I'm not necessarily demeaning the rest intervals between each exercises; indeed, that is a great way to INCREASE demands (demands being work output per unit time, amount of muscle inroading- instead of intensity, which is defined by most as physical exertion relative to capacity at the moment; in that regard, 100 percent intensity can not be exceeded, so if one were to train at 100 percent intensity it would be obvious that they could subsequently further increase to 101% - that's impossible), although the weight being used will likely be lower.
Perhaps I misspoke. It would make sense to me that all potential variations have value, though relative to the individual, in varying degrees. This means that if you have rest intervals between each exercise, and the sets, that is a good way to increase demands. The same can be said for not using timed rest intervals between each exercise, but exclusively between sets. Either method is probably a good way to avoid biological homeostasis in a system that adapts very quickly and very specifically to stress.
However, for progress to be tracked affectively, there needs to be at least one constant among the variations. This can be anything, but remember that related affects (such as fatigue from timing rest intervals at a rate, e.g. one minute, will influence the amount of weight used) may render it difficult to track progress based solely on certain variables, such as the load and reps performed, unless REST INTERVALS ARE THE SAME EVERY WEEK (a constant). Essentially, the best way to gauge actual progress would be as follows:
Two set training
Reps: 8 reps (3/3 cadence)
Load: 100 pounds
Rest Interval: 1 minute
Reps: 6 reps (3/3 cadence)
Load: 100 pounds
-
Now assume at the next workout you performed the following workout:
Reps: 11 reps (3/3 cadence)
Load: 100 pounds
Rest Interval: 1 minute
Reps: 8 reps (3/3 cadence)
Load: 100 pounds
-
We now can be absolutely certain that progress was made, because results were achieved through identical (at least according to values we can control) means. Now assume that, instead of timing the rest intervals and not tracking cadence (3 seconds up 3 seconds down), this following workout was performed:
Reps: 9 reps (no set cadence, rest pause on last rep)
Load: 100 pounds
Rest Interval: 4 minute
Reps: 8 reps (no set cadence, rest pause on last 2 reps)
Load: 100 pounds
-
Reps: 10 reps (no set cadence, rest pause on last two reps)
Load: 100 pounds
Rest Interval: 8 minute
Reps: 7 reps (no set cadence, rest pause on last rep)
Load: 100 pounds
Clearly, progress potentially has been made, but biomechanical factors (like locking the joints and removing tension) and physical energy replenishment (rest intervals diverse in their quality) might have stunted progress because, while more work was performed, the exercise becomes less demanding. Or, perhaps, no positive or nevative changes occured in the system. The point is it's impossible to tell affirmatively.
Another way to track progress would to have the load, reps and/or TUT constant (as I believe P-Funk does) and improve rest intervals (e.g. improve by 5 seconds every week). The point of this convoluted diatribe is simply that progressive overload isn't exclusively related to the load being moved, time under muscular tension, or repetitions performed; it can also be related to the physical work (force generated to cause displacement) or metabolic work (energy expended) divided by the time.