| The encrypt/decrypt set is obviously only one of a large set of do/undo pairs and, rather than getting too bound in the messinesses of a single case or so, we might look for a general solution. None of the obvious seem quite right: {tol} has been examined already, {na'e} gets the wrong results too often, {fa'e} regverses orders but in fact the order of steps in the process is often not reversed but the reversal comes in a different place (the steps in coding and decoding are often the same, just the nature of the input differs -- though, of course, there are many cases where the two processes are quite different, but still not reversals), and {dut} is almost the same as {tol} at least in definition. So we need to rethink what is involved here and work more on it. |