The amount of work that went into this project is difficult to overstate. Hundreds of hours in the archives, several thousand scans, dozens of meetings, and countless late nights all came together to create this digital archive. It has been both an academic endeavor and a genuine passion project, and there are many people who all helped make this a reality. There is also much to be said about the creation of a digital archive like this from a technical point of view. It was a rigorous process of organizing and reorganizing files, of manual data entry, and a lot of troubleshooting.

Initially, this project began as an idea for a class, and therefore was limited in scope to the first decade of Weathervane's existence. It since expanded into covering from 1935 to 2000, spanning all of the content available at the University of Akron archives. The process of collecting and compiling all of the data now available went approximately as follows:

1: Figure out what plays were put on, in what order, and organize the physical files to reflect the order. Much of the playbill collection was amassed and organized by one man, Mr. Chisna "Cue" Henry Fleming. Because the collection was composed of both Mr. Fleming's efforts and a few other people's donations, there were some difficulties faced in finding every playbill and ensuring the plays were in the correct order. Then, each play title and its opening date were manually input into a spreadsheet for further organization.

2: Scan each page of the playbill twice for different file types. A 16-page playbill would therefore require 32 scans, and with each scan taking a minimum of 3 minutes (assuming all went well and nothing needed rescanned), this easily took the most time out of any part of the project. Early playbills were only 2 to 8 pages, but by the 1990's, they expanded to 30-32 pages. 

3: While the scans are running, input the relevant data into the master spreadsheet. As mentioned before, I created a spreadsheet that would allow for easier searches for names. For a while, I attempted to copy-paste the information into the spreadsheet to minimize errors and streamline the process, but that ended up taking significantly more time and caused issues with format, so each name was instead input manually. Thankfully, most of the playbills had multiple copies, so while one was being scanned, I was flipping through another copy.

4: Once the full playbill is scanned, digitally process it and upload it onto the University of Akron's archive website via ContentDM. I had to convert half of the files from JPEGs into PDFs, run a program to make them text-searchable, combine files to make one large PDF version of a full playbill, and then upload each one individually. This process was also rather tedious and presented the most opportunities for things to go wrong, either with the entry of metadata or the conversion process. Metadata, for lack of a better definition, is data about data- this includes things like the original date of the play, when it was uploaded and digitized, copyright information, etc.

5: Rinse and repeat for every playbill.

This is an oversimplified account, naturally, but I hope it adequately conveys the technical expertise that was required for each step of the process. It took time to refine the process and get everything as streamlined as possible, with many experiments with varying results along the way. This is also not to mention the other materials such as photographs and newsletters, although they followed a similar structure. 

There are so many people who all aided in the creation of this archive as well, and their efforts and kindness deserve to be acknowledged as essential- without them, this project would never have gotten off the ground. You may also read a little about Mat Cruz, the graduate behind the project, or explore the original spreadsheet of plays and names.