Both questions are defective as the site words them: the first does not exclude points or the empty set, and the second reads Erdős's dimension hypothesis as the ambient dimension. The site labels the problem DISPROVED and its commentary says that "the answer to both is in fact no, as shown by Rudin [Ru58] (conditional on the continuum hypothesis)"; its thread has no comments. The corrected Statement follows Erdős's own words, which the curator's credit presupposes for the first question: [Er44], printed p. 445, asks "Is it true that every connected set contains a connected subset not homeomorphic to it? (Points do not count as connected sets.)", p. 443 states the same convention, and [Er82e], p. 77, asks for a subset "which is not a point"; the change inserts "(Points and the empty set do not count as connected sets.)". For the second question [Er44], printed p. 446, asks whether "every connected set of dimension greater than 1 contains $2^c$ connected subsets", and [Er82e], p. 77, whether "every connected set (in a Euclidean space) of dimension greater than one contains more than $c=2^{\aleph_0}$ connected subsets"; the hypothesis is on the set's own dimension, and the change inserts "of dimension greater than one". The site's wording reads that phrase as the ambient dimension $n\ge2$. The curator's verdict follows [Er82e], where Erdős writes that "Mary Ellen Rudin using the continuum hypothesis found the required counter examples" to both questions. For the first question Rudin's theorem gives countable complements, and the paper neither states nor proves that every nondegenerate connected subset of her set is homeomorphic to the set; that step is asserted by [Er82e] and the site and has not been found here. The first defect is already in [Er82e], which excludes points for the subset only; the second is the site's. The $2^c$ form of [Er44] is a variant.