Home.Land.Security Official Project Launch!!

Project:


In the wake of the murder of Trayvon Martin, there is heightened awareness of violent actions against and harassment of folks from marginalized communities by the police or otherwise. While there are existent sites that address traditional notions of safety, particularly in the form of mapping crimes, there are little or no resources available for people from these communities–people of color, queer folks, and undocumented people– to gauge how safe they are in their communities or in the communities that they are visiting. Our site is designed for vulnerable citizens and visitors of San Francisco, who wish to better identify safety risks from their viewpoint, including, but not limited to, officer involved shootings and targeted arrests for what we call “quality of life” crimes.

Home.Land.Security is an interactive learning and planning tool for the vulnerable citizens and tourists of San Francisco, who could potentially be the target of attacks similar to the killings of Oscar Grant and Trayvon Martin. We aim to show how neighborhoods and demographics are disproportionately targeted for intimidation, harassment, and excessive use of force by mapping officer involved shootings, quality of life crimes (which represent crimes in which vulnerable groups are typically targeted with excessive force), and user reported incidents that can expand past traditional notions of safety and crime.

This site highlights spatial patterns of excessive force and aid both citizens and policymakers in preventing future incidents. This site will also give people a forum to map their negative experiences with law enforcement officials or other citizens.

Check out Home.Land.Security.

Diagrams:

Team Metro:
Jordan “Google Earthling” Rosencranz
Jordan was the utility man. In addition to putting nearly 60 officer involved shootings on the map, he was instrumental in adding key functions such as activating the user reported crime form tool and switching toggle tools from checkboxes to radio buttons. He also added a search bar.
Ben “Likes Shiny Things” Palmquist
Ben made our site shine with CSS and bootstrap and took the lead on many of our debugging projects. He also handled the finicky crimespotting data, while taking the lead on coding our custom function, which ranks neighborhoods by non-traditional safety measures.
Pamela “Devil in the Details” Stephens
Pam took the reins on the GIS projects, publishing a myriad of demographic layers with clear/user friendly legends. She was the lead formatter of the initial accordion-style site design and paved the way for final design. Pam also helped format our charts.
Will “Wordsmith” Dominie
Will published all of our written content on the site, allowing the user to better understand the purpose layer and function. Will was also a GIS superstar and put in a great deal of work behind the scenes to ensure that the neighborhood ranking function worked properly. Extremely detail oriented, Will worked with divs and charts to create an intuitive user interface.

Evaluation:
Things that worked great!
o Our styling was super stylin.
o We presented our data really well. It looks good and is easy to read/understand.
• Things that worked less great!
o Forms?
o We have this strange bug in the site where our demographic layers don’t always align with the map itself. It only happens some of the time, usually after loading the map, clicking on a neighborhood to display graphs, and then changing the demographic layer. It is a disturbingly fickle bug, only occurring in some browsers and only at certain window sizes. Our only hypothesis is that it may be due to an overzealous div (a number are rigged to pop up on click ) that changes the window size. Since the map is set to 100%, perhaps the map resizes to meet the new size (we do observe a faint shift on click), but forgets to resize the map server layer? While we are usually very willing to take responsibility for our mistakes, we kinda suspect that their error might be more on google or ESRI’s end.
o It was difficult to connect the searchbar to our user reporting forms.
o Crimespotting was really annoying and difficult to work with. Sometimes they would turn their API off and then back on again randomly.

• The future of Home.Land.Security
Following our successful launch on Yohman, we will be shopping our site around to organizations that might be interested, such as copwatch.

Added functionalities:
a. Expand the project to outside of San Francisco, either by creating national website, or similar sites for other cities and counties.
b. Add a resources page for those reporting harassment that want to take action
c. Integrate more positive indicators of security for our targeted population such as availability of cultural resource centers
d. Add introductory window/awesome video to help capture the narrative of the site akin to Will’s framing in the presentation)

Documentation:
o User interaction: User interaction is a key feature of our site’s design. We attempted to create an interface that encourages exploration, allowing users to explore the data presented in a nonlinear manner. In particular, we wanted users to be able to make comparisons between different neighborhoods and datasets, exploring how race, immigration, location and other factors influence policing and safety. Toward this end, we included a variety of tools for people to interact with the site. These include:
• A search bar: This allows users to easily locate a place or address.
• Forms: We have included a form that allows users to report on incidents that threaten their safety, or the safety of others. These incidents are stored in a database, and can be shown on the map.
• Radio Buttons: We included a number of layers that users can toggle using checkboxes or radio buttons.
Custom function: We authored a function (metro.showRank) that displays the relative ranking of a neighborhood and color codes the display to reflect how a safe a neighborhood is. We think its pretty neat.

o Custom layers: We created oodles of custom layers, including some with some with some really slick legends that Pam created in illustrator. These layers include:
•An Officer Involved Shooting (OIS) KML, which we later converted to an ArcMap mapserver.
•A neighborhood master layer, in which we converted our data layers (census demographics, OIS and crime statistics) to SF neighborhood boundaries using dissolve and some spatial joins. This layer was the base for our demographic layer. It also defines neighborhoods that users can click on to get more data on the safety of a neighborhood, which is shown at the bottom of the map.
•A number of demographic and neighborhood layers to allow users to explore the characteristics of their neighborhoods.

oWe also include Purpose, About Us, and FAQ tabs.

Week 8: Website Assignment

The Proposed Mid City Bus Stops website just got a little bit more interesting. Now users can not only toggle amenity layers on and off, but they can also assess the areas they’re looking at based on selected demographics (from 2000 though, yikes…not enough time to produce nice updated shapefiles!!). In addition, users that have a better sense of where they want to look at can search addresses and zip codes. Fun times for everyone involved.

PS: The messy bus stop stuff from last time is gone. Perhaps I can work it out in the future.

Itching for more? Click HERE.

Week 7 Assignment

This week I played with adding ArcGis layers into the proposed bus stop maps. After agonizing over the prospective layers I could add, I decided to include rail lines (excluding Expo line because I already had this shapefile from earlier this year) and libraries. Unfortunately neither rail nor libraries are very close to my area, so it’s not a very exciting map. Also, it seems that my existing bus stops are not being called.

Check it out for real here!

Home.Land.Security Status Report

This week our team’s been working on pulling in four data sources for our map of health and safety in San Francisco.

(1) Officer-involved shootings: We found a list of all officer-involved shootings in the city of San Francisco from 2000 through 2011 and have geocoded this information and created twelve layers (one for each year) that can be toggled on and off.

(2) Grocery stores we have a Google Places layer showing grocery stores within the city and (3) Crimespotting: Drawing data from San Francisco Crimespotting, we are in the process of mapping crime data from throughout San Francisco. As of May 1, we have most of the data mapped, but are still working to create check boxes to toggle layers on and off, to group different types of crimes within three main crime categories (violent, property, and quality of life), and to find a way for two-word crime types in the API (such as “SIMPLE ASSAULT”) to translate into one-word arguments for the get JSON call (like “simple_assault”).

(4) Twitter: The Twitter API is proving to be a little more difficult than we had thought. There is an overall issue with getting any geocoded tweets on the map no less any geocoded tweets by our search terms (so far we’ve tried immigrant but once the code is in place we would like to use more pointed search terms. Because the query returns are over the limit every time, the map only loads about half of the time which is also an issue.

Once we finish our Crimespotting and Twitter maps, we will combine these four maps into one elegant map and work on formatting and presentation for the midterm.

Week 4: Website Assignment

The latest version of the Proposed Mid City Bus Stops website now features toggling! I decided not to include the bound part of the code because everything was outshining the proposed stops which is kind of the point of the whole thing. Admittedly the color scheme got a little bit nauseating this time, but I had to dedicate most of my efforts to making the site work this week.

Home.Land.Security Updated Proposal

A. Revised proposal topic and description

How safe is LA? New IT firm to re-present the city’s safety in a new light

We are pleased to announce Home.Land.Security., a new IT firm specializing in mapping the safety of Los Angeles’ communities for the city’s most vulnerable people.

In February of this year, Trayvon Martin, a 16-year-old African American was shot and killed by George Zimmerman as he walked to his girlfriend’s house in Sanford, Florida. The killing, which appears to have been racially motivated, has triggered anger and sadness nationwide. It has also sparked controversy and attention to the safety of Black and Latino teenagers—especially after talk show host Geraldo Rivera suggested that Trayvon’s death was due to his wardrobe choice (a hooded sweatshirt) rather than his racist attacker. This project is a response to incidents like Trayvon’s death, and seeks to bring to light the larger systemic forces that make neighborhoods unsafe and unhealthy.

In recent years, there has been a growing recognition of the interrelationships between health, safety, and the built environment and the dire need to create safer spaces. Public health professionals have begun to think more holistically about health, questioning the social and spatial determinants of wellbeing. Planners too are thinking beyond their traditional domain, questioning how streets, housing, neighborhoods and cities may affect the health of their residents.

We believe that this growing attention to creating safe and healthy spaces is a positive development, but that it does not go far enough. It has failed to address the acute threats to the safety of queers, immigrants, people of color, women, low-income people, the homeless, people with disabilities, and other communities.

Home.Land.Security. seeks to fill this gap by visually representing unsafe, threatening and harmful spaces as experienced by the most vulnerable. We will include some traditional health riskscapes, such as pollution and access to healthy food and open space, but will also include less traditional threats such as immigration raids, racist/homophobic discourse and hate crimes and police violence. Our team is working around the clock to assess which data sources are most appropriate to explore this topic.

Our site will redefine the current state of health and safety in a number of at-risk neighborhoods and communities within Los Angeles County. We aim to publish this information on the Los Angeles Community Action Network’s (LA CAN) website.

B. Functionalities 

Pan and Zoom Search: This interactive feature will allow the user to analyze health and safety layer data specific to an address or neighborhood, similar to Oakland Crimespotting. Layers within the map boundaries will pop up automatically, depending on what the user chooses to see.

Health and Safety Layer Selection: The user may select from a variety of categories, such as hate crime occurrences or immigration raids. These layers will be viewable in isolation or in combination with others.

Neighborhood Characterization: Demographic layers from 2012 will give the user a more complete picture of the neighborhood/city they are viewing.

Split Screen Comparison: Users will be able to compare two neighborhoods/cities at once with the split screen option.

Safety/Health Score Report: Similar to Walk Score, this site can generate a safety/health report from a 0 to 100 scale.

C. Wireframe diagram

D. Storyboard

E. Datasets

Twitter API

Yelp API

LA Times API

Air Quality Index

Google Maps API

Crimemapping API

F. Milestones

Week 6: Import all health/safety/demographic layers to map, make map searchable by neighborhood/city/region. Begin working on other interactive features to be finished by finals week

Finals: Write code that creates tables with demographic information, health and safety statistics. Write code that allows user to isolate and combine layers. Write code for report generator that produces safety/health score based on all available layers. Write code for interactive self-reporting form to allow users to report hate crimes and immigration raids.

G. Concerns

Some of the functionalities described may be too advanced for us to achieve within the timeframe, such as filtering through twitter feeds and generating score reports. Some data such as hate crime occurrences may be difficult to find/obsolete. Establishing pollution buffers could be difficult if we try to compare to real time traffic data.