Previously, on How Can People With Websites Share What Books They’ve Read With Each Other?:
- Tom Critchlow suggested making a kind of distributed GoodReads using JSON,
- Matt Webb suggested using modified RSS,
- and Jeremy Keith suggested using HTML microformats.
I expect there are other episodes I’ve missed, but anyway. They all sound fine.
Having spent some time writing code to track the past 22-40 years of my reading (22 consistenly, 40 with gaps) I thought I’d share a few things I’ve done or thought about, in terms of the data. Some of this will be too specific to my needs to be useful to a generalised effort, but still.
As well as having start and end dates for when I read something, I have a boolean “Finished?” field. Because I want to acknowledge the occasions when I’ve spent time with a book but not made it all the way through.
Because I was back-filling reading from lists of books I’d read in particular years, I didn’t have detailed start and end dates for some of my reading. In these cases (a) I’d use that year for only the end date (leaving the start date blank), and (b) I’d use an extra field for specifying the granularity of those dates.
As always when I have to record dates of variable granularity I end up adding a
granularityfield for each date, using values taken from Flickr’s API. A nice thing is that the HTML
<time>tag doesn’t require a complete date (or time); its
datetimeattribute also accepts partial
Tom says his idea needs “further research into a universal book ID solution.” ISBNs go some way towards this, but they’re not ideal – the same book in different editions, and so different countries, will have different ISBNs. Do you want to say “I’ve read this edition of Generation X” or “I’ve read Generation X”? Plus, old books won’t have ISBNs. And some modern books – e.g. some self-published ones – aren’t going to have them either.
I’ve wondered what would be more useful, and where else I could link to other than (ugh) Amazon. Maybe OpenLibrary is a good bet (e.g. Generation X there)? They have Open Library IDs (OLIDs), and also support ISBNs, OCLC Numbers and LCCNs.
As well as tracking the reading of books, I also track the periodicals I read. So I have a field to distinguish between the two. Because this is enough of a distinction for me, that’s fine, but I can imagine adding this to a more universal tool would create problems of categorisation. “We should add ‘manifesto’!” “Where’s pamphlet?” “Is this a ‘journal’ or a ‘magazine’?”
Similarly, I would like to add categories or tags to my archive of reading, mainly so I can view lists of all the books I’ve read about “history” or “technology” or whatever. But, in terms of getting something simple and universal off the ground, this sounds like opening a big taxonomic can of worms/maggots/eels.
Each of my publications can be part of a series, a field that I currently only use for periodicals. e.g. a particular issue of the LRB is collected in the London Review of Books series. I also want to be able to group books like this but haven’t yet decided how to manage that, because a series of books – e.g. all volumes of C.P. Snow’s Strangers and Brothers series – doesn’t feel like the same thing as a series of periodicals.
When adding authors to a publication I can specify their role (author, editor, translator, illustrator, etc.) and the order in which they appear, which often feels important.
Another wrinkle, when thinking about modeling all this, depending on how fussy you want to be, is that occasionally an “author” is a group or organisation, rather than an individual, e.g. (“Folk Horror Revival”).
Each publication on my site has an optional URL field for linking to its official site — some books have their own website, and some periodicals have a page for each issue. There’s also a second URL field, that I use to link to my own notes or review of the book. (UPDATE: I’ve realised since that I should allow for zero or more links, each a pair of URL and title (“My notes”, “Official site”, etc.))
I keep thinking of adding a text field to hold a short review and, whenever I do so, I have to remember that this should be related to a reading and not a publication — I will probably change my mind or notice different things when reading a book years apart.
As I say, many of these points are specific to my own uses, quirks and obsessions. And no doubt people who deal with data about books professionally will have even more considerations. But, still, there are some thoughts.