It is finally time to discuss the process for working No-Match Quotations! I know you’ve all been waiting with bated breath to find out more about my approach so, without further ado, let’s get started!
Now, there are two ways to work No-Matches (aww yeah – two technical boring things!): the database and a spreadsheet. These quotations start in the database, where users will look for the correct informational match such as author, title, etc. Each author used in the dictionary has a specific tag, along with each of their works, so you could feasibly search for all quotes written by Shakespeare in the dictionary. It’s a really nifty system, but in cases when the quote doesn’t have that information/tag attached, they can slip through the cracks and the subsequent results would be incomplete. And that’s where the cool work of matching these quotations come in!
The Database TM
Okay, it’s not really trademarked, but it does sound cool. Anyway, the database was created as a very fancy way to match quotations to the correct source. When you click “Start Matching,” the database will present you with a quote and possible matches next to it, which allows you to hopefully find a match among them. If you find one for the quote, you mark it as a correct match, if not, you skip it – it’s all pretty straightforward!
If you’re feeling adventurous, and have the training, you can go a bit more in-depth if you don’t find a match in the results. To do this you would start a search, find the correct result, and create a custom match. That involves a lot more research and a bit more training on the XML tag specificity (not actual XML training, I’m not that cool), because tags must match exactly or when you try to find, for example, all the quotes written by William Shakespeare you’ll only end up with a partial list. The rest of the quotes could end up tagged to the wrong author or under a different spelling, like Billiam Shakespeare. While that would be pretty funny, we probably shouldn’t do that.
The Spreadsheet
This version of matching is more work but it’s my favorite! However, I’m a nerd who loves research, so it’s not surprising. The spreadsheet approach also requires using multiple spreadsheets, so I hope you love spreadsheets. We start with the No-Match, which are quotations that came from the database (not TM) that never found a match. These require a bit more research to either determine who wrote it and/or the work it came from. Also needed is the Bibliography spreadsheet, which is a list of all our current tags for authors, works, and the links for the works we’re pulling from. We pull from the same edition for each specific quote (for example, the 1711 edition of I’m A Real Book, rather than 1711 and 1832) so that we’re being as consistent and accurate as possible.
Once we have searched and found the (hopefully) correct author, which can take a bit of time, we’ll go to that Bibliography sheet and find the link to the work we’re sourcing from. Now pull it up, find the quote (again, hopefully), and then plug that completed information into the No-Match sheet.
And with that, great job! You’ve got yourself a completed quotation (done in one of two ways, you skilled person, you) and there’s only a few thousand to go