Rant

The Arecibo telescope collapsed on Dec 1 2020, but it died 15 years earlier. Federal funding was cut to the point where maintenance kept getting more expensive. There were some negotiations over Universities taking over the maintenance, but damage kept adding up and the longer the negotiations kept going the more they would have to spend to get it back to fully-repaired.

Scott Manley has said it better than I can, but to summarise - Arecibo was built with the “defend against Soviet nukes” budget, which was more generous than the “advance science” budget it eventually ended up in.

There are bigger radio telescopes, but it could light up its targets in radio. Imagine shining a giant flashlight on Mercury to get a better look, versus looking out into the darkness at Mercury. Progress has gone backwards; NASA doesn’t have its own ships taking people to the ISS, there are no supersonic commercial airliners, and we lost the biggest telescope of its type.

I was reminded of this when Today I found out talked about a pre-Apollo proposal for a military base on the moon. It would have been expensive, but the USA spent ~$25.4bn on the whole Apollo programme, against about $111bn on the USA military in Vietnam. The time period is similar enough that I’m not going to adjust for inflation. Where would we be if we valued research as much as bombs?

Code

In 1974 Arecibo sent the first deliberate transmission to the stars. Radio signals have leaked for as long as we’ve had radio ¹, but Arecibo’s message was like a flashlight instead of a lantern.

As a binary string, it is as follows:²

message <- "00000010101010000000000001010000010100000001001000100010001001011001010101010101010100100100000000000000000000000000000000000001100000000000000000001101000000000000000000011010000000000000000001010100000000000000000011111000000000000000000000000000000001100001110001100001100010000000000000110010000110100011000110000110101111101111101111101111100000000000000000000000000100000000000000000100000000000000000000000000001000000000000000001111110000000000000111110000000000000000000000011000011000011100011000100000001000000000100001101000011000111001101011111011111011111011111000000000000000000000000001000000110000000001000000000001100000000000000010000011000000000011111100000110000001111100000000001100000000000001000000001000000001000001000000110000000100000001100001100000010000000000110001000011000000000000000110011000000000000011000100001100000000011000011000000100000001000000100000000100000100000001100000000100010000000011000000001000100000000010000000100000100000001000000010000000100000000000011000000000110000000011000000000100011101011000000000001000000010000000000000010000011111000000000000100001011101001011011000000100111001001111111011100001110000011011100000000010100000111011001000000101000001111110010000001010000011000000100000110110000000000000000000000000000000000011100000100000000000000111010100010101010101001110000000001010101000000000000000010100000000000000111110000000000000000111111111000000000000111000000011100000000011000000000001100000001101000000000101100000110011000000011001100001000101000001010001000010001001000100100010000000010001010001000000000000100001000010000000000001000000000100000000000000100101000000000001111001111101001111000"

The idea being that an alien might pick up this string and notice its length:

str_length(message)

## [1] 1679

Then they might notice that \(1679 = 23 \times 73\), and that 23 & 73 are prime. If they like 2-dimensional grids as much as us, they might decode it thus:

message_vector <- message %>% 
  str_split("") %>% 
  unlist() %>% 
  as.numeric()

list(x = seq_len(23), y = seq_len(73)) %>% 
  cross_df() %>% 
  mutate(message = message_vector) %>% 
  filter(message == 1) %>% 
  mutate(y = -y) %>% 
  ggplot(aes(x=x,y=y)) + geom_point() + coord_fixed() + theme_void()

Or incorrectly as:

list(x = seq_len(73), y = seq_len(23)) %>% 
  cross_df() %>% 
  mutate(message = message_vector) %>% 
  filter(message == 1) %>% 
  mutate(y = -y) %>% 
  ggplot(aes(x=x,y=y)) + geom_point() + coord_fixed() + theme_void()

Hopefully they find the first one more meaningful than the second!

Postscript

I updated {blogdown}, and the old theme did not play nicely with the updates. Eventually I decided that I like the default theme, and made a couple of tweaks. For example, the site logo is what Big Sleep decided An Operational Researcher looks like.

As with every time I touch the theme, the commit messages on GitHub tell a story of growing despair.