This weekend, I made contact with Christian Gastmeier. He is a student at COGS working under the supervision of David MacLean. He is working remotely from Sudbury, Ontario. It is a capstone project: developing an AI-driven NLP pipeline for Spatiotemporal extraction.
I have included two paragraphs from his proposal:
“ Background
Since 2013, Bob Maher has maintained an extensive digital record of geographic observations through his blog ‘The Ernest Blair Experiment’. This body of work contains over a decade of high-fidelity local knowledge regarding the Annapolis Valley. However this data is currently ‘unstructured’, existing as a narrative text rather than a spatial database. Transforming this narrative into a Geographic information System (GIS) requires Natural Language Processing (NLP) to bridge the gap between human story telling and provincial resource management.
Purpose
The purpose of this project is to develop a reproducible cloud-based AI pipeline using ArcGIS to extract geographic themes from unstructured text. Leveraging the full ArcGIS online (AGOL) suite this project will:
- automate extraction
- Geocode and Validate
- Cloud- Native Visualization
- Multi-media Integration
- System Scalability
Where things stand today (Monday, May 4,2026) ?
- Christian emailed his Capstone project proposal
- We had telephone conversation
- Bob agreed to provide hard copy ( 13 volumes of his ernestblairexperiment blog
- Christian is working remotely from Sudbury, Ontario.
- Deliverables:
- Spatiotemporal Cloud Database
- Automated Extraction Pipeline
- The Ernest Blair Digital Atlas
This blog will be sent to a wide range of industry/academic partners.
If you are interested in the full project proposal, contact Christian at W0529818@nscc.ca Future blog posts will track our progress.
Edward is unavailable for a few days. He suggested that I post this blog, as is.
Ernest Blair Experiment 