Tag Archives: linkedin

Building Egypt 2.0: When Institutions Fail, Crowdsourcing Surges

I recently presented at Where 2.0 and had the chance to catch Adel Youssef’s excellent talk on “How Location Based Services is Used to Build Egypt 2.0.” He shared some important gems on digital activism. For example, while Facebook allowed Egyptians to “like” a protest event or say they were headed to the streets, check-in’s were a more powerful way to recruit others because they let your friends know that you were actively in the location and actually protesting. In other words, activists were not checking into a place per se, but rather creating an event and checking into that to encourage people to participate in said event.

Adel also shared some interesting insights on how location-aware mobile tech-nologies are being used to build a new Egypt. “After the revolution, the police force just disappeared, there is no police; and there is no traffic control. But this drove more crowdsourced traffic control, crowdsourced police, crowdsourced services. And this has been happening in the last year alone. Crowdsourcing revolution. But not a revolution to overthrow a tyrant but a revolution to build a developed country. [...] People going to clean the streets, planting trees, repainting the streets. And they are feeling ownership of their campaign.”

Adel shared several other crowdsourcing initiatives in his talk, from OneYad (matching volunteers) and Zabatak (monitoring corruption) to EntaFeen (check-in’s for good), Bey2Ollak and Wasalny (both addressing the problem of road traffic). I’m excited by all this innovation happening elsewhere than Silicon Valley and hope these platforms will go mainstream beyond the region in the near future. Indeed, I just signed up for the OneYad beta because I really think this kind of tool could be used in the West.

Adel: “We see a lot of crowdsourced networks built after the revolution because we need to build the country and we want to do this bottom-up, want to do it by the people, you want to empower the people.” The point, for Adel, is to go “from social networking to social working” and thus fill the gaps in services that institutions are failing to provide. This reminded me of Tunisian Ambassador Mohamed Salah Tekaya’s remarks last year: “During the Arab Spring, we have seen the power of Twitter and Facebook… Now we need to use the power of LinkedIn.”

Passing the I’m-Not-Gaddafi Test: Authenticating Identity During Crisis Mapping Operations

I’ve found myself telling this story so often in response to various questions that it really should be a blog post. The story begins with the launch of the Libya Crisis Map a few months ago at the request of the UN. After the first 10 days of deploying the live map, the UN asked us to continue for another two weeks. When I write “us” here, I mean the Standby Volunteer Task Force (SBTF), which is designed for short-term rapid crisis mapping support, not long term deploy-ments. So we needed to recruit additional volunteers to continue mapping the Libya crisis. And this is where the I’m-not-Gaddafi test comes in.

To do our live crisis mapping work, SBTF volunteers generally need password access to whatever mapping platform we happen to be using. This has typically been the Ushahidi platform. Giving out passwords to several dozen volunteers in almost as many countries requires trust. Password access means one could start sabotaging the platform, e.g., deleting reports, creating fake ones, etc. So when we began recruiting 200+ new volunteers to sustain our crisis mapping efforts in Libya, we needed a way to vet these new recruits, particularly since we were dealing with a political conflict. So we set up an I’m-not-Gaddafi test by using this Google Form:

So we placed the burden of proof on our (very patient) volunteers. Here’s a quick summary of the key items we used in our “grading” to authenticate volunteers’ identity:

Email address: Professional or academic email addresses were preferred and received a more favorable “score”.

Twitter handle: The great thing about Twitter is you can read through weeks’ worth of someone’s Twitter stream. I personally used this feature several times to determine whether any political tweets revealed a pro-Gaddafi attitude.

Facebook page: Given that posing as someone else or a fictitious person on Facebook violates their terms of service, having the link to an applicant’s Facebook page was considered a plus.

LinkedIn profile: This was a particularly useful piece of evidence given that the majority of people on LinkedIn are professionals.

Personal/Professional blog or website: This was also a great to way to authenticate an individual’s identity. We also encouraged applicants to share links to anything they had published which was available online.

For every application, we had two or more of us from the core team go through the responses. In order to sign off a new volunteer as vetted, two people had to write down “Yes” with their name. We would give priority to the most complete applications. I would say that 80% of the 200+ applications we received were able to be signed off on without requiring additional information. We did follow ups via email for the remaining 20%, the majority of whom provided us with extra info that enabled us to validate their identity. One individual even sent us a copy of his official ID. There may have been a handful who didn’t reply to our requests for additional information.

This entire vetting process appears to have worked, but it was extremely laborious and time-consuming. I personally spent hours and hours going through more than 100 applications. We definitely need to come up with a different system in the future. So I’ve been exploring some possible solutions—such as social authentication—with a number of groups and I hope to provide an update next month which will make all our lives a lot easier, not to mention give us more dedicated mapping time. There’s also the need to improve the Ushahidi platform to make it more like Wikipedia, i.e., where contributions can be tracked and logged. I think combining both approaches—identity authentication and tracking—may be the way to go.