There are certainly many questions that arise from asking a broad question about “representative-ness.” What does representation imply? Would we be simply be comparing a part to a whole—Penn’s rare books collection being the part and the rest of the world’s holdings being the whole, thus trying to decipher how representative our collection is out of all the preserved novels out there in terms of quantity? Or, are we speaking of more complex representative qualities, such as: How well does our collection represent sentiments of certain decades? Does our collection accurately reflect trends or changes that occur in novels of the same time period, as shown by the worldwide collection? And departing from the text: How well does our collection represent (un)common printing methods of certain years/decades? Is our collection an accurate representation of material that people of specific socioeconomic groups were able to access (through systems like circulating libraries or private purchase)?
A practical solution to figuring out the “representative-ness” of Penn’s RBC in many different terms: Create a branch of a website like Worldcat that not only tracks the first editions of rare books, but all editions of those books. It would do this by searching the catalogues of all online databases that include old/rare books, targeting the online databases of colleges, universities, and major libraries. Then it would take those books, catalog them, and be searchable using the grouping method that we talked about, used commonly in online shopping (what was that called again?). Some of the filters could be year published, open-access availability, edition (if known), and collection location (this last grouping could lead back to each individual collection’s more detailed website, since the purpose of the greater website would be more like a bibliography). If a system like this were in place, then it would be easier to determine the representative-ness of our collection. Perhaps this would truly be the “Transatlantic Fiction Project,” since it would certainly involve collections across the world, instead of what we’re doing which seems more collection-specific.
An issue I can see running into here is the fact that not every library is completely online yet (apparently even Penn still uses a card catalog for some things), so the world database of rare books wouldn’t be complete… although it would never really be complete, since there are certainly rare books out there that are privately owned and will never be declared. I’m also unsure if there are copyright laws that prevent taking bibliographic information without permission or payment, which could be a problem; another issue could be increasing the possibility of mechanical errors, since I imagine that in the transfer or calling up of information from one smaller website to a greater one could welcome many instances of typos, misspellings, eliminated words, errors in labeling/tagging of entries, etc. These all seem sort or minor, compared to the greater good of having such a website. Ideally, many of the books would eventually be digitized and made into searchable PDFs at no cost to the reader (we can only hope about this last part, I suppose).
This is just an idea, but perhaps another way we could determine the “representative-ness” (or maybe I mean relevancy) of Penn’s RBC is by examining the reviews that are associated with the text. Maybe we could infer how popular or relevant the text was to people by examining the reviews and noting how many reviews of the books there were. Perhaps this would give us some insight into how well our collection represents books of the time: If it turns out that there are very few reviews of all the books Penn holds (which I know isn’t true), then maybe our collection represents a more elite or unrepresentative portion of the whole. This is an oversimplification, but I still think that reviews might be useful when examining these books (or possibly, even something that we might eventually want to include in the database as a supplement to the novels).