1 of 37

Reimagine 911

Harvard Day of Service

2 of 37

We’re people-centered problem solvers

Showing that with the mindful use of technology

Government can work well for everyone

3 of 37

Our goal

A resilient government that effectively and equitably serves everyone

4 of 37

Our Values

Listen first

Include those who’ve been excluded

Act with intention

1

2

3

5 of 37

How we work

Demonstrate what can be done through action

Collect feedback about what works and what doesn’t

Build something small and use it with real people

Our Values

Listen first

Include those who’ve been excluded

Act with intention

1

2

3

6 of 37

The �Reimagine 911

Project

7 of 37

Our goal

The Reimagine 911 project is focused on helping cities develop more emergency response alternatives by identifying and understanding opportunities visible in the data.

8 of 37

Not All 911 Calls Need Police

50%

of mental crisis calls they receive from 911 can be helped over the phone alone.

1 in 3

Black men born in 2001 will be sent to prison sometime in their lives.

25%

of 911 calls are for crimes in progress. Most are not emergencies that require police.

9 of 37

Send government and community service responders when law enforcement is not needed.

10 of 37

The challenge ahead of us

The availability and uniformity of incident data is essential to knowing when to divert 911 calls.

This is a recurring challenge we heard over and over again through the various workgroups for Transform911, to the dozen plus interviews we had with experts in the field.

11 of 37

This is where you come in…

12 of 37

Data Standardization

13 of 37

Gathering Metadata

… but first

14 of 37

Data Dictionaries

… but first

15 of 37

Two Quick Things…

16 of 37

Stuff to Know About

Our 911 System

17 of 37

Speaking plainly..

911 was originally a tool of racial oppression

18 of 37

19 of 37

911 is a highly decentralized system

20 of 37

Vocab term!

PSAP » Public Safety Answering Point

Aka “call center”

Aka “911 districts”

21 of 37

People call 911 for many, many things - not just emergencies.

22 of 37

There are many 911 stakeholders for each PSAP

23 of 37

911 calls are deployed by many different agencies …so expect a mess.

24 of 37

We’ll need good context for these datasets and city to perform good analysis.

25 of 37

PSAP, Census, and dataset boundaries are arbitrary and may not match up.

26 of 37

Activity 1: �Data Dictionary

Understanding exactly what

The 911 data means

27 of 37

Data Dictionaries

Understanding what the data means

A data dictionary is a collection of descriptions of the data objects or items in a data model for the benefit of programmers and others who need to refer to them. Often a data dictionary is a centralized metadata repository.

911 open data across cities define terms differently.

For example: Location data in San Diego, CA is Address, City, State. But in Seattle, WA, it is Precinct, Sector, Beat.

  • What do these terms actually mean in each city?
  • Are these terms conceptually similar across cities?

These are questions we seek to answer when completing the data dictionaries for a city’s 911 data.

https://www.techtarget.com/searchapparchitecture/definition/data-dictionary

28 of 37

Activity 2:

City Metadata

Gathering local context about

the 911 datasets

29 of 37

City Metadata

Gathering local context

30 of 37

City Metadata

Gathering local context

Group 1: �911 Administrative Structures

31 of 37

City Metadata

Gathering local context

Group 2: �Community Service Programs

32 of 37

Activity 3:

Data Standardization

Helping us make 🍎 to 🍎�comparisons between 911 calls

33 of 37

Data Standardization

Apples-to-apples comparisons

Seattle, WA

Phoenix, AZ

Phoenix, AZ

APCO

Code

FIRE APARTMENT

MULTI-DWELLING FIRE

RESID FIRE

STRUCTURAL FIRE

34 of 37

Data Standardization

Apples-to-apples comparisons

35 of 37

Data Standardization

Apples-to-apples comparisons

Missing Descriptions

36 of 37

Data Standardization

Apples-to-apples comparisons

Cryptic Inter-Agency Codes

37 of 37

Data Standardization

Apples-to-apples comparisons

Demo Time!