diff --git a/.gitignore b/.gitignore
index fa4c381..b9da7ce 100644
--- a/.gitignore
+++ b/.gitignore
@@ -227,7 +227,7 @@ vignettes/*.pdf
 .Renviron
 
 # pkgdown site
-docs/
+# docs/
 
 # translation temp files
 po/*~
diff --git a/docs/changelog.md b/docs/changelog.md
new file mode 100644
index 0000000..094a2e9
--- /dev/null
+++ b/docs/changelog.md
@@ -0,0 +1,54 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/)<!-- , and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html) -->.
+
+## [Unreleased]
+
+### Added
+
+- Auto-generated Markdown files using Frictionless (resolves #64)
+- Adds empty template SQLite databases (one using integer and the other using string IDs) to hold a network formatted in accordance with GMNS. This is a convenience for users who may wish to exchange GMNS-formatted networks via databases rather than `.csv` files.
+    - Also adds SQL create statements used to create these. The database and create statements are placed in `usage/database`
+- Documentation website (where you're hopefully reading this now!) at https://zephyr-data-specs.github.io/GMNS/
+
+### Changed
+
+- In the json schema files, the ID fields (`node_id`, `link_id`, etc.) has always been defined as type "any", allowing text or integer to be used. This follows the example of `unique_ID` as defined in GTFS. It is certainly possible that an application may further restrict ID fields to be integer (issue #83).
+- The automated generation of markdown has resulted in several non-substantive changes to the markdown files:
+    - Their names are in lower case (e.g., `node.md`) rather than mixed case (e.g., `Node.id`)
+    - Datatypes for the ID fields are "any", carried straight from the json files
+    - Datatypes are number (rather than numeric), string (rather than text)
+
+## [0.95] - 2024-06-14
+
+### Added
+
+- Add expanded example network (Cambridge, MA)
+
+### Changed
+
+- Branch names were changed from `development` -> `master` to `develop` -> `main` (fixed #62).
+- Directory layout changed as follows (fixes #63):
+    - `Conversion_Tools` -> `usage\conversion`
+    - `Images` -> `docs\img`
+    - `Small_Network_Examples` -> `examples`
+    - `Specification` -> `spec`
+    - `Specification_md` -> `docs\spec`
+    - `Validation_Tools` -> `usage\validation`
+
+## [0.94] - 2023-06-15
+
+### Added
+
+- Config information is now stored as a new `config` table at the dataset level (fixes #59)
+- New `curb`_seg table to provide a separate segment object for curbside regulations, which may change at different locations (and more frequently) than segment-level changes to the travel lanes
+- Addition of zone tables to two examples (fixed #56)
+
+### Changed
+
+- Adjustments to the signal phasing tables to fix #57
+- Standardized `bike_facility` type names to fix #58
+- Updates to validation code to use Zephyr-hosted repo (GMNSpy#2)
+- Minor updates to reflect spec changes above