Changes

After reading through the paper ["The USPTO Patent Assignment Dataset:Descriptions and Analysis"](https://www.uspto.gov/sites/default/files/documents/USPTO_Patents_Assignment_Dataset_WP.pdf) and taking into account are needs for the Assignment database (the ability to connect the two databases together, the ability to trace conveyance of a patent over time) I decided on the following structure for the Assignment database.

Purpose: store all the assignors from all the assignments Each entry represents: the one assigning/granting the license for the entry in the ===ASSIGNMENT===table  Primary key: rf_id and assignor_name All columns (variables): rf_id, assignor_name

Purpose: represent the conveyance of a particular patent overtime Each entry represents: an individual transaction for a patent (not unlike assignment, except this table has information about the assignment in relation to other assignments for the patent Primary key: rf_id All columns (variables): Two possibilities, depending on whether we want data to be repeated or not. Not-repeated (with exception of primary/foreign key): rf_id, next_transaction (also rf_id), num_conveyance (number in order of conveyance by date), acknowledgment_date (in ASSIGNOR for the assignment data paper but would be moved here. Could originally be stored in ASSIGNOR until CONVEYANCE is generated), execution_date (same deal as acknowledgment date, should probably be moved here; if there are multiple execution dates, we will store the last one) conveyance_type Repeated, also include: assignee, assignor, patent_id 

Purpose: store extra information relevant to the patents represented in ASSIGNMENT Each entry represents: information about the documentation and the patent for a particular entry in ASSIGNMENT Primary key: rf_id All columns (variables) when created: rf_id, app_num (application document USPTO number), app_date, app_country, pgpub_num (pre-grant publication document USPTO number), pgpub_date, pgpub_country, grant_num (granted patent document USPTO number), grant_date, grant_country, invention_title, language (of invention title – could be potentially useful and interesting to investigate), reel_num, frame_num (reel_num and frame_num were in ASSIGNMENT for the uspto data paper, but I think they would be more useful here) 

Purpose: represent assignment transactions – central table for the database Each entry represents: an individual patent assignment transaction Primary key: rf_id All columns (variables): rf_id, assignment_id (US patent assignment id), correspondent_name, correspondent_address1, correspondent_address_2, correspondent_address3, correspondent_address4, record_date (date recorded with USPTO), last_update_date, page_cnt (page count of assignment record)