On 11 Aug, Yukihiro Matsumoto wrote: > > In message "[ruby-talk:00613] Re: Bug in Array#clone!" > on 99/08/11, GOTO Kentaro <gotoken / math.sci.hokudai.ac.jp> writes: > [...] > What information should be copied by dup? Any opinion? Sorry, no opinion! But as `clone' and `dup' are grouped together in the manual like: clone dup Returns the copy of the object. For cloned object, ... I have silently assumed, that they are synonymous. A proposal: Whenever methods do the same work with little difference you could group them, and differ them in the description. For example: getc readchar Returns a character read from the receiver. In case of EOF, getc returns nil, and readchar will raise an EOFError. Or so... But functions that do different things like `dup' and `clone', it would be better, IMHO, to separate them in the description and point to each other to show the difference. Ruby has some alias names like `IO#each' and `IO#each_line'. They are described together like `clone' and `dup'. So it is not too difficult to get the impressions, that all names described in that fashion are, more or less, equal. What do you think? > > matz. \cle