06 Aug '03 02:30>
Originally posted by iamatigerThe flaw is that your SSS does not map to the reals, I'm afraid. Reals do not necessarily have some number N of digits, but rather an actual denumeable infinity of digits. It may be clear to try and work out for yourself the proof of the following:
That seems to me to be an assertion of some fact rather than an identification of a flaw in my argument. I'll try to put my argument more clearly.
If a set S(N) is the set of all combinations of N digits (from 0 to 10), excluding combinations with trailing zeros, then, if the allowed values of N are the natural numbers SS, which is the set of all S(N) ...[text shortened]... e reals). However SSS has been defined to be the Union of a countable number of countable sets.
"Any countable union of countable sets is countable, or equivalently, no uncountable set is a countable union of countable sets."
If you just start going ahead in a general way, the specifics of why your above argument fails can be made apparent. I'd be glad to look at abything you say here...this is quite an interesting thing š.
The reals are the union of an uncountable number of countable sets, or the union of a countable number of uncountable sets. They are the union of all possible subsets of the naturals (first case), or equivalently the union of all continuous intervals.