Sharing Python Tools - Practical Solutions and Considerations
As a Python enthusiast, I've integrated automation into many of my daily GIS tasks over the past decade. However, I recognize that not everyone shares my passion for coding. At work, I'm always willing to answer questions and demonstrate programmatic solutions to everyday problems. But I also aim to avoid being preachy or condescending.
When I develop solutions that could benefit my team, I consider various methods for sharing them:
| Solution | Adoption Rate | Reason |
|---|---|---|
| Jupyter Notebooks | Low | People are generally reluctant to learn how to use them. |
| Python Script Geoprocessing Tools (ArcGIS Pro) | High | Team members are already familiar with these tools. |
| Python Scripts with UI (Tkinter) | High | Simple and easy to use, can be deployed with a .bat file. |
| Streamlit Apps | Low | Cannot run them on their own, security concerns related to running them in the cloud. |
Each of these solutions has its own pros and cons, but I have found ways to use each of them in different situations.
Important Constraints
| Constraint | Challenge | Solution |
|---|---|---|
| Python Package Installation | Installing new Python packages is not easy. | Constrain solutions to use only packages available by default in ArcGIS Pro. |
| Running Python Code | Running Python code correctly can be difficult. | Ensure solutions run within ArcGIS Pro or use simple deployment methods like .bat files. |
| Jupyter Notebooks | Notebooks are difficult to use unless interested. | Avoid using notebooks unless there is a strong push for their adoption. |
| Hosting Web Apps | Not easy to host web apps on the intranet. | Explore internal hosting options or justify the need for a dedicated server or cloud solution. |
| External Python Code Execution | Cannot use arcpy outside of ArcGIS Pro (e.g., Streamlit). |
Use alternatives like geopandas for geoprocessing and folium for mapping. |
Additional Suggestions
Documentation and Training
- Provide clear and concise documentation and training sessions.
- Create short tutorial videos to demonstrate tool usage effectively.
User Feedback
- Regularly solicit feedback to understand user pain points and improve tools.
- Incorporate user suggestions to enhance user-friendliness.
Incremental Adoption
- Start with small projects to demonstrate the value of automation.
- Gradually introduce more complex solutions as users become more comfortable.
IT Collaboration
- Work with the IT department to explore secure options for hosting web apps.
- Advocate for dedicated resources for automation and web app hosting.
Hybrid Solutions
- Use interfaces that users are familiar with and gradually introduce new ones with proper support.
- Consider hybrid solutions that combine simpler tasks automated within ArcGIS Pro with more complex external tools.
Community Building
- Foster a community of practice within the team for sharing scripts, tips, and tricks.
- Encourage collaboration and knowledge sharing to build a supportive learning environment.
By focusing on these areas, we can bridge the gap between technical and non-technical users, making Python-based automation more accessible and beneficial for everyone on the team.