🚀 go-pugleaf

Len Weisberg writes:
 > Oliver Schoett writes:
 > Kyle> Maybe the attribute should be "exposed" instead of "read".
 > Kyle> That's generally what VM means by "read".  It can't really know
 > Kyle> wheyther you've read the message of course.
 > 
 > > That's one of my pet peeves with vm and gnus as compared to rn.  They
 > > mark a message read when it is first exposed.  But this is simply
 > > wrong, as I have not yet read it (for example, I might get
 > > interrupted).  I prefer the approach of rn, where an article is marked
 > > 'read' upon leaving it with a command whose meaning is "I've read this
 > > one, go to the next one".
 > 
 > Hear, Hear!  Thanks for not giving up at the phrase "of course" .
 > So there is a natural way of telling vm that a message has been read.

Well, this is the way I originally implemented it, but I had
lotsa trouble deciding which commands should mark a message as
read and in what contexts.

Marking a message as read after it's been exposed seems natural
to me.  But perhaps VM is being too presumptious to even hope to
get this right.  I hate programs that are too smart for their own
good.  Maybe VM should lose this feature altogether.