Okay, John, I've experimented with a table with wheels (traymobiles, we
used to call them 60 years ago, when I was young). I've done wipes
according to your recipe. What I've found is a replica of the package
dropping on the conveyor belt. The two surfaces make contact and "weld"
together into a static friction situation after a very, very short interval
of time.
Now you and Chuck are right in claimng that the frictional force acting
during that short time interval does work. As far as the "make-and-break
welds" model of frioction goes, the welds formed as contact is made do not
break in this situation. Thus, at those small regions of real contact, the
frictional force changes its point of application and the traymobile (or
package) both experience acceleration and have work done on them by the
frictional force.
If at any time the belt is accelerated, work will again be done on the
package, this time by a static friction force.
So, in some specialised situations (and, John, your tryamobile is sure
specialised) frictional forces can do work but in general they do not.
Tyhe thiung to do is not to claim that they do (as in the standard texkbook
illustration of a block sliding to rest on a table with a rough surface)
but to calclate the chjange in kinetic energy by using the CM equation and
look at energy alterations by applying the full first law of thermodynamic
to a properly chosen system (here the block and the table,
See also Arons "A Guide to Introductory Physics Teaching" page 128 and
Sherwood and Bernard "Work and heat transfer in the presence of static
friction" AJP, 1984, pp1001-1007.