search

UCDavisLibrary / ava

American Viticultural Areas
https://ucdavislibrary.github.io/ava/
Creative Commons Zero v1.0 Universal
60 stars 56 forks source link

Created winter_highlands #922

Closed alisonsnwong closed 11 months ago

kurtishouser commented 11 months ago

Are there any to plans to handle the en dash "–" differently? See my Long Valley-Lake County comment.

kurtishouser commented 11 months ago

Also, it's best practice to merge new/updated AVAs into master in accordance with their effective dates. So Long Valley-Lake County, High Valley and North Coast before Winter Highlands.

MicheleTobias commented 11 months ago

Why does it matter what order they go in? We did not worry about that when we started the dataset because the goal was just to get them digitized well. Students started with easier boundaries and worked their way up to harder ones.

kurtishouser commented 11 months ago

With software development it's best practice for the master branch to keep the app in a stable state between numbered releases. As it applies to a dataset like this, it would be best practice for the master branch to represent all AVAs at a particular point in time.

So basically wait for Long Valley-Lake County, High Valley and North Coast to be completed and then in quick succession, merge them into master, run the shapefile and aggregation scripts, and merge those files into master. Then do a release -> 1.3.

Do the same for Winter Highlands and release again -> 1.4.

Combining all into a single release is valid too.

The order in which the data is digitized and pull requests are submitted doesn't matter, just how they're merged into master. Ideally there would be a one-to-one relationship between the FR document and the pull request but for large changes that affect many AVAs, separate PRs/AVA certainly make things easier to manage and track progress.

The 1.2 release with Gabilan Mountains was a milestone that brought everything current as of that date. Now the project is primarily adding and updating AVAs as new documents are published so keeping things in chronological order is easy.

These are just suggestions for best practice.