into the global tweetstream-of-consciousness in an unstructured way, fine, but if you want to contribute something and have it much more likely to be picked up and used for good, well, flip over to the “crowdspotter” Twitter applet tab, or text using some #standardformat and Twitter peels off the reports to the appropriate database.   I’d even have no problem with Twitter leveraging the collected data and selling it into aftermarkets (c.f. my iPhone severe weather spotting post); heck, this might even give it a sustainable, Google-esque business model.   As an aside, if some topics were so critical (political oppression) that niche deployments and “editors” like Ushahidi were still warranted, there’s no reason such deployments couldn’t stand to leverage and benefit from a more structured and distributed common platform of “lower quality input” reports described above

The global proliferation of cell phones – and, eventually, smartphones – opens up far too many possibilities for crowdspotting and democratization and empowerment to let it languish.   Niche deployments are a good thing, but will face an uphill battle if participants have to find their way to the niches; the critical factor for crowdsourcing and crowdspotting is the huge volume of potential participants.   This seems like a huge disruptive technology opportunity waiting for someone to pick up and run with …


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Note: HTML is allowed. Your email address will never be published.

Subscribe to comments

%d bloggers like this: