Today, NiJeL is proud to announce the launch of Unite Arizona (ImmigrantHarassment.com), a web resource provide both a way for Arizonans to anonymously report harassment, intimidation, raids/sweeps, and an outlet for unreported criminal activity via SMS (text message), Twitter, email, or the web. Unite Arizona is currently live and accepting SMS data at 602-824-TALK (8255), Twitter updates with the hashtag #MHRSAZ, and emails at report@immigrantharassment.com.

With the passage of Arizona Senate Bill 1070, global media, politicians, human rights groups, and concerned individuals have turned their attention toward Arizona, rightfully concerned about the negative impacts of this new law. Minority groups in Arizona have been and will continue to be subject to verbal and physical harassment and intimidation from organized hate groups, some members of law enforcement and xenophobic Arizonans. Moreover, many more victims will likely cease to report crimes out of fear of detention and deportation due to this law.

NiJeL created Unite Arizona (ImmigrantHarassment.com) to provide an outlet for harassment, intimidation and unreported crime. These incidents will be filtered by the type of incident and visualized on a participatory map and a timeline for the community to see. Unite Arizona uses the Ushahidi Platform: free and open source software designed to gather real-time, crowdsourced data for crisis response.

Incoming data can be tagged by location, category, date and time, and each report can include references to news items, photos and video. Trusted site administrators are charged with mapping and coding incoming messages, approving and verifying each incident, scoring the reliability of the source and indicating the probability that the event is real. Users of the site can also rate the importance of incidents, promoting those that are particularly egregious. Finally, anyone can sign up to receive alerts of approved incidents, filtered by location. With this system we intend to provide a powerful reporting platform for victims and activists, an alert system for crisis responders, and a compelling visualization of the scale and scope of harassment, intimidation and unreported crime in Arizona.

There are a number of ways to help us with this project:

Moderation Volunteering
If you would like to help us moderate reports of harassment, intimidation and unreported crime and comments form the public, please contact us using the from under the contact us tab. You will need to go though a background check process and attend a training session to learn how to use the internal moderation tools. Thank you!

Organizational Support
If your organization would like to show support for this effort and would like more information about how to get involved, you may also use the contact us form under the contact us tab. Thank you!

Donations
We are also accepting donations to help us support our volunteer coordinators, train new moderators, disseminate SMS and other site information, and improve the site technology among other items. Any amount would be much appreciated. Please see the About Us page on Unite Arizona to dontate using PayPal. Thank you!!

I’m not sure why I didn’t see this video until now, but both Nancy and Lela talk extensively about their thoughts on ICCM 2009 on this video of conference highlights. Nicely done!

Hot off the presses! We blogged a while back about JD’s Ignite talk – Empirically Derived ‘Fault Line’ Analysis: A Proposed Early Crisis Warning Tool at ICCM 2009, and now we have an extremely well produced video to share of the talk. Because of the speed of the talk, following JD through the model might be a challenge, but if you’re interested in talking a more in depth look at the model you can read more about it here.

Thanks again to Patrick Meier, Jen Ziemke and the team of JCU students who filmed and produced each of these videos. Thank you!!

NiJeL was honored to be invited to the Innovations in Mobile Data Collection for Social Action in the Middle East workshop next week in Amman, Jordan, hosted by UNICEF Innovation and MobileActive.org. Our executive director, JD, will have the pleasure of attending this workshop and is looking forward to showcasing some of NiJeL’s most recent work and learning from the impressive roster of attendees! You can follow JD’s updates from the workshop on Twitter here: @nijel_mapping

Here’s a quick description of the workshop:

UNICEF Innovation and MobileActive.org invite you to attend a three-day workshop on distributed and real-time data collection, monitoring, and visualization of data with mobile technology.

What is this About?

With the ubiquity of mobile technology, data collection and monitoring of key indicators from the ground up by affected populations is now possible. Mobile technology in the hands of people can now be more than a person-to-person communication medium but can be used for capturing, classifying and transmitting image, audio, location and other data, interactively or autonomously.

By involving people in defining and participating in their own data collection, this approach can address significant unmet challenges in large-scale data collection for public health and citizen participation.

In this three-day workshop, we will explore the critical issues, technologies, and architectures involved in collecting and utilizing data-from-below, bringing together the key technology and research leaders on distributed data collection and distribution in the Middle East.

What are the Goals?

• An exploration of key issues in citizen-driven data collection in the Middle East.  These include technologies, systems, architecture, tools, standards, and people, among others.
• Kick-start a regional working group / community around open-source data collection, aggregation and visualization using mobile technology
• Map the landscape in the Middle East of applications/technologies, developers, and key thought leaders around real-time distributed data collection, monitoring, and visualization using mobile technology?
• Help UNICEF build a roster of potential partners, possible vendors, academic institutions of interest, and groups or individuals to advance UNICEF regional goals.
• Prototype new products or improvements of existing products about distributed data collection.
• Please note that the workshop will be held primarily in English (with some Arabic sessions).

The impetus for the workshop is UNICEF’s national-scale project in Iraq collecting data from various populations about key indicators and use that data to effect policy and programmatic changes that can improve the lives of children.

As part of this work, MobileActive.org, a global community of people using mobile technology for social impact, and UNICEF partnered to explore, with key leaders in the Middle East, critical issues on large-scale, citizen-driven and bottom-up data collection.

And here’s a list of the impressive organizations invited to be there:

Al Jazeera
ArabiaGIS
Cisco
Development Seed
Gapminder
Institute for War & Peace Reporting
Intel
Meedan
Mercy Corps
MIT Media Lab
Open Data Kit
Open Mobile Consortium
Rootspace
Royal Scientific Society of Jordan
Sharek961
SMEX Beirut
Souktel
Thoughtworks
UNICEF Iraq
Union of Arab ICT Associations
Ushahidi
Women in IT, Lebanon
Zain

Many thanks to Chris C. at MIT Media Lab for facilitating our involvement at this workshop, and to Katrin Vercalas at MobileActive for inviting us to attend. Thank you!!

Nancy, Lela, Layal and JD all attended the first annual International Conference on Crisis Mapping (ICCM) a few week ago at John Carrol University in Cleveland, OH. Co-founded and organized by Jen Ziemke and Patrick Meier, this was a top notch conference loaded with some of the best minds in crisis early warning, humanitarian operations and logistics, software development and user interface design, and we at NiJeL were proud to be a part of this high powered dialog. In this post, I’ll be discussing my 5 minute Ignite Talk on a model we’ve developed that might be useful in conflict early waring applications, but look for future posts from Nancy, Lela, Layal. Lela will be discussing her participation on a the Crisis Mapping Visualization and Crisis Mapping Analytics panel, and Layal will be discussing her overall impression of the conference and what she took away from it.

My Ignite Talk focused on a crisis early warning predictive model that we derived after viewing data from the UNDP’s Crisis Recovery and Mapping Analysis (CRMA) team (for more information about the CRMA, take a look at Margunn Alshaikh’s ICCM Ignite Talk about CRMA’s work – fascinating!). The CRMA team led a massive participatory mapping project over the last 2 years across Sudan to better understand spatially the threats and risks to peace, actors, natural resources and other indicators of peace or conflict. Over the summer, Patrick asked us along with Andrew Turner at Geocommons to attempt to derive a model using this data that would predict where conflict might occur – the “fault lines” along which you might increase your intervention if you are confident in the model’s predictive capacity. The model we derived we shared publicly for the first time at ICCM 2009. Our slides follow:

As you can see from the slides, our model is entirely theoretical at this point – we have yet to have the opportunity to test it on either historical data or on currently acquired field data such as in the case of the CRMA project in Sudan.

The model that we derived (below) is a relatively straightforward “gravity based” model where the main operative principal is that the further you are away form a potential conflict flash point, the less of an impact that threat has on you in your current location.

In the model, each term represents a rasterized spatial data layer. Vj in each term is the value of the layer at cell j – for instance if conflict over water resources is an issue, then Vj might be a measure of  water availability at cell j. Dij is the distance from the cell i (the cell being evaluated) to cell j and -alpha, -beta…-gamma, are all distance friction coefficients, meaning that the further cell i and cell j are from each other, the less of an impact that data point will have over the calculation of Ci. These distance friction coefficients are exponential decay factors and will be empirically derived for each layer from any spatial conflict data that we can get our hands on. The coefficients A, B…X are layer weighting coefficients and should be assigned to each layer through expert local opinion. For the CRMA data, we have asked the CRMA team to rank order the data layers they think would be of importance in this model and tell us qualitatively, how “close” one layer is to another in terms of rank. Is there a wide gap between the top ranked layer and the second ranked layer?

Once we have each layer weighted correctly, we can calculate Ci – the strength of the fault line at cell i – for each cell in the grid. The resultant grid we calculate is a so called “violence risk surface.” As I said during my talk, this surface should show areas of relatively high risk for potential conflict, but clearly will not be a substitute for a trained analyst to predict where conflict might occur. I likened this resultant violence risk surface to software a radiologist uses to identify areas in an x-ray or CT scan that might be of interest for further consideration – this model might suggest areas of potential conflict that an analyst might not otherwise have noted.

Several people talked with me about this model after my talk and I’ve already connected with a few folks about potentially using historical and current data to derive the distance friction coefficients and to improve the model in other ways. Jeffrey Villaveces at UN OCHA Columbia has graciously offered to send us data on conflict between the government and the FARC, and others have shown interest as well.  For those of you who attended ICCM 2009 and would like to connect with me again on this topic, feel free to send me a message at jd ‘at’ nijel.org. Thanks!

We have created a Google Earth file highlighting Conflicts in Africa from 2007-2008. This file was produced with data from the state department’s Humanitarian Information Unit and in collaboration with Patrick Meier, a Doctoral Research Fellow at the Harvard Humanitarian Initiative. The Google Earth file shows regions where Armed Conflict, Inter-Communal Strife, and Political Violence have taken place. Patrick has previously blogged about this work on his blog iRevolution.

The original HIU data were provided to us a GIS shapefile, which we converted into KML for viewing in Google Earth.

You can download the KMZ Google Earth file here:

http://nijel.org/conflict_map/Africa_Conflicts_Without_Borders.kmz

And the original HIU pdf map of “Africa: Conflicts Without Borders 2007-2008″ here:

http://nijel.org/conflict_map/hiu_cwob_v081110_300dpi4.pdf

An article this morning in Directions Mag talkes extensively about “an interactive Map Center” built for a UN Humanitarian Information Center website in response to Cyclone Nargis. While I think it’s great the company, IDV Solutions, that built the map could deploy it in a week and there are some very sophisticated features bundled in the application, a few big red flags came up right away for me. First, their solution is tied exclusively to Microsoft SharePoint – integrating it with any other system does not seem to be possible. SharePoint and their solution, Visual Fusion, are both proprietary – we have no idea how these applications work under the hood and if someone else, say at the UN, wanted to replicate this system for another crisis, even if they had the expretise to do so, they could not deploy it themselves. That’s one reason why Ushahidi stands out. It’s open platform will allow any organization with the expertise to deploy their application, whereas with this solution, the UN has to keep coming back to the well. Finally, it appears to be entirely Flash-based on the front end, which might be painfully slow in low bandwidth conditions, or might not work at all on older machines with older browsers. This might look fantastic in the UN headquartes in New York, but in the field I can forsee hardware and bandwidth issues being too much for the system to overcome.

Our Crisismapper friends over at Ushahidi recently lent their alpha code to Al Jazeera to deploy War on Gaza, which is allowing civilians subjected to the fighting to post incidents via SMS, Twitter, and web form. Erik Hersman of Ushahidi has an excellent blog post here detailing how the deployment came about. Here at NiJeL, we’re very interested in how this deployment ultimately handles Arabic language support – our work on HarassMap is just beginning and we’re hoping to support both Egyptian Arabic in Arabic script and Egyptian Arabic in semi-standardized Latin script, so there could be many lessons to learn from Ushahidi and Al Jazerra’s work on this vital project.

I did a quick search of the ground level feeds and information for the attacks. I’ll add more as I find them.

The map below of the attacks is linked on a number of posts and is from http://tr.im/1jj1

View Larger Map

Twitter Feeds:

• http://twitter.com/mumbaiattack
• http://twitter.com/Gauravonomics
• http://twitter.com/mumbai

Blogs:

• http://indiauncut.com/iublog/article/a-night-out-in-mumbai/

Mumbai Help Blog: http://mumbaihelp.blogspot.com/2008/11/can-we-help.html

Photos

• http://www.flickr.com/photos/vinu/show/with/3061372661/

News:

• http://www.mahalo.com/Mumbai_Terrorist_Attacks

Lela