Spoilers and Metadata

Spoilers

I don't think that many people like it when a story is "spoiled" for them. This is evidenced by that fact that, when referencing a story, it is pretty common for any plot details to be prefaced with a "spoiler warning". This allows readers to stop if they don't wish to have such details revealed to them.

Personally, I too don't generally like it when a story is spoiled for me. I prefer to enjoy the story in the way that the creator intended, with suspense and tension intact. There is simply much less excitement built up throughout the story if I already know what is going to happen beforehand.

However, knowing some or all of the main plot points does not necessarily stop me from reading a book or watching a film for the first time. Such knowledge is often difficult to avoid, particularly in the case of classic or popular stories which are referenced very frequently.

The canonical example of a spoiler is probably the revelation that Snape kills Dumbledore (it's so widespread that a spoiler warning doesn't seem warranted here). I've no idea how people who haven't yet read or watched the Harry Potter series could escape having heard this spoiler. Luckily, there is quite a lot more to the Harry Potter story than just that particular event. Perhaps one would even be more inclined to read Harry Potter after hearing the spoiler so many times, just to find out why it's apparently such a big deal.

Metadata

Now I want to turn my attention to the idea of metadata, which is data about data. For the rest of the article I'm going to specifically discuss books, but similar ideas could easily be applied to other media as well.

When I read a book, I generally know a bit about it before I start. Usually I at least know the title, the genre, and the author. I'll often check the number of pages, or at least have a rough idea based on the book's thickness. Perhaps I'm familiar with other work by the same author and can anticipate the book's style. I probably also have some idea of when and perhaps where the book was published. But none of this information is part of the actual plot of the book. It's all metadata. The data is the words in the book (the plot). The metadata is additional information about or related to the plot or book, but not the plot itself.

For example, it's clear that the length of a book is related to the the book's plot. A shorter, less complex plot requires fewer words to express, and thus requires a book with fewer pages. But it's also clear that the length of the book isn't part of the plot itself. Knowing that my copy of The Count of Monte Cristo is 441 pages long doesn't add anything to the story. As such, the length of a book is metadata.

Metadata as Spoilers

I was thinking about this recently while reading a sci-fi novel. As I was approaching the end of the book, I began speculating about how the plot would be resolved. It seemed like there was still a lot that needed to happen to reach a suitable conclusion. A few pages later, a large plot twist showed me that I was wrong and that the path to resolution was in fact much quicker than I had anticipated.

I then started to wonder if the plot twist was made more predictable because of the discrepancy between the small number of remaining pages and what felt like a lot of plot points left to tie up. In reality, I didn't speculate enough to predict the twist because the book was good and I just got on with reading it. But, if I'd thought about it a bit more, perhaps I would have suspected that some sort of plot twist was required to end the story relatively quickly. Perhaps I would have even predicted what the twist would be. Would this have been a bad thing? Would the excitement of the story have been lessened for me?

What if I had read the novel without knowing its length? It wouldn't be possible for me to speculate on the speed of resolution if I didn't know when the book would end. In an extreme scenario, I could conceivably read a book without any context: no knowledge of the title, author, publication details, or length (imagine an E-reader that supplies the user with nothing but the book's text).

However, my taste in books is not unbounded. I judge books by their covers and other metadata in order to choose ones that I'm reasonably certain I'll enjoy. But what if I had a book recommendation system that knew everything I'd read and how much I'd liked it, and then simply offered me a stream of text of unknown length from a book of its choice? Could this be more enjoyable to read than a book about which I have metadata? Frankly, I don't know. I do however suspect it could at least expose me some good works that I may not otherwise choose myself.

I'd be curious to hear from people who actually do something like this, and read books (or consume other media) knowing very little metadata at the outset. What pieces of metadata do you try to hide from yourself? Are there some that you consider more revealing than others?

I'm also interested in folks who go the other way and like to know a lot about a book before reading. Are you unbothered when you are spoiled? Do you actually enjoy being spoiled and seek out plot details beforehand?