```
# remove all varibles
rm (list = ls())
########################################################
########################################################
##
## A couple of introductory comments about basic math
##
########################################################
########################################################
#--------------------------------------------
# how many zeros are there in a power of 10?
#--------------------------------------------
# The number of zeros (before the decimal point) in a
# POSITIVE power of 10 is the same as the exponent
10^1 # 10 (one zero)
```

`[1] 10`

`10^2 # 100 (two zeros)`

`[1] 100`

`10^3 # 1000 (three zeros)`

`[1] 1000`

`10^4 # 10000 (four zeros)`

`[1] 10000`

```
# etc ...
# The number of zeros after the decimal point in a
# NEGATIVE power of 10 is one less than the absolute value of the exponent
10^-1 # 0.1 (no zeros after the decimal point)
```

`[1] 0.1`

`10^-2 # 0.01 (one zero after the decimal point)`

`[1] 0.01`

`10^-3 # 0.001 (two zeros after the decimal point)`

`[1] 0.001`

`10^-4 # 0.0001 (three zeros after the decimal point)`

`[1] 1e-04`

```
# etc ...
#-------------------------------------------------------
# Intro to Scientific Notation in R
#
# Multiplying numbers by powers of 10
#-------------------------------------------------------
# If you multiply a number by a POSITIVE POWER of 10 the decimal point
# will move to the RIGHT by the number of positions as expressed by the exponent.
# Example:
1.234 * 10^0 # 1.234 : 10^0 is 1 so this doesn't change the first number
```

`[1] 1.234`

`1.234 * 10^1 # 12.34 : 10^1 is 10 so this moves the decimal to the right by 1 position`

`[1] 12.34`

`1.234 * 10^2 # 123.4 : 10^2 is 100 so this moves the decimal to the right by 1 position`

`[1] 123.4`

`1.234 * 10^3 # 1234.0`

`[1] 1234`

`1.234 * 10^4 # 12340.0`

`[1] 12340`

```
# etc ...
# If you multiply a number by a NEGATIVE POWER of 10 the decimal point
# will move to the LEFT by the number of positions as expressed by the exponent.
# Example:
123.4 * 10^-1 # 12.34 move decimal point 1 position to the left
```

`[1] 12.34`

`123.4 * 10^-2 # 1.234 move decimal point 2 positions to the left`

`[1] 1.234`

`123.4 * 10^-3 # 0.1234 move decimal point 3 positions to the left`

`[1] 0.1234`

`123.4 * 10^-4 # 0.01234 move decimal point 4 positions to the left`

`[1] 0.01234`

```
# etc ...
# These types of calculations can be used as a shorthand to write very large numbers
# and very small numbers in a very compact way. For example:
#
# 9.234 * 10^100 is shorthand for
# the very very large number 9234000000...(total of 97 zeros)...0000
# i.e. 97 zeros after the 9234
#
# 9.234 * 10^-100 is shorthand for
# the very very small number 0.00000......(99 zeros)...0009234
# i.e. 99 zeros between the decimal point and the 9234
#
# The above calculations are examples of "Scientific notation".
# Scientific notation is used as a shorthand for writing very big numbers
# (and very small numbers - see below)
#---------------------------------------------------------------------
# R has a shorthand notation for writing these types of calculations.
# Instead of writing 1.2345*10^6, you could instead write 1.2345e6
#
# The "e" in the number stands for "exponent". The "e" is understood
# to be read as "times ten to the power of". The number after the "e"
# is the exponent for the power of 10.
# EXAMPLE - all of the following are the same exact number:
#---------------------------------------------------------------------
1234500 # this is the same
```

`[1] 1234500`

`1.2345 * 10^6 # this is the same 1234500`

`[1] 1234500`

`1.2345e6 # this is the same`

`[1] 1234500`

```
# By default R will display values in scientific notation if the number is
# very very big or very small. For example:
# very large number - by default, R shows this value in scientific notation
12300000000 # 1.23e+10
```

`[1] 1.23e+10`

```
# very small number - by default, R shows this value in scientific notation
0.000000123 # 1.23e-07
```

`[1] 1.23e-07`

```
# We can also write scientific notation in our code if we like.
1.23e+9 / 1e+8 # 12.3 , same as 1230000000 / 100000000 but a easier to read :)
```

`[1] 12.3`

`1.2e-4 # 0.00012`

`[1] 0.00012`

`0.0000000000123 == 1.23e-11`

`[1] TRUE`

`0.00000000001231 == 1.23e-11`

`[1] FALSE`

```
# If R returns some very large numbers or some very small numbers in a vector,
# the ALL of the numbers in the vector will be displayed in scientific notation.
# Some typical arithmetic - nothing surprising.
100 / c(100, 20) # 1 5
```

`[1] 1 5`

```
# Same code as above but the last value will be very large so ALL of the
# results are expressed in scientific notation.
100 / c(100, 20) # 1 5
```

`[1] 1 5`

`100 / c(100, 20, 0.001) # 1e+00 5e+00 1e+05`

`[1] 1e+00 5e+00 1e+05`

```
# numbers are all low enough to display without scientific notation
3 ^ (1:5) # 3 9 27 81 243
```

`[1] 3 9 27 81 243`

```
# Some of the larger numbers are displayed in scientific notation so
# R displays ALL of the numbers in scientific notation
3 ^ (1:30) # 3.00e+00 9.00e+00 2.70e+01 8.10e+01 2.43e+02 etc ...
```

```
[1] 3.000000e+00 9.000000e+00 2.700000e+01 8.100000e+01 2.430000e+02
[6] 7.290000e+02 2.187000e+03 6.561000e+03 1.968300e+04 5.904900e+04
[11] 1.771470e+05 5.314410e+05 1.594323e+06 4.782969e+06 1.434891e+07
[16] 4.304672e+07 1.291402e+08 3.874205e+08 1.162261e+09 3.486784e+09
[21] 1.046035e+10 3.138106e+10 9.414318e+10 2.824295e+11 8.472886e+11
[26] 2.541866e+12 7.625597e+12 2.287679e+13 6.863038e+13 2.058911e+14
```

```
#-----------------------------------------------------
# MORAL OF THE STORY - don't become alarmed
#
# Occasionally, you will see R displaying numbers in
# scientific notation. Don't become confused. Understand
# that these are just "regular numbers" being displayed in
# a more concise format. Any math that is done with these
# numbers is the same as if you did the same math with the
# equivalent non-scientific-notation format.
#-----------------------------------------------------
#----------------------------------------------------------
# PRACTICE
#----------------------------------------------------------
# QUESTION - Write the value 4.98e+5 as a "regular" number
# QUESTION - Write the value of 4.98e-3 as a "regular" number
# what are the values of the following expressions?
#
# 1e-2 + 2e-1
#
# 9.876e5
#
# 5.23e4 + 1000
#
#
# What will R display for the following numbers?
#
# 12340000000000 (ten zeros)
#
# 0.0000000000123 (ten zeros)
#########################################################################
# 2022 - BEREN - ONLY COVERED UP TO HERE IN THIS FILE
# YOU ARE NOT RESPONSIBLE FOR THE MATERIAL BELOW THIS POINT IN THIS FILE
#########################################################################
#########################################################################
# 2022 - WILF - ONLY COVERED UP TO HERE IN THIS FILE
# YOU ARE NOT RESPONSIBLE FOR THE MATERIAL BELOW THIS POINT IN THIS FILE
#########################################################################
############################################################
############################################################
#
# Roundoff error in Decimal (base-10) and Binary (base-2)
#
############################################################
############################################################
# The numbers what we humans use every day are made up of 10 different digits i.e 0,1,2 .. 9
# These numbers are therefore called "base-10" numbers or "decimal" numbers.
#
# It may sound strange, but in order to perform many calculations, computers internally use numbers
# that only have 1 and 0 as digits. These numbers are known as "base-2" or "binary" numbers. When computers display numbers to
# people, the numbers are converted to the "base-10" (or "decimal") numbers that we are
# used to seeing.
#
# Right now we will NOT discuss the intricate details of how
# computer work with Binary (base-2) numbers.
# (For more info. you can refer to the powerpoint on Canvas.)
# HOWEVER, you should understand the following ...
#
# The fractions that can be converted to an exact number with a decimal point are
# different for base-2 (aka binary) numbers and for base-10 (aka decimal) numbers.
# For binary numbers (i.e. numbers that computers use internally)
# fractions that can be reduced to have a power of 2 in the denominator
# can be represented by a terminating decimal point value.
# Other fractions cannot.
#
# This issue leads to "roundoff" error very frequently when doing math
# in computer programming- similar to 1/3 + 1/3 + 1/3 ...
#
# R masks this issue by displaying what you might think a number is.
# However, the actual value for the number might be something else.
#
# You can use the print.default (or just print) function with the
# digits argument to display the ACTUAL value that R stores for a number
0.1
```

`[1] 0.1`

`print(0.1, digits=22)`

`[1] 0.1000000000000000055511`

`0.3`

`[1] 0.3`

`print(0.3, digits=22)`

`[1] 0.2999999999999999888978`

`0.6`

`[1] 0.6`

`print(0.6, digits=22)`

`[1] 0.5999999999999999777955`

`1.1`

`[1] 1.1`

`print(1.1, digits=22)`

`[1] 1.100000000000000088818`

```
# This issue can lead to "roundoff" errors. There are ways to deal with
# this but it can get a little involved for this early in the course.
# for right now, just accept it. We'll revisit this issue later.
# It is well known that the fraction 1/3 cannot be directly represented by
# a number with a decimal point. The best we can do is 0.333... You would need
# an infinite number of 3's after the decimal point to accurately represent the
# fraction 1/3. Therefore 0.333 + 0.333 + 0.333 is NOT the same as 1/3 + 1/3 + 1/3.
# Similarly 1/11 is 0.090909... You cannot represent one eleventh exactly using
# a finite number of digits.
# In binary (i.e. base-2) a similar problem exists. However, the numbers that cannot
# be represented exactly are different from those in decimal. For example, if you
# would convert 1/5 to binary, you would not get an exact representation.
# Therefore R will not be able to represent the value of 1/5 exactly
# using standard "numeric" values.
= 1/5
oneFifth # R displays a rounded version of the value oneFifth
```

`[1] 0.2`

`print(1/5 , digits=22) # you can see the actual value stored by specifying the # of digits after the decimal point`

`[1] 0.2000000000000000111022`

`print(1/5 * 1, digits=22)`

`[1] 0.2000000000000000111022`

`print(1/5 * 2, digits=22)`

`[1] 0.4000000000000000222045`

`print(1/5 * 3, digits=22)`

`[1] 0.6000000000000000888178`

`print(1/5 * 4, digits=22)`

`[1] 0.8000000000000000444089`

`print(1/5 * 5, digits=22)`

`[1] 1`

```
# The same is true for 1/10
= 1/10
oneTenth oneTenth
```

`[1] 0.1`

`print(oneTenth, digits=22)`

`[1] 0.1000000000000000055511`

`print(oneTenth * 1 , digits =22)`

`[1] 0.1000000000000000055511`

`print(oneTenth * 2 , digits =22)`

`[1] 0.2000000000000000111022`

`print(oneTenth * 3 , digits =22)`

`[1] 0.3000000000000000444089`

`print(oneTenth * 4 , digits =22)`

`[1] 0.4000000000000000222045`

`print(oneTenth * 5 , digits =22)`

`[1] 0.5`

`print(oneTenth * 6 , digits =22)`

`[1] 0.6000000000000000888178`

`print(oneTenth * 7 , digits =22)`

`[1] 0.7000000000000000666134`

`print(oneTenth * 8 , digits =22)`

`[1] 0.8000000000000000444089`

`print(oneTenth * 9 , digits =22)`

`[1] 0.9000000000000000222045`

`print(oneTenth * 10 , digits =22)`

`[1] 1`

```
######################################################################################
#######################################################################################
#
# EXPLORING R's VARIOUS OPTIONS
#
# R has over 80 different options that you can set. For most of the time we will
# choose to just leave the defaults that are set when R is installed. However,
# sometimes it is helpful to change some of these options.
# options() # see all of R's options and their values
#
# When working with roundoff error issues it is helpful sometimes to automatically
# see more digits after the decimal point when you perform R commands.
# You can do that through R's options. See the following for more info.
#
# options() # see all of R's options and their values
#
# names(options()) # see just the names of the options without their values
#
# getOption("digits") # see the value of the digits option (default for new installation is 7)
#
# options(digits=22) # set the number of max # of digits that R will display for a number
#
#
#######################################################################################
#######################################################################################
options() # see all of R's options and their values
```

```
$add.smooth
[1] TRUE
$browserNLdisabled
[1] FALSE
$callr.condition_handler_cli_message
function (msg)
{
custom_handler <- getOption("cli.default_handler")
if (is.function(custom_handler)) {
custom_handler(msg)
}
else {
cli_server_default(msg)
}
}
<bytecode: 0x000001575a3d7438>
<environment: namespace:cli>
$CBoundsCheck
[1] FALSE
$check.bounds
[1] FALSE
$citation.bibtex.max
[1] 1
$continue
[1] "+ "
$contrasts
unordered ordered
"contr.treatment" "contr.poly"
$defaultPackages
[1] "datasets" "utils" "grDevices" "graphics" "stats" "methods"
$demo.ask
[1] "default"
$deparse.cutoff
[1] 60
$device
function (width = 7, height = 7, ...)
{
grDevices::pdf(NULL, width, height, ...)
}
<bytecode: 0x0000015757dace30>
<environment: namespace:knitr>
$device.ask.default
[1] FALSE
$digits
[1] 7
$echo
[1] FALSE
$editor
[1] "notepad"
$encoding
[1] "native.enc"
$example.ask
[1] "default"
$expressions
[1] 5000
$help.search.types
[1] "vignette" "demo" "help"
$help.try.all.packages
[1] FALSE
$help_type
[1] "html"
$htmltools.preserve.raw
[1] TRUE
$HTTPUserAgent
[1] "R (4.3.1 x86_64-w64-mingw32 x86_64 mingw32)"
$install.packages.compile.from.source
[1] "interactive"
$internet.info
[1] 2
$keep.parse.data
[1] TRUE
$keep.parse.data.pkgs
[1] FALSE
$keep.source
[1] FALSE
$keep.source.pkgs
[1] FALSE
$knitr.in.progress
[1] TRUE
$locatorBell
[1] TRUE
$mailer
[1] "mailto"
$matprod
[1] "default"
$max.contour.segments
[1] 25000
$max.print
[1] 99999
$menu.graphics
[1] TRUE
$na.action
[1] "na.omit"
$nwarnings
[1] 50
$OutDec
[1] "."
$pager
[1] "internal"
$papersize
[1] "letter"
$PCRE_limit_recursion
[1] NA
$PCRE_study
[1] FALSE
$PCRE_use_JIT
[1] TRUE
$pdfviewer
[1] "C:/PROGRA~1/R/R-43~1.1/bin/x64/open.exe"
$pkgType
[1] "both"
$prompt
[1] "> "
$renv.consent
[1] TRUE
$repos
CRAN
"https://packagemanager.posit.co/cran/latest"
$scipen
[1] 0
$show.coef.Pvalues
[1] TRUE
$show.error.messages
[1] TRUE
$show.signif.stars
[1] TRUE
$showErrorCalls
[1] TRUE
$showNCalls
[1] 50
$showWarnCalls
[1] FALSE
$str
$str$strict.width
[1] "no"
$str$digits.d
[1] 3
$str$vec.len
[1] 4
$str$list.len
[1] 99
$str$deparse.lines
NULL
$str$drop.deparse.attr
[1] TRUE
$str$formatNum
function (x, ...)
format(x, trim = TRUE, drop0trailing = TRUE, ...)
<environment: 0x000001575770a7c8>
$str.dendrogram.last
[1] "`"
$tikzMetricsDictionary
[1] "000390-sciNotation_roundoffError_v005-tikzDictionary"
$timeout
[1] 60
$try.outFile
A connection with
description "output"
class "textConnection"
mode "wr"
text "text"
opened "opened"
can read "no"
can write "yes"
$ts.eps
[1] 1e-05
$ts.S.compat
[1] FALSE
$unzip
[1] "internal"
$useFancyQuotes
[1] FALSE
$verbose
[1] FALSE
$warn
[1] 0
$warning.length
[1] 1000
$warnPartialMatchArgs
[1] FALSE
$warnPartialMatchAttr
[1] FALSE
$warnPartialMatchDollar
[1] FALSE
$width
[1] 80
$windowsTimeouts
[1] 100 500
```

`# learn more about the options command ?options `

`starting httpd help server ... done`

`names(options()) # see just the names of the options without their values`

```
[1] "add.smooth"
[2] "browserNLdisabled"
[3] "callr.condition_handler_cli_message"
[4] "CBoundsCheck"
[5] "check.bounds"
[6] "citation.bibtex.max"
[7] "continue"
[8] "contrasts"
[9] "defaultPackages"
[10] "demo.ask"
[11] "deparse.cutoff"
[12] "device"
[13] "device.ask.default"
[14] "digits"
[15] "echo"
[16] "editor"
[17] "encoding"
[18] "example.ask"
[19] "expressions"
[20] "help.search.types"
[21] "help.try.all.packages"
[22] "help_type"
[23] "htmltools.preserve.raw"
[24] "HTTPUserAgent"
[25] "install.packages.compile.from.source"
[26] "internet.info"
[27] "keep.parse.data"
[28] "keep.parse.data.pkgs"
[29] "keep.source"
[30] "keep.source.pkgs"
[31] "knitr.in.progress"
[32] "locatorBell"
[33] "mailer"
[34] "matprod"
[35] "max.contour.segments"
[36] "max.print"
[37] "menu.graphics"
[38] "na.action"
[39] "nwarnings"
[40] "OutDec"
[41] "pager"
[42] "papersize"
[43] "PCRE_limit_recursion"
[44] "PCRE_study"
[45] "PCRE_use_JIT"
[46] "pdfviewer"
[47] "pkgType"
[48] "prompt"
[49] "renv.consent"
[50] "repos"
[51] "scipen"
[52] "show.coef.Pvalues"
[53] "show.error.messages"
[54] "show.signif.stars"
[55] "showErrorCalls"
[56] "showNCalls"
[57] "showWarnCalls"
[58] "str"
[59] "str.dendrogram.last"
[60] "tikzMetricsDictionary"
[61] "timeout"
[62] "try.outFile"
[63] "ts.eps"
[64] "ts.S.compat"
[65] "unzip"
[66] "useFancyQuotes"
[67] "verbose"
[68] "warn"
[69] "warning.length"
[70] "warnPartialMatchArgs"
[71] "warnPartialMatchAttr"
[72] "warnPartialMatchDollar"
[73] "width"
[74] "windowsTimeouts"
```

`getOption("digits") # see the value of the digits option (default for new installation is 7)`

`[1] 7`

```
options(digits=22) # set the number of max # of digits that R will display for a number
# You can see now that these numbers have roundoff error
1/5
```

`[1] 0.2000000000000000111022`

`1/10 `

`[1] 0.1000000000000000055511`

```
# These numbers do not
1/2
```

`[1] 0.5`

`1/4`

`[1] 0.25`

`1/8`

`[1] 0.125`

```
# set digits option back to the default
options(digits=7) # set the number of max # of digits that R will display for a number
# You don't see the roundoff error anymore but it is still there ...
1/5
```

`[1] 0.2`

`1/10 `

`[1] 0.1`

```
# A somewhat more involved explanation ...
#############################################################
#############################################################
##
## Floating point numbers and "roundoff errors"
##
## (A "floating point number" refers to a number that
## contains a decimal point. The term "floating point" is
## a reference to a technique that
# computers use to process numbers with decimal points in them.
## We will not cover exactly why they
## are called "floating point" numbers, but the reason is
## related to the fact that multiplying a number by a power of 10
## simply moves the decimal point.)
##
#############################################################
#############################################################
# Some fractions that we write do not have an exact representation
# as floating point numbers. For example, many people will write
# 1/3 as 0.333, however there really should be an infinite number of 3's in the floating point
# version. 0.333 is just a rough approximation. 0.3333 is closer to 1/3
# and 0.33333 (five 3's) is even closer. However, you will NEVER get
# an exact representation of 1/3 unless you write an infinite number
# of 3's after the decimal point.
#
# Therefore 1/3 + 1/3 + 1/3 = 1
# When converted into floating point numbers becomes, approximately
#
# 0.333 + 0.333 + 0.333 = 0.999 (which looks wrong!)
#
# Some fractions can be converted exactly, eg.
# 1/2 is 0.5 exactly
# 1/4 is 0.25 exactly
# 1/5 is 0.2 exactly
# 1/8 is 0.125 exactly
# 1/10 is 0.1 exactly
# 1/16 is 0.0625 exactly
# 1/25 is 0.04 exactly
#
# In general if a fraction can be reduced to an equivalent fraction whose
# denominator is a power of 2 , a power of 5 or a power of 10, then the
# fraction can be represented by an exact terminating decimal.
#------------------------------------------------------------------
#
# Base-10 (aka Decimal) numbers vs Base-2 (aka Binary) numbers
#
#------------------------------------------------------------------
# Languages such as Hebrew, Arabic and English use different alphabets.
# However, some languages such as French, Spanish and English use the same
# alphabet. However, the words they form often mean different things in the
# different languages. For example the word "pie" in English is a tasty treat
# but "pie" in Spanish means a foot. In order to really understand what the word
# "pie" means you need to know if the word is Spanish or English!!!
# In a similiar way, there are different "numeric languages".
# "Numeric languages", such as "Roman numerals", use different "alphabets".
# For example, the number that we know as one hundred and twenty three, i.e. 123
# is written in Roman numerals as CXXIII.
#
# However, some "numeric languages" use the same digits that we recognize but
# in a different way. The internal workings of computers are designed to use
# a different "numeric language" than the numbers that people are used to seeing.
# Internally, computers process information using a "numeric language" known as
# "base-2" or "binary". For example, the number "one hundred and twenty three", i.e. 123,
# can be written in binary (or base-2) as 01111011. Every base-2 (or binary) number
# is written using just zeros and ones. For example, the binary number
# 1001100100100101 is the equivalent of the decimal (base-10) number
# 39205 (i.e. thirty nine thousand two hundred and five).
#
# When a computer gets a decimal (base-10) number from a peson, the computer first converts
# the decimal (base-10) number into a binary (base-2) number. The computer then
# performs any processing it needs to do. Finally, the result is translated back
# into the decimal form and displayed to the person.
#----------------------------------------------------------------------------------
#
# Decimal (base-10) and binary (base-2) numbers can represent the exact same integers.
#
#----------------------------------------------------------------------------------
# Any integer that we can write in base-10 (or decimal) can also be written, in a different
# form in base-2 (or binary). For example, the following are the decimal and binary
# equivalents for a few numbers:
#
# DECIMAL (base-10) BINARY (base-2)
# 0 0
# 1 1
# 2 10
# 3 11
# 4 100
# 5 101
# 6 110
# 7 111
# 8 1000
#----------------------------------------------------------------------------------
#
# Decimal (base-10) can NOT precisely represent some "real" numbers (i.e. fractions)
# and there will be "roundoff" error.
#
#----------------------------------------------------------------------------------
# In the number system we use every day, some fractions can be perfectly represented
# by "real" numbers (i.e. numbers with decimal points) while others cannot.
# For EXAMPLE:
# 1/2 is EXACTLY 0.5
# 1/4 is EXACTLY 0.25
# 1/5 is EXACTLY 0.2
# 1/8 is EXACTLY 0.124
# 1/10 is EXACTLY 0.1
#
# However, the following fractions CANNOT be represented EXACTLY by a number with a decimal point.
# These fractions lead to "repeating digits after the decimal point"
# EXAMPLE:
# 1/3 is 0.333... (the 3's repeat forever)
# 1/6 is 0.1666... (the 6's repeat forever)
# 1/7 is 0.142857142857... (the 142857's repeat forever)
# 1/9 is 0.111... (the 1's repeat forever)
#----------------------------------------------------------------
#
# Binary (base-2) can NOT represent other "real" numbers
#
#----------------------------------------------------------------
# Binary numbers have similar limitations but for DIFFERENT fractions.
# Any fraction that has a power of 2 in the denominator can be represented
# exactly in binary. However, other fractions cannot.
# For example:
#
# 1/2 is EXACTLY 0.1 (in binary)
# 1/4 is EXACTLY 0.01 (in binary)
# 1/8 is EXACTLY 0.001 (in binary)
# 1/16 is EXACTLY 0.0001 (in binary)
# 3/4 is EXACTLY 0.11 (in binary)
# 3/8 is EXACTLY 0.011 (in binary)
# (any fraction for which the denominator is a power of 2 can be represented exactly in binary)
#
# However:
#
# 1/3 in binary is 0.010101... (the 01's repeat forever)
# 1/5 in binary is 0.00110011... (the 0011's repeat forever)
# 1/6 in binary is 0.0010101... (the 01's repeat forever)
# 1/10 in binary is 0.000110011... (the 0011's repeat forever)
# (fractions whose denomiators are not a power of two cannot be represented exactly in binary)
```