Originally Posted by HowlerKarma
IF this is "necessary" as a bridging thing-- you know, to scaffold computation of differing place values-- then I'd advocate making the "0" a subscript zero.

That tallies nicely with how scientists tend to carry non-significant figures through calculations-- and we do-- in order to avoid rounding errors.

At the same time, however, it is still important to be able to evaluate-- at any point in the process, at any point in the calculation, at any point now or in the future-- where there is a question of experimental error, and how large the error might be.

Okay, this makes sense and bundles some important ideas.