add 2025 roadmap

Author: nlebovits

docs/ROADMAP.md

@@ -1,85 +1,75 @@
-# Clean & Green Philly Summer 2024 Roadmap
+## Clean & Green Philly 2025 Roadmap
 
-## 🗓 Roadmap Overview
+### Overview
 
-Welcome to the project roadmap for Clean & Green Philly! After launching v1.0.0 in spring of 2024, we will spend the summer shifting our focus from frontend tasks to building out a more robust data pipeline and a project CI/CD workflow. We will still continue to work on front end stuff, but will have a more balanced distribution across all parts of the project, while also gathering input through user research.
+Welcome to the project roadmap for Clean & Green Philly! This is a living document that serves to outline the scope of what we intend to work on in the coming year. We intend to make quarterly updates to this document as the year progresses.
 
-## Front End
+We launched v1.0.0 of [cleanandgreenphilly.org](https://www.cleanandgreenphilly.org/) in the spring of 2024. Since then, we have made substantial updates to our data pipeline, incorporating parcel-level data for every property in Philadelphia, while undertaking several months of user research and workshops to understand where we want to go next.
 
-### 🌐 Relevant Organizations Page
+In 2025, we plan to begin taking advantage of our new data pipeline to pursue new analytical possibilities, and to implement the necessary updates to the website that we’ve identified through the user research process. At a high level, this will entail: 1) rebranding to more clearly articulate our aims as an organization and avoid confusion with the Mayor’s Office of Clean & Green Initiatives; 2) rewriting site content to align with the rebrand, as well as adding improved contact and donations pages; 3) focusing our front-end development on improved web and mobile accessibility for the cleanandgreenphilly.org site, 4) streamlining our data pipeline and adding improved data quality controls, 5) establishing more robust CI/CD, 6) exploring the possibility of using machine learning to identify vacant properties.
 
-- **Goal:** Create a page listing organizations actively working on related initiatives.
-- **Key Actions:**
-  - Research and compile a list of relevant organizations.
-  - Design and develop the webpage to display this information in an accessible format.
+#### Organizational Rebrand 🎨
 
-### 📝 Grant Writing Opportunities
+**Goal:** The Executive Director will recruit pro bono support to help Clean & Green Philly rebrand based on our strategic plan, in order to clearly articulate our aims as an organization and how we are distinct from other initiatives (especially the Mayor’s Office of Clean and Green Initiatives). This Code for Philly team will support the integration of this rebrand into the website as needed, e.g., by updating site content, graphics, colors, etc.  
+**Key Actions:**
 
-- **Goal:** Provide a comprehensive list of grant writing opportunities for local organizations.
-- **Key Actions:**
-  - Identify and list potential grant sources.
-  - Connect users to detailed grant writing resources and guides.
+- Recruit pro bono technical support for branding, marketing, graphic design, content writing.
+- Develop a branding kit with appropriate material for use in the newsletter, site, and elsewhere.
+- Communicate new branding guidelines for use on the website; update the site accordingly.
 
-## Dataset Enhancements
+#### Content Rewrite ✍️
 
-### 🔢 Development Risk Indicator
+**Goal:** Following the organizational rebrand, the Executive Director will secure support in writing new site content to ensure alignment with our rebrand and strategic plan. Additionally, new material is needed for technical documentation (e.g., a data dictionary and clearer explanations of data points like our “priority level” designation) and the “Transform” page.  
+**Key Actions:**
 
-✅ **Completed June 2024**
+- Identify and draft needed technical documentation (note that this will require some concerted thought based on comparable examples and best practices); mock and implement actual UI.
+- Conduct user research and a comparative analysis to identify what content is missing on the website.
+- Survey domain experts to identify content missing from Get Access and Transform pages.
+- Recruit one or more content writers to update content for these pages.
+- Coordinate with the UX team on mocking and implementation.
+- Consider mocking and implementing/updating the following pages:
+  - Contact
+  - Donations
+  - Grant opportunities for local organizations
+  - A page listing organizations actively working on related initiatives
 
-- **Goal:** Add a development risk metric based on building permits per census block.
-- **Key Actions:**
-  - Calculate a z-scored count of building permits issued in the past year per census block.
-  - Scale the data from high to low, noting the approximation to avoid a false appearance of precision.
+#### Improving Web and Mobile Accessibility 📱
 
-### 📊 Tangled Titles Model
+**Goal:** In order to support less tech-saavy users, we will prioritize enhancements to web and mobile accessibility for cleanandgreenphilly.org.  
+**Key Actions:**
 
-- **Goal:** Develop a model to identify properties with tangled titles.
-- **Key Actions:**
-  - Reach out to Pew about getting their dataste on tangled titles.
-  - Build and implement a predictive model to identify potential tangled titles in our dataset dataset.
+- Build on the user journeys mapped out in the 2024 community workshop slides, adding these to the website to help guide new users.
+- Create a tutorial for the “Find Properties” page using the [React Joyride guided tours package](https://github.com/gilbarbara/react-joyride).
+- Implement fixes to pain points as identified by the UX team in our user workshops.
+- Consider adding an “FAQ” page to the site, if needed.
 
-### 🏗 Negligent Developers Data
+#### Enhancing Data Pipeline ⚙️
 
-✅ **Completed June 2024**
+**Goal:** Improve the maintainability and functionality of the data pipeline by improving code quality and adding better data quality control checks.  
+**Key Actions:**
 
-- **Goal:** Include data on negligent developers in our dataset.
-- **Key Actions:**
-  - Define what constitutes negligent development.
-  - Implement a workflow to identify likely cases of negligent developers in our dataset.
+- Refactor the data pipeline to improve the maintainability of the codebase (considering using tools like `vulture` and `radon` for this purpose).
+  - Relatedly:
+    - https://github.com/CodeForPhilly/clean-and-green-philly/issues/717
+    - https://github.com/CodeForPhilly/clean-and-green-philly/issues/706
+- Add formal, consistent data quality controls using libraries like `dbt`, rather than our current ad hoc checks that exist in the pipeline.
 
-## Back-End
+#### Improving CI/CD 🚀
 
-### 📦 Containerization
+**Goal:** Improve our CI/CD process to reduce the workload on project leads and ensure the maintainability of the codebase.  
+**Key Actions:**
 
-- **Goal:** Containerize the entire backend setup to improve deployment efficiency and allow for migration to Google Cloud Services (GCS).
-- **Key Actions:**
-  - Build out a Dockerfile to run the entire ETL pipeline in the cloud.
-  - Deploy the container to GCS.
+- Add tests for key parts of the data pipeline, using a sample dataset stored on GitHub, or mocking data when necessary.
+- Ensure that tests are integrated into the GitHub Actions workflows to prevent breaking code from being added to the codebase.
+  - Along the way, address: https://github.com/CodeForPhilly/clean-and-green-philly/issues/830
+- Eliminate redundant GitHub Actions workflows while also making sure that the remaining ones all function as intended.
+- Review documentation to ensure that it is up to date and accurate.
 
-### ☁️ Porting to Google Cloud
+#### Exploring ML 🧠
 
-- **Goal:** Migrate our pipeline to Google Cloud.
-- **Key Actions:**
-  - Automate data updates with scheduled cloud functions to refresh our dataset monthly.
-  - Document migration steps and best practices for future reference.
+**Goal:** Explore the possibility of using ML with our dataset to identify vacant properties.  
+**Key Actions:**
 
-### 🧪 Implementing Data Quality Tests
-
-- **Goal:** Implement data quality assurance measures using tools like `dbt`.
-- **Key Actions:**
-  - Define data quality metrics and tests.
-  - Integrate `dbt` into our data processing pipelines.
-  - Schedule and monitor data quality checks to ensure continuous data integrity.
-
-### 🔄 Setting Up CI/CD
-
-- **Goal:** Establish a Continuous Integration/Continuous Deployment (CI/CD) workflow to streamline development and deployment.
-- **Key Actions:**
-  - Develop CI/CD pipelines using GitHub Actions including automated testing, building, and deployment.
-
-### 📧 Automated Stakeholder Communication
-
-- **Goal:** Automate the process of sending targeted email updates to stakeholders such as Habitat for Humanity about new, eligible properties.
-- **Key Actions:**
-  - Set up email triggers in our system to send out monthly updates.
-  - Customize email content based on stakeholder interest and relevance.
+- Perform EDA to identify key features in the dataset associated with vacancy indicators.
+- Explore basic modeling approaches (e.g., logit, RF) as proofs of concept to demonstrate the viability of using ML to identify vacant properties.
+- Work with key stakeholders to identify scalable ways to supplant the City’s outdated vacancy by building our own model.