remove the old skool files

README.mdown

@@ -7,6 +7,38 @@ Instructions
 ============
 Please follow the instructions on our [wiki](https://github.com/cooperhewitt/collection/wiki)
 
+Collections items as JSON files
+============
+
+This departments, exhibitions, objects, people, periods, roles and types folders
+contain Cooper-Hewitt collection data exported as individual
+[JSON](http://www.json.org/) files.
+
+File names and parent directory structures are generated using the unique ID of
+the object itself. For example the full path for [iRobot's Roomba vacuum
+cleaner](http://collection.cooperhewitt.org/objects/18704235/) (ID #18704235)
+is: [objects/187/042/35/18704235.json](blob/master/objects/187/042/35/18704235.json)
+
+Paths are generated by chopping numeric identifiers in to groups of three (or
+less) starting front to back. A reference implementation can be found in the
+[utils.py](/blob/master/bin/utils.py) library.
+
+Objects as JSON files (in Git(Hub))
+============
+
+It should be noted that the performance of Git (and GitHub) when dealing with
+very large repositories is still less than ideal. It appears to be a problem
+inherent to Git and its indexing processes.
+
+Which is to say: Everything still works but it can be slow. In practice this has
+the side-effect of making some features of GitHub, like pull requests, nearly
+impossible because the web application is configured to time out before indexing
+(of a very large repository) has completed.
+
+There is a good thread about the problem and some potential solutions over here:
+
+http://thread.gmane.org/gmane.comp.version-control.git/189776
+
 Usage Guidelines
 ================
 Are there any restrictions?