for blogpost: http://benjaminlmoore.wordpress.com/2014/01/09/wikipedia-is-dead-long-live-the-pedia/

wikiConsumption.R

require(ggplot2)

lm_eqn <- function(m) {
  # fromm: http://stackoverflow.com/a/13451587/1274516
  l <- list(a = format(coef(m)[1], digits = 2),
            b = format(abs(coef(m)[2]), digits = 2),
            r2 = format(summary(m)$r.squared, digits = 3));
  if (coef(m)[2] >= 0)  {
    eq <- substitute(italic(y) == a + b %.% italic(x)*","~~italic(r)^2~"="~r2,l)
  } else {
    eq <- substitute(italic(y) == a - b %.% italic(x)*","~~italic(r)^2~"="~r2,l)    
  }
  as.character(as.expression(eq));                 
}


views <- read.csv("http://reportcard.wmflabs.org/data/datafiles/rc/rc_page_requests.csv")
views$Month <- as.Date(views$Month)

ggplot(views, aes(x=Month, y=English/10e6)) + 
  geom_line() + geom_smooth(method="loess", col="red") + 
  labs(y="Pageviews (Million)", x="", 
       title="en.wiki monthly pageviews 2008-14") +
  theme_bw()
  

mobile <- read.csv("http://reportcard.wmflabs.org/data/datafiles/rc/rc_page_requests_mobile.csv")
mobile$Month <- as.Date(mobile$Month)

ggplot(mobile, aes(x=Month, y=English.Mobile/10e6)) +
  geom_line() + labs(y="Pageviews (Million)", x="",
       title="en.wiki mobile monthly pageviews 2008-14") +
  geom_smooth(method="lm", col="darkgreen") + 
  theme_bw() + geom_point() + 
  annotate("text", x = as.Date("2013-01-01"), y = 50, col="darkgreen", size=6,
     label = lm_eqn(lm(English.Mobile/10e6 ~ Month, mobile)), parse = TRUE)

uv <- read.csv("http://reportcard.wmflabs.org/data/datafiles/rc/rc_comscore_region_uv.csv")
uv$date <- as.Date(uv$date)
# filter non-matched dates
uv <- uv[uv$date %in% views$Month,]
views <- views[-nrow(views),]
vvu <- data.frame(month=views$Month, views=views$English, uv=uv$World)

ggplot(vvu, aes(x=month, y=views/uv)) + 
  geom_line() + geom_smooth(method="loess") +
  labs(y="Mean pageviews per unique visitor", x="") +
  theme_classic()