3 or 4 compression algorithms is alot.
For initial standardization work, we ought to try to agree on one.
Only with VERY strong arguments should we concede to have two.
This does not preclude private usage, but the focus at this stage
of standards work ought to be to develop a BASIC capability with
interoperable open systems compressed data exchange, rather than
to assume or plan for a wide range of such algorithms. It just
makes the implementation work more difficult and makes
interoperability less likely.