7th Anniversary

I reached a million views on 2017 September 27th.

Near the start of September, I had wondered if I would reach a million before my 7th anniversary, which is today. I used the traffic to this site to predict when I would hit the mark.

library(ggplot2)
library(dplyr)
library(cowplot)

# traffic as of 2017 September 4th
d <- read.csv('https://davetang.org/site_stat/blog_20170904.csv')
d$date    <- as.Date(d$date)
d$day     <- factor(weekdays(d$date), levels = c('Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday'))
d$weekend <- grepl(pattern = "^S", x = d$day)
d$month   <- factor(months(d$date), levels = month.name)
d$quarter <- factor(quarters(d$date))
d$year    <- format(d$date, "%Y")
d$cumsum  <- cumsum(d$views)

group_by(d, year) %>% summarise(views = sum(views))
# A tibble: 5 x 2
   year  views
  <chr>  <int>
1  2013  87415
2  2014 253844
3  2015 283973
4  2016 223065
5  2017 135922

head(cumsum(d$views))
[1] 130 399 657 803 855 908

ggplot(d, aes(x = date, y = cumsum(views))) +
  geom_point() +
  geom_smooth(method='lm')

Not the best fit.

Use only 2017 data to predict.

d_2017 <- d[d$year >= 2017,]
head(d_2017)
           date views       day weekend   month quarter year cumsum
1441 2017-01-01   117    Sunday    TRUE January      Q1 2017 848414
1442 2017-01-02   220    Monday   FALSE January      Q1 2017 848634
1443 2017-01-03   453   Tuesday   FALSE January      Q1 2017 849087
1444 2017-01-04   520 Wednesday   FALSE January      Q1 2017 849607
1445 2017-01-05   549  Thursday   FALSE January      Q1 2017 850156
1446 2017-01-06   382    Friday   FALSE January      Q1 2017 850538

ggplot(d_2017, aes(date, cumsum)) +
  geom_point() +
  geom_smooth(method = 'lm')

Much better fit.

Perform a linear regression.

fit <- lm(cumsum ~ date, d_2017)

# will I hit a million before my 7th anniversary?
predict(fit, data.frame(date = as.Date('2017-10-01')))
      1 
1001525

predict(fit, data.frame(date = as.Date('2017-09-28')))
       1 
999830.9

# not far off the real date
predict(fit, data.frame(date = as.Date('2017-09-29')))
      1 
1000396

I have mentioned the traffic to this site several times before. The main reason I do so is because it's one of the means by which I assess whether this site is still relevant and useful to people. I can only assume that some of the visitors have found the posts useful.

What's new since my last anniversary post? The biggest change is that I have started working with single cell data. In addition, I have been a bit more active with blogging. The period between my 5th and 6th anniversary, I only wrote 13 posts. Between my 6th and 7th anniversary, I wrote 29 posts, which is over two a month, and was my aim! I hope to keep it up.

As usual, let me know if you have found my blog useful. It really does motivate me. Have fun.