Google recently released a simple little tool that essentially lets you run trends on word usage over the last 500 years. The tool searches over 500 Billion words contained in 1 million of the books they have scanned. I just played around with it for a few minutes and found some interesting beer related trends.
Beer and Ale had similar usage rates 200 years ago, with beer now the dominant term (lager is surprisingly unpopular).
There was a large spike in homebrewing interest during prohibition and then it crashed back down until the early 1990s.
Interest in wine is more variable than beer over the last 300 years.
There were spikes of interest in hops around 1700 and 1800, anybody have a theory on this one?
That's all I found, if you find any other good ones leave a comment.
"hops" is also the 3rd-person-singular form of the verb "to hop". It seems that this verb used to be more popular than it is now.
ReplyDeletehttp://ngrams.googlelabs.com/graph?content=hops%2Chop%2C+hopping&year_start=1600&year_end=2000&corpus=0&smoothing=3
That's awesome. Thanks for sharing.
ReplyDeleteDuring the early days of this country, before refrigeration/canning, people largely grew/raised stuff to feed themselves and/or barter with their neighbors. Since hops needed no preservation, they were valued as a CASH crop.
ReplyDeleteDuring the 1800's hop growing was a huge deal around here, and everyone would take time off from farming to work picking hops. (More cash flow). It was the social event of the year, for many. (They even had a huge-ass hop pickers picnic that warranted it's own train to transport people to and from the event).
Then the bottom fell out of the market and something (powdery mildew?) pretty much wasted the hop crop. After that, it moved to the PNW, where the majority of US hops still come from. (But Central New York is on the comeback trail....promise ;-) )
Sorry, I have go all pedantic engineer here since statistical signal processing is my area.
ReplyDeleteFirst, you should not have cropped out the scale on the graphs. More importantly, you have to be careful with the "smoothing" function. It looks like its simply averaging the data over some number of years. Look at the hops chart with smoothing set to 0 (http://ngrams.googlelabs.com/graph?content=hops&year_start=1600&year_end=2000&corpus=0&smoothing=0). The variance gets higher the further back you go. This is the result of several factors, which I won't assume I know. Since the data from back then is much less correlated, simply averaging isn't valid. So the peaks on the hops chart you show are probably real, but not as high as they look. As to what it means, that should be left to experts (Ron and Martyn).
(Oooh, Google lets you access the raw data. I wish I had more time.)
It is interesting that "porter" follows almost the same path as "ale".
ReplyDeleteLots of interesting thoughts, thanks.
ReplyDeleteTim, Blogger cropped the y axis, but if you click on a graph you can see the complete image. The smaller size picture was unreadable, and I didn’t want to force people to click on each one.
Agreed on the smoothing function, wasn’t trying to draw any hard conclusions. Google starts with it on by default. The issue with these search terms is that they are so infrequent that the data is more variable than it would be with more common terms.
Gotcha, sounds like Google is trying to be too smart.
ReplyDeleteThe graph for "homebrewing" is interesting. It tapers off in the late '90's then really takes off in 2002. The "Amber Waves" episode of Good Eats first aired in October, 2002. I think this show got a lot of folks interested in brewing - folks like me who had never heard of it before
ReplyDeleteThat episode was one of the things that got me into the hobby, despite the problems I now notice when watching it.
ReplyDeleteInteresting to see the negative correlation with "beer" and "ale"
ReplyDelete