See #1405141

I guess I'm sharing the first part of this opinion:

The choice between OP_CAT, the Great Script Restoration or anything in-between is purely engineering.

To me: OP_CAT alone is bad engineering. We know people will have to do atrocious things to convince Script to do what they want, so we should do better.

but without the followup tweet weakening the assertion by proposing to introduce a whole bunch of opcodes instead of GSR if "GSR is deemed too big".