Just to be technically accurate, the limit of 0.999... stands at 1.
Not quite. The series represented by 0.999... does
converge to a limit of 1, but that limit then is the value of the series. It is not simply something 0.999... gets close to; it is its value.
As a matter of definition, the value of a series is the limit of the corresponding sequence of partial sums.
L = Sum(k=1 to infinity) a
k <==> L = lim(N -> infinity) Sum(k=1 to N) a
k
0.999... is identically 1 and provably so.
A series can be thought of
informally as a summation over infinitely many terms, in fact the sigma notation supports that conceptual view, but it runs into trouble because the informal model comes with a process concept wherein a partial summation gets closer and closer, but never quite reaches its final value. You cannot get to infinity by counting, so the idea of completing the summation conflicts with the apparent step-by-step process.
The formal definition lacks this defect and maintains full consistency.
Doron is mired in a process concept. He muddies the water more with his just use a different infinity approach, too. There are only just so many rational numbers possibly involved in the process view of 0.999.... Giving special consideration to aleph
9 (for example) doesn't change that.