What abut Unicode? Do you allow accented characters and the like? If so, then the standard range character classes are an outright bug (to loosely paraphrase Larry). Use the POSIX character classes for this.

Here is a regex using look ahead assertions and POSIX character classes which meets all of these requirements

$username = <>;
print "Yay!\n" if $username =~ /(?!    #look ahead . . .
                               ^    #and see if at the start of the st
+ring
                               \d+  #we have one or more digits
                               $)   #until the end of the string. If n
+ot . . .
                               ^    #at the start of the string
                   [[:alpha:]]   #allow an alphabetic character
                  [_.-0-9[:alpha:]]   #followed by an alphanumeric cha
+racter, underscore, dot, or dash
                           {0,29}   #0 to 29 times
                                $   #to the end of the string
                              /x;
[download]

Note: That if Unicode character are really not allowed (ie you actually DO want ASCII A-Za-z) then replace [:alpha:] in the example appropriately

Update: Not long after posting this I realized some mild errors if had because of an incomplete reading of jimbobn's original question. $face = red. This has been corrected Cheers,
Erik

Light a man a fire, he's warm for a day. Catch a man on fire, and he's warm for the rest of his life. - Terry Pratchet