Distribution of languages on Twitter - plot order, freqpoly and grid.arrange different sized plots

dist.r

#!/usr/bin/env Rscript

library(ggplot2)
library(stringr)
library(gridExtra)

fmt <- function(){
    function(x) format(x,nsmall = 2,scientific = FALSE)
}


Y = read.delim("./distro.csv", sep=",", header=TRUE)
Y$rat <- 100 * Y$count/sum(Y$count)
Y$lan <- with(Y, factor(Y$language, levels=Y[order(-rat), ]$language))


p1 <-     ggplot(data=Y) +
    geom_bar(aes(lan, rat), color="orange", fill="orange", alpha = 0.5, stat="identity") +     
    xlab("Language") + ylab("% of Tweets") +
    opts(legend.position = 'none',
       panel.background = theme_rect(fill = "#545454"),
       panel.grid.major = theme_line(colour = "#757575"),
       panel.grid.minor = theme_line(colour = "#757575"),
       title = "Twitter Language Distribution")

p2 <- ggplot(data=Y) +
    geom_freqpoly(aes(lan, rat, group=1), size=2,  color="orange", alpha = 0.5, stat="identity") +
    scale_y_log10(labels = fmt(), breaks = c(0.001, 0.002, 0.005, 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1,2,5, 10)) +
    xlab("Language") + ylab("% of Tweets") +
    opts(legend.position = 'none',
       panel.background = theme_rect(fill = "#545454"),
       panel.grid.major = theme_line(colour = "#757575"),
       panel.grid.minor = theme_line(colour = "#757575"),
       title = "Twitter Language Distribution - Detail of Tail (log scale)")


png(filename = "distro.png", width = 800, height = 600, units = 'px')
  print( grid.arrange(
    p1, p2, ncol = 1, heights=unit.c(unit(0.3, "npc"),unit(0.7, "npc")) )
  )
dev.off()

distro.csv