RE: [humanmarkup-comment] Base Schema-channel

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

The channel shouldn't know about time.  That is chronemic information. 
A simulation (if I understand you) defines a chronemic for a virtual
time with a begin and end value, or at least a begin value (given 
that the end time might be indeterminate in some cases).

Channel is messy.  At the barest, it is just a connector for 
routing events into (sensory).   The problem here is getting events out. 
So Rex is considering naming sensory and communication channels. 
That seems to be overkill if the only semantic the base needs 
is in and out to indicate the directionality of the message/event. 
If sensory, we know we have five to six types.   For communication, 
it is tough because we can have lots of these.   Is that a problem?

len

-----Original Message-----
From: Rob Nixon [mailto:rnixon@qdyn.com]

The only thing I'd add to this thread is the need for something along the following
lines.

We are running a series of if-then simulation scenarios.  In one case, the
simulation carries through from it's previous "restart" with information that it has
accumulated from it's previous runs.  (i.e. the 1990 simulation has restarted 3
times, so each virtual character would have 3 mappings from virtual Jan. 1st to
"Human World"-real time.   Information and communication would be tied to a dated
list :
sim1-01/01/1990:17:50:33->06/04/2002:06:23:05
sim2-01/01/1990:17:50:33->06/05/2002:12:43:04
sim3-01/01/1990:17:50:33->06/06/2002:14:06:39

The catch here is that information and communications could be carried through from
the previous iteration and we'd need to be able to roll back to the state of the
vchar at any given point in its iterations but still be able to stamp the
communications with a local virtual (or simulation) time.

I'm not entirely sure how much of a problem this would be under the current
"channel".