🚀 go-pugleaf

In article <1992Dec31.180958.22688@chance.gts.org> john@chance.gts.org (John R MacMillan) writes:
>| Well, the historical reason this wasn't necessary/desirable/done is
>|called finger.
>
>[...]
>
>|  I regularly finger people, then look to see when my message was sent.
>|Result? I know if it was read.
>
>Actually, finger has the same problem as many existing read receipts,
>and return receipts: incorrect user expectation.
>
>Here Dave, a knowledgeable computer professional, arrives at a result
>that is simply not true, because fingerd does not (and can not) know
>if I have read my mail.

And here's one possible reason why finger may not be reliable: depending on
your backup method, the atime on spool files may be munged, and fingerd
will think the file has been "read".  True it has, but not by a user.  :-)

The latest version of GNU finger lets you customize what long output the
fingerer sees on a per-user basis.  So for me I've programmed it to look at
the mtime on my personal inbox.  As a matter of habit, when I incorporate
new mail and save my inbox I usually read it right then and there.

However, this still isn't a 100% reliable method.  Really, the best way to
find out if someone has read their mail is to ask in the mail message
"please send me a reply when you've seen this" (and if they choose not to
do this that's their business).  All this nonsense about UA automated
read-receipt replies is a waste of time.  People seem to think computers
are capable of solving all sorts of problems that they're not, particularly
when human capriciousness comes into play.