Christopher Browne wrote:

> 
> I don't see there being a material benefit to be found in switching over
> to a BINARY format. 

This might be true if the binary data is only small; in case of large 
data and small indexes (something like 100GB TOAST, 20MB index size) 
this is certainly not the case.

  Certainly not when it would lead to a loss of
> portability.

I agree, though for my use-case (only i86) it is acceptable, I used a 
hard-coded slon version. I'd prefer a better COPY format anyway.

Regards,
Andreas