Harry Potter Fandom on AO3, Example Analysis
Do we want to use this space to share specific analyses and techniques? To give it a go, here is something I pulled together this week end in answer to a comment by snowgall. I scrapped metadata of all public works under the “Harry Potter - J. K. Rowling” fandom on AO3, to see if there was anything interesting in the number of works and creators posting in recent years. This data doesn't address their primary concerns about back dating works, but the trends since 2007 are interesting enough. This post describes the results, and if anyone wants to know more about the how and why, I can elaborate in the comments.
I was a little worried that studying the Harry Potter fandom would be depressing. It's been years since any new material came out, so I assumed that productivity and participation would be on the decline. But apparently, it's not. Totally not. Below the cut are more details on how very wrong I was, with graphs.
First off, some basic descriptors
There are over 70 000 works on AO3 in the Harry Potter fandom. A lot works predate the archive, so we can only assume that AO3 holds a fraction of what has been written in the last two decades. However, AO3 has supported the importing fic from other sites, there are some some which are dated pre 2007.
The final publication date for the books was back in 2007, and the final movie in 2011. So if people were losing their enthousiasm for the series in the last three years, or getting distracted by other media currently in production, I won’t blame them. However, it looks like that isn’t what’s going on, at least in terms of the production of fanworks (posted to AO3).
The average number of works per day has been increasing steadily, and the most productive days for the fandom (on AO3) were this year, in 2015. Some of the variability is in the same forces affect other fandoms I've looked at, including the New Years bump (Late December to early February). There are likely HP fandom specific events behind other bouts of concentrated activity. Can any one explain the jump in mid November 2009?
So people are still making and posting fanworks in the Harry Potter fandom. Cool. But are people still joining the fandom? To look at this, I counted the creators contributing to the HP fandom per week, categorised by when they first started contributing to the fandom. So the first creator wave are those whose first works date before 2005, next those with works dated before July 2007, and those in each the following two years, and then, as steam continues to pick up, those contributing as of every half year. Just look at the graph, it’s much easier to interpret.
Two points from this:
1. There are a number of long term contributors in the HP fandom. Some are still active ten years after their first works!
2. There are still piles of creators posting in Harry Potter works for the first time. In any given week, approximately half of contributors have been posting in the fandom for less than six months.
Being from outside the HP fandom (OK, I posted a potterlock podfic last week...) I wonder if maybe all this activity is actually from folks writing crossovers or playing in Hogwarts AU. So I checked.
Here is a plot of works per month since 2007, showing three (exclusive) categories of Harry Potter related works:
1. Sole fandom works: those which list only “Harry Potter - J. K. Rowling” in their fandom field.
2. Multi fandom works: those which list two or more fandoms, including HP, but do not use AU tags.
3. AU works: Those who which fall under the hierarchical “Alternate Universe - Harry Potter Setting” tag, which mostly do not list Harry Potter as a fandom.
The trendlines agree with the eyeballing: All of these types of works are on the increase, though the works which are strictly HP are still the most common, by far. So again, I was wrong about how the Harry Potter fandom is continuing to grow.
But surely multi-fandom works and AU’s becoming more common? In this last plot I report the percentage of Harry Potter related works which are either multifandom or AU. Both trend upwards, linearly, meaning they are growing component of the works being published. Barring big changes, we can expect that next year 15% of works posted to HP will be multi-fandom, and 7% will be of other fandoms set in the Harry Potter Alternate Universe.
Curious as to which fandoms get mixed with Harry Potter the most? You can read the lists in the first page of this spreadsheet, but here are a few tidbits: Sherlock is the most common, followed by Supernatural, while, surprisingly, Buffy the Vampire Slayer and Les Miserables near the top of the lists as well.
Note: While I defined these categories exclusively, to count a work as either multi-fandom or AU, not both, the fact that the AU tags only become active in late 2009 suggest there is likely some confusion in how these terms are used. Also tracking AU's in tags is hellish and I wouldn't be surprised if a number of works have escaped the parent tag.
Note 2: all of these graphs can be interactive. Click on them to get the web version worth playing with. They are all embeded in this spreadsheet.
I was a little worried that studying the Harry Potter fandom would be depressing. It's been years since any new material came out, so I assumed that productivity and participation would be on the decline. But apparently, it's not. Totally not. Below the cut are more details on how very wrong I was, with graphs.
First off, some basic descriptors
There are over 70 000 works on AO3 in the Harry Potter fandom. A lot works predate the archive, so we can only assume that AO3 holds a fraction of what has been written in the last two decades. However, AO3 has supported the importing fic from other sites, there are some some which are dated pre 2007.
The final publication date for the books was back in 2007, and the final movie in 2011. So if people were losing their enthousiasm for the series in the last three years, or getting distracted by other media currently in production, I won’t blame them. However, it looks like that isn’t what’s going on, at least in terms of the production of fanworks (posted to AO3).
The average number of works per day has been increasing steadily, and the most productive days for the fandom (on AO3) were this year, in 2015. Some of the variability is in the same forces affect other fandoms I've looked at, including the New Years bump (Late December to early February). There are likely HP fandom specific events behind other bouts of concentrated activity. Can any one explain the jump in mid November 2009?
So people are still making and posting fanworks in the Harry Potter fandom. Cool. But are people still joining the fandom? To look at this, I counted the creators contributing to the HP fandom per week, categorised by when they first started contributing to the fandom. So the first creator wave are those whose first works date before 2005, next those with works dated before July 2007, and those in each the following two years, and then, as steam continues to pick up, those contributing as of every half year. Just look at the graph, it’s much easier to interpret.
Two points from this:
1. There are a number of long term contributors in the HP fandom. Some are still active ten years after their first works!
2. There are still piles of creators posting in Harry Potter works for the first time. In any given week, approximately half of contributors have been posting in the fandom for less than six months.
Being from outside the HP fandom (OK, I posted a potterlock podfic last week...) I wonder if maybe all this activity is actually from folks writing crossovers or playing in Hogwarts AU. So I checked.
Here is a plot of works per month since 2007, showing three (exclusive) categories of Harry Potter related works:
1. Sole fandom works: those which list only “Harry Potter - J. K. Rowling” in their fandom field.
2. Multi fandom works: those which list two or more fandoms, including HP, but do not use AU tags.
3. AU works: Those who which fall under the hierarchical “Alternate Universe - Harry Potter Setting” tag, which mostly do not list Harry Potter as a fandom.
The trendlines agree with the eyeballing: All of these types of works are on the increase, though the works which are strictly HP are still the most common, by far. So again, I was wrong about how the Harry Potter fandom is continuing to grow.
But surely multi-fandom works and AU’s becoming more common? In this last plot I report the percentage of Harry Potter related works which are either multifandom or AU. Both trend upwards, linearly, meaning they are growing component of the works being published. Barring big changes, we can expect that next year 15% of works posted to HP will be multi-fandom, and 7% will be of other fandoms set in the Harry Potter Alternate Universe.
Curious as to which fandoms get mixed with Harry Potter the most? You can read the lists in the first page of this spreadsheet, but here are a few tidbits: Sherlock is the most common, followed by Supernatural, while, surprisingly, Buffy the Vampire Slayer and Les Miserables near the top of the lists as well.
Note: While I defined these categories exclusively, to count a work as either multi-fandom or AU, not both, the fact that the AU tags only become active in late 2009 suggest there is likely some confusion in how these terms are used. Also tracking AU's in tags is hellish and I wouldn't be surprised if a number of works have escaped the parent tag.
Note 2: all of these graphs can be interactive. Click on them to get the web version worth playing with. They are all embeded in this spreadsheet.