1
$\begingroup$

Let $C$ be the standard middle thirds Cantor set in the interval $[0,1]$. The "endpoints" of $C$ have very simple numerical values that can be listed off: $$0,1,1/3,2/3,1/9,2/9,6/9,7/9,1/27,2/27,7/27,8/27,... $$

What I am looking for is some numerical values of the "nonendpoints" of $C$. Specifically, is there a dense subset of the nonendpoints which has some nice pattern to its values? Can you list off some of these for me?

$\endgroup$
3
  • 3
    $\begingroup$ There's a pretty simple characterisation of the Cantor set in terms of base-$3$ expansions. $x\in C$ if and only of $x$ has a base-$3$ expansion in which no $1$ occurs. $\endgroup$ May 5, 2016 at 18:52
  • $\begingroup$ @DanielFischer ok, but could you list a dense set of these for me? I need to make a table $\endgroup$ May 5, 2016 at 18:57
  • $\begingroup$ Cantor set has empty interior, so it has no "non-endpoints". $\endgroup$
    – jdods
    May 5, 2016 at 21:39

1 Answer 1

2
$\begingroup$

For example,

$$\frac 14 = \frac{0}{3} + \frac{2}{3^2} + \frac{0}{3^3} + \frac{2}{3^4} + \frac{0}{3^5} + \frac{2}{3^6} + \cdots$$

is in the Cantor Set.

Edit: Actually, elements of the Cantor Set are of the form

$$\sum_{n=1}^\infty \frac{a_n}{3^n}$$

where $a_n$ is any sequence consisting of only $0$ and $2$'s.

So, when $a_n$ is not almost constant (an almost constant sequence is a sequence which is constant from some point on) then the corresponding number is a non-endpoint.

Edit2:

$$\frac{2}{25} = \frac{0}{3} + \frac{0}{3^2} + \frac{2}{3^3} + \frac{0}{3^4} + \frac{0}{3^5} + \frac{2}{3^6} + \cdots$$

$$\frac{11}{12} = \frac{2}{3} + \frac{2}{3^2} + \frac{0}{3^3} + \frac{2}{3^4} + \frac{0}{3^5} + \frac{2}{3^6} + \cdots$$

For further examples, you can take:

$a_n = 0,0,0,2,0,2,0,2,0,2,0,2,0, \dots$

$a_n = 2,0,2,0,2,2,0,2,0,2,0,2,0, \dots$

$a_n = 0,2,2,0,0,2,2,0,0,2,2,0,0, \dots$

$a_n = 0,0,0,0,2,2,2,2,0,0,0,0,2,2,2,2, \dots$

$a_n = 0,0,0,2,2,0,0,0,2,2,0,0,0,2,2, \dots$

etc.

$\endgroup$
3
  • $\begingroup$ interesting. so I'm thinking we could we just change finite numbers of terms at the beginning and somehow end up with a dense set of nonendpoints... $\endgroup$ May 5, 2016 at 19:14
  • $\begingroup$ i understand what you are saying in your edit, but I really want numerical values! $\endgroup$ May 5, 2016 at 19:19
  • $\begingroup$ Truncate the series above for some sequences of 2's and 0's. $\endgroup$
    – jdods
    May 5, 2016 at 19:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .