fermi/README.md

# A minimalist calculator for fermi estimation

This project is a minimalist, stack-based DSL for Fermi estimation. It can multiply and divide scalars, lognormals and beta distributions.

## Motivation

Sometimes, [Squiggle](https://github.com/quantified-uncertainty/squiggle), [simple squiggle](https://git.nunosempere.com/quantified.uncertainty/simple-squiggle) or [squiggle.c](https://git.nunosempere.com/personal/squiggle.c) are still too complicated and un-unix-like.

## Usage

Here is an example

```
$ go run f.go
5000000 12000000
=> 5000000.0 12000000.0
0.002 0.01
=> 13859.5 86583.4
30 180
=> 706832.8 9167656.0
/ 48 52
=> 14139.1 183614.7
/ 5 6
=> 2573.1 33632.0
/ 6 8
=> 368.4 4893.5
/ 60
=> 6.1 81.6
```

Perhaps this example is more understandable with comments and better units:

```
$ sed -u "s|#.*||" | sed -u 's|M|000000|g' | go run f.go
5M 12M # number of people living in Chicago
=> 5000000.0 12000000.0
0.002 0.01 # fraction of people that have a piano
=> 13859.5 86583.4
30 180 # minutes it takes to tune a piano, including travel time
=> 706832.8 9167656.0
/ 48 52 # weeks a year that piano tuners work for
=> 14139.1 183614.7
/ 5 6 # days a week in which piano tuners work
=> 2573.1 33632.0
/ 6 8 # hours a day in which piano tuners work
=> 368.4 4893.5
/ 60 # minutes to an hour
=> 6.1 81.6
# ^ piano tuners in Chicago
```

You can see a recording in action here:

[![asciicast](https://asciinema.org/a/fygBtg0XDc1iVajArdQn9b9CA.svg)](https://asciinema.org/a/fygBtg0XDc1iVajArdQn9b9CA)

## Installation

```
make build
sudo make install
f # rather than the previous go run f.go
```

Why use make instead of the built-in go commands? Because the point of make is to be able to share command-line recipes. 

## Usage together with standard Linux utilities

```bash
f
sed -u "s|#.*||" | sed -u 's|M|000000|g' | f

cat more/piano-tuners.f | f
cat more/piano-tuners-commented.f | sed -u "s|#.*||" | sed -u 's|M|000000|g' | f

tee -a input.log | go run f.go | tee -a output.log
tee -a io.log | go run f.go | tee -a io.log

function f(){
  sed -u "s|#.*||" | 
  sed -u "s|//.*||" | 
  sed -u 's|K|000|g' | 
  sed -u 's|M|000000|g' | 
  sed -u 's|B|000000000|g' | 
  /usr/bin/f
}
```

Note that these sed commands are just hacks, and won't parse e.g., `3.5K` correctly—it will just substitute for 3.5000

## Tips & tricks

Conceptually clearer to have all the multiplications first and then all the divisions

## Roadmap 

- [x] Write README
- [x] Add division?
- [x] Read from file?
- [x] Save to file?
- [x] Allow comments?
  - [x] Use a sed filter? 
- [x] Add show more info version
- [x] Scalar multiplication and division
- [ ] Program into a small device, like a calculator?
- [-] Think of some way of calling bc
- [x] Think how to integrate with squiggle.c to draw samples
  - [x] Copy the time to botec go code
  - [x] Define samplers
  - [x] Call those samplers when operating on distributions that can't be operted on algebraically
- [ ] Think about how to draw a histogram from samples
- [x] Display output more nicely, with K/M/B/T
- [x] Consider the following: make this into a stack-based DSL, with:
  - [x] Variables that can be saved to and then displayed
  - [x] Other types of distributions, particularly beta distributions? => But then this requires moving to bags of samples. It could still be ~instantaneous though.
- [x] Figure out go syntax for
  - Maps
  - Joint types
  - Enums
- [ ] Fix correlation problem, by spinning up a new randomness thing every time some serial computation is done.

Some possible syntax for a more expressive stack-based DSL

```
1B to 20B
* 1 to 100
/ beta 1 2 # or b 1 2
=: x # content of the stack at this point saved into x

1 to 10
10 to 100
=: y # content of the stack at this point saved into y

x # put x on the stack
- y # substract y from the content of the stack. Requires interpreting x and y as list of samples
```