Fasten Your Seat Belts
November 2nd, 2007Today’s post is about the humble paper clip. (Hey, I can’t write about weighty topics every week, right?) Sure, it’s a great fastener, but the Wikipedia link talks about using paper clips in many diverse applications, from outward signs of government solidarity to lock picks. I have my own unconventional use for them: as a bookmark. (Sorry guys, I’m not McGuyver here.)
My problem is that when I handle or lend books to others, the normal bookmarks I put in the book inevitably fall out, a problem which paper clips fortunately don’t have. Though I’m surely not the first person in the world to think of using paper clips as bookmarks, in my particular case I stumbled on the concept at work, where paper clips are both readily available and plentiful.
I don’t just use paper clips to keep my place, though. The large number of available “bookmarks” makes it easy to mark anything I deem too important to forget, or anything that I want immediate access to should I use that book as a reference.
Thus, since I’ve started my paper clip annotating, I’ve been riddling my books with them, inserting them any place I see that’s even remotely interesting. It’s been a remarkably effective system for me so far. (Having enough metal on me to set off the detector at the airport is a secondary benefit.) These days, when I pick up a book I’ve read before I usually just head straight for the clips.
Paper clip bookmarking isn’t a perfect system, however. Books that are very succint and content-rich like “What the Numbers Say”) quickly accumulate an awful lot of paper clips, and end up lopsided like an accordion and hard to stack other books on. You can compensate for this by spreading out paper clips across the top of pages, but it only goes so far. Plus, there’s the odd looks you get from friends when you lend them books filled with paper clips (which I do often). I just tell them I’m a paperclip collector, which usually stuns them into silence.
Anyway, when I sat down and thought about it, the number of paper clips I put in a book is actually fairly representative of how much lasting value I got out of reading it. This is not an objective measure, even by my own standards, as it is tempered by what I already know - I won’t mark the same concept twice in two different books. And some books have really dense passages, which I can only mark with one paper clip. Conversely, some books are more about a vague “reading experience”, and I don’t bother to “clip” those either. Still, especially for technical works, it’s a good guide.
Which also means that - wait for it - those clips are an excellent source of data. (See how neatly that tied in? It’s almost like I’m a professional writer already.) With this data-oriented perspective in mind, I went ahead and tallied the paper clips for all the books I’ve “clipped” and wrote down how many pages they had marked. Since some books are obviously bigger than others, to be fair I also used a very rigorous and scientific size classification scheme: I eyeballed all the books and called them “small”, “medium”, “large”, and “extra large”. And to give an you a relative measure, I divided the number of pages by the number of clips to give you the pages per clip (PPC) ratio.
The weird thing about the PPC ratio is that lower is better. That is, lower pages per clip means there are more paper clips relative to the book’s length. The clip-per-page ratio probably makes more intuitive sense, but that’s a decimal ratio, since there are many more pages in a book than paper clips. And as I read in “What the Numbers Say”, people don’t work very well with decimals. It’s so bad that we don’t even have a real language for longer decimals. We have to read out every number and then maybe tack on an awkward “thousdandths” or “millonths” or whatever. So, I went with the inverse ratio even if it’s slightly more confusing.
Thus tallied, I slapped all that data into a spreadsheet, complete with fun charts and tables. That may seem like overkill (and it is, though my long-time reader(s) are surely getting used to my tendency towards that by now) but I got to use a lot of what I know about charts and tables to make the spreadsheet and it was a good opportunity to practice.
Anyway, on to my findings: Here are my top 3 books, ranked by pages per clip (actually 4, since there was a tie for 3rd):
*Not including appendicies, bibliographies, indexes, notes, references, or the like.
Of course, I think all those books are great. Heck, I even wrote about “Show Me the Numbers” last week and “Super Crunchers” three weeks ago in this very blog. (Hey, it’s only shameless self-promotion if you have a big audience.) The other 2 books up there are equally as interesting, though I won’t have time to go into them here.
As a side note, I should mention that a high PPC ratio means I think highly of a book, but the opposite is not necessarily true. That is, just because a book doesn’t a have a lot of paper clip bookmarks doesn’t mean it’s not a good read, because there’s a lot of reasons why that might be the case (of which being a bad book is only one reason).
How about general findings, then? Keeping last week’s post about charts in mind, I made a couple of charts to help visualize trends.
The first chart is a general scatter plot split into four quadrants (click the thumbnail for a bigger version):
The second chart is the same scatter plot, but color and pattern coded by page size, so you can see if there’s any relation between pages per clip and page size:
(In a twist of self-reference that would make Douglas Hofstadter proud, “Show Me the Numbers” is the book with the best PPC ratio, and the graphs used to display that fact were taken directly from a charting example in “Show Me the Numbers”. Now if I can just get the author to reference my blog in his next edition, I can REALLY make your head hurt…)
So, trends. First, the norm seemed to be books without a lot of pages or paper clips. I guess great, compact ideas are not common in the books I read, even though I really enjoy most of them.
Second, lengthier books do not seem to be any higher quality for me than shorter ones. In fact, most of the books with a lot of paper clips were often some of the shortest. I don’t think that’s going to surprise anyone, but I thought it was interesting. (Which begs the question of why my posts are so long. And that’s because, uh…you can’t use paperclips on a webpage. And I’m not responsible for any damage to your monitor if you try.)
Third, page size and pages per clip are weakly correlated, at best (though this is a small data set). This is unexpected because you would think a smaller page size would really help things. With smaller pages, there are more opportunities to add clips for the same amount of words. If anything, though, it seems like the larger the page size, the more paper clips I’ll put in a book. Even then, the correlation does not seem strong, visually speaking. As you can see from the medium-sized books, books of all page sizes (except the one anomalous “Extra Large” book) are scattered throughout the plot.
And so a new measurement is born: PPC (pages per paper clip). At last, a quantitative measure of a “good book”. Now when I lend a friend a book, I don’t just have to say “you should read this, I think you’ll like it”. I can instead use authoritative pronouncements like “of COURSE you should read this book, it’s got under a 15 PPC! Duhhhh.” Then I’ll hand them an accordion-shaped book that will mark them as a security risk at the airport. And they’ll be thoroughly confused, of course. But the number will live on in my heart - and in a whole pile of books that just never seem to stack right…
| | | del.icio.us |
November 2nd, 2007 at 9:15 am
Two things: It looks like the book you love so much is called “Show me the Numbers,” not “Show me the Data.” I was trying to find “Show me the Data” on Amazon and failed. Also, if you want to get rid of your undataful SML classification scheme, Amazon has data on the dimensions of every book, so you could use a clips per words per cubic inch.
Good catch, Matt, I appreciate it. Turns out I probably got the title wrong because of a “show me the data” joke I made in an earlier post, ha! That’s also a good idea about the Amazon dimensions. I considered getting a word count from them, but book dimensions is just as good, if not better. My only worry is overkill at this point, I think. Bringing in cubic inches would be hardcore.
Thanks again for reading and commenting!
- Dave
November 2nd, 2007 at 6:25 pm
You know I just started to learn how to pick locks, so for me a paper clip and a bic pen with metal clip are the best tools for the job.
I, too, know a lot about lock picking. If you ever want to talk shop, let me know. I’ve definitely picked a lock with a paper clip before.
- Dave
November 9th, 2007 at 2:10 pm
I see you got the charts hooked up the way you wanted with thumbnails, etc. Nice job.