[no to] leading zeros / Quasipalindromes?

[Frank Ellermann]

Marc LeBrun wrote:

> anagrams (amongst hundreds of others) just for "encyclopedia":
>    Nice code play
>    Open delicacy
[...]
>    Do an epicycle
>    Dope in a cycle
[...]

Now this cries for a "fini,word" sequence, where EIS offers
some anagrams of itself, encoding A = 1, B = 2, etc., 99 = end,
and 0 = space... ;-)

Your other ideas:  I like the current set of keywords, Neil has
eliminated the less useful "done", "look", "huge", and "part".

In some sequences "base" is missing, most of this has to be
addressed manually.  Bits of a constant could be tagged with
"base,cons", but in that case I won't insist on a "base" (and
not bother Neil with a COMMENT).  It starts to get weird for
programs if the base for a "cons" is greater than 10, and then
"base,cons" would be really useful.

Displaying a "cons" 13,14,10,13 as 13141013 would be too weird,
if that's only a joke D,E,A,D with base 16 digits represented
in decimal notation...

...in that special case "base,cons,word" could be used.  OTOH
most constants are shown as decimal expansion, the binary case
is rare, continued fractions have their own keyword, and other
forms are very unusual (at least for constants).

One famous case is to abuse base 3 digits 1 and 2 instead of
binary digits 1 and 0 for binary strings with leading zeros -
in my terminology these are strings and no numbers.  In the
same terminology your anagrams, Amarnath's quasi-palindromes,
or Patricks real palindromes are all strings, requiring "base".

                             Bye, Frank