wizjo

ds <- read.csv("dataset.csv")

## rainbow palette with gradients!
palette(rainbow(10, alpha=.5))

plot(c(rep(1,5), rep(.2,5)), ds[,4], pch=21, cex=(seq(1,10)-1) %% 5 +1, col=ds[,2]+1, bg=ds[,2]+1, xlim=c(-1,19), ylim=c(-1,7), axes=F, 
     xlab="x axis name", ylab="y axis name", main="Give a name to your chart")

for(i in 1:8){
  par(new=T)

wizjo

ds <- read.csv("that_csv_file.csv")
names(ds) <- c("test", "time", "city", "renderstart")
ds$time <- as.POSIXct(strptime(ds$time, format="%m/%d/%y %H:%M"))
ds$city <- as.character(ds$city)
ds$test <- as.character(ds$test)

cities <- names(sort(table(ds$city)))
tests <- names(sort(table(ds$test)))

wizjo

# > head(ds)
#                  hour activated organic viral 
# 1 2011-05-22 17:00:00        28       9    19 
# 2 2011-05-22 18:00:00        47      12    35 
# 3 2011-05-22 19:00:00        42      13    29 
# 4 2011-05-22 20:00:00        57      13    44 
# 5 2011-05-22 21:00:00        48      19    29 
# 6 2011-05-22 22:00:00        86      19    67 

# > dim(ds)

wizjo

## protovis style color cat10
palette(c("#1f77b4", "#ff7f0e", "#2ca02c", "#d62728", "#9467bd", "#8c564b", "#e377c2", "#7f7f7f", "#bcbd22", "#17becf"))
palette(rainbow(11, alpha=.8))

wizjo

grid = ximport("grid")

words = ["data", "visualization", "mining", "analysis", "design", "UX"]
colormode(RGB, range=255)
colors = [color(255,30,25), color(124,155,220),color(100,100,100), color(197,206,216)]
fontlist = ["Arial Black", "Trebuchet MS", "Georgia"]

def randContent(x, y, w, h, style=None):
  for i in range(40):
    x1 = random(0,w/2)

wizjo

scl = 3
r = .3 ## roundness of the rectangles
size(100*scl, 100*scl)
w=10*scl

mat1=[
        [0,2,2,0,2,2,2,0], 
        [1,0,0,1,2,0,0,2],
        [1,0,0,1,2,0,0,2],
        [1,0,0,1,2,0,0,2],

wizjo

## here's how you do a "SELECT col1,col2,col3,col4, count(1) AS cnt FROM ds_subs_less_table GROUP BY 1,2,3,4" thing in R ##

# first, find out all the unique combos of col1,col2,col3,col4
subs.geo <- data.frame(unique(cbind(
  as.character(ds.subs.less$source_tz), as.character(ds.subs.less$source_country),
  as.character(ds.subs.less$target_tz), as.character(ds.subs.less$target_country)
  )))
names(subs.geo) <- c("source_tz", "source_country", "target_tz", "target_country")

# next, figure out the frequency of each combo shows up

wizjo

select sub.query_id, sub.result, sub.sql, 
       avg(sub.time_lapse) as avg_time,
       max(sub.started_at) as last_run_at,
       count(1) as cnt
  from
(
select qe.query_id, 
       qe.result,
       qe.started_at,
       (qe.finished_at - qe.started_at) time_lapse,

wizjo

SELECT FIRST_VALUE(pkey_id) OVER(PARTITION BY cnt ORDER BY cnt DESC, pkey_id)
       AS first_id
  FROM (
SELECT pkey_id, 
       COUNT(*) AS cnt 
  FROM dimension_attribute_histories
 GROUP BY 1
       )sub
LIMIT 1

wizjo

SELECT sub2.period_id, sub2.time_id, sub2.network_id, COUNT(1) as cnt
FROM (
SELECT sub1.period_id,
sub1.time_id,
sub1.network_id,
SUM(sub1.pct_squared) AS dept_concentration_hhi
FROM (SELECT sub.period_id,
sub.time_id,
sub.network_id,
sub.job_type,