Art conservators worked this out decades ago. When a restorer applies a consolidant deep inside a structurally failing sculpture, the treatment may be irreversible — the material can't be fully extracted without destroying what it saved. Barbara Appelbaum, writing in the Journal of the American Institute for Conservation in 1987, argued that conservators should stop fixating on reversibility and focus instead on re-treatability: when this object needs care again, what options remain open?
The distinction is sharper than it looks. A refunded purchase returns the money but not the delivery window the buyer planned around. A recalled message clears the inbox while the recipient's memory stays full. An application reopened after rejection restores the process; the deadline it depended on expires quietly in the background.
For any system acting on someone's behalf, the useful question is how much of the affected person's future choice a given action consumed. An undo that restores the prior state but not the prior range of options has reversed nothing that mattered.
Reversibility asks: can we return to the prior state?
Re-treatability asks: when someone needs to act again, what options survive?
Where the gap shows up:
- Canceled booking — the slot is released, but the original reservation may already be taken
- Retracted email — message removed; the recipient already read it
- Reversed payment — funds returned, but the merchant's perishable inventory window closed
- Reopened application — process resumed while the consequential deadline has already passed
Why it matters for design: Two actions can both be technically "reversible" yet leave radically different ranges of remedy. What separates them is how many future options each one preserves — for every party affected, not just the one who initiated the undo.