Text 3767, 43 rader
Skriven 2019-01-31 19:20:24 av WKITTY42 (1:124/5013)
     Kommentar till en text av PSI-JACK
Ärende: Re: Message Pointers on n
=================================
On 09/07/14, Psi-Jack said the following...
 
 Ps> One more quirk I can suggest to iron out.
 Ps> 
 Ps> When a user posts a new message, and their message read pointer for that
 Ps> message base is already at the last message before their own new post, it
 Ps> would be very nice if Mystic would automatically adjust the message
 Ps> pointer to mark the user's own new post as read, because, they literally
 Ps> just wrote it, but newscan shows it.

i'm between the options on this... on the one hand, it kinda makes sense but
what happens if new messages have been tossed into the message area while the
user was writing their message? this is why JAM (not sure about SQUish at the
moment) has two read pointers... with only one pointer (lastread), the user
could miss messages if it was set to be equal to the post they just made...
with two pointers, lastread and highread, then maybe setting highread to their
post might be OK but i err on the other side and say not, not really... just
because they wrote it doesn't mean that they have read up to that point...

the deal with two pointers is that while reading messages, one might jump up
to the next post in a thread (mystic doesn't yet support this) and then return
to where they were to continue reading... their highread would point to the
highest message number they have read but the lastread pointer shows where
they are in the list of posts... i might read a message and then jump on up to
the next reply to that post without reading the other in between... when i'm
done, i can return to the post where i jumped from and go on from there... if
i drop my connection after the jump and before i return back, then yes, the
lastread and highread would both point to the same message and i would "loose"
those i skipped unless i specifically went back hunting for them...

 Ps> I don't know, that's just my own thought. I think opinions on others on
 Ps> this might be good to get to see if they feel the same. 

when (if?) the JAM (and SQUish) capabilities are fully implemented, things
should be clearer and easier to follow...

disclaimer: it has been over a decade since i used and tested SQUish bases...
the main software i run was the ones that created JAM bases so i'm much more
familiar with them at this point in time but there's still some aspects of
them that i've forgotten...

--- Mystic BBS v1.10 A51 (Windows)
 * Origin:  (46:1/132)