PRINT does not ignore UNSET values.

[hostilefork]

Red, like Rebol, will silently squelch unsets in COMPOSE (for example):

red> compose [10 (print "Hello") 20]
Hello
== [10 20]

In Rebol, this applies to PRINT as well. In Red, the word "unset" appears:

red> print [10 (print "Hello") 20]
Hello
10 unset 20

I believe the preferable output in that case would be Rebol's output:

Hello
10 20

(I know this seems nitpicky but I am trying to build a new philosophy of PRINT, and you have asked for complete proposals, so it's easiest to start with the "hopefully this is an agreement" parts in building that.)

[dockimbel]

Agreed. The difference comes from the way FORM (which is used internally by PRINT) processes unset! values. I see no reason not to have the same behavior as Rebol in this case.

[hiiamboris]

I think this has been fixed in Red 0.6.2. form () now produces "", consistent with R3.