I was approached yesterday by a friend who asked me if I knew a good way to encode (compress) a string of numbers so that it would be as short as possible, while still perfectly readable by a human.

I think "here and there" is a bit of an understatement. And besides, it doesn't work: >>> b'\xa7'.decode('ascii') Traceback (most recent call last): File "<stdin>", line 1, in <module> ...

The approach used by the demo is to encode each categorical value on the fly. An alternative, which is more efficient but slightly more complicated, is to precompute all possible categorical values, ...

Anyone have some C# sample code or class to decode/encode base64 data? I'm new to C#, and I thought this sort of thing would be built-in to .NET. Just looking for a black box that will take a b64 ...

I wouldn't be so quick to claim that there are never character set or encoding related bugs in Perl or Tcl applications as a direct result of DWIM not always doing the _right_ thing. Perhaps not ...