This has probably been discussed many times before but I want to resurrect it to get your thoughts. 

I have seen a considerable difference in the life of the T-12 lamps vs the T-8s and wanted to get your opinion as to why you think it is?   Obviously we're only getting half if even that much of the life from the T-8s lamps as compared to the T-12s.  It is most noticeable in the T-8s that are on and off more often or on for shorter periods of time.  I see better life out of the T-8s when they are left on for 8+ hours a day but even still, we would get more hours out of the T-12s they replaced and the T-12s were more forgiving to frequent on/off cycles and short usage.  Is it the lamps because of the lower mercury content, or is it the ballasts, or both? 

I see the issue with Philips, GE, and Sylvania.  Most noteably with Sylvania and Philips.  The older gen T-8s from Philips seem to last better and longer then the new generation of T-8s from Philips- which have been very disappointing in their reliability.  GEs early generation T-8s were crap but seem to have gotten much better as time has gone on with their product.  Sylvania has been in the middle and is generally what i am using, though without doubt their T-12s were the best compared to their counterparts from GE and Philips. 

The T8s are (generally) instant start via HF.  They do NOT like frequent switching.  Started once a day or on 24/7 they will last longer than a T12. 

American electronic T8 ballasts, operates their lamps, exacly like SOX gear operates LPS lamps: Starts the lamp with cold cathode with the electrode heated to thermionic emission, by the discharge, meaning that T8 shouldn't be started more than one time a day (Conditions where streetlighting and security lighting are the only applications that holds this fact).
Dimming isn't possible, since there is no electrodes heating supply, so if the lamp is underdriven, it operates with cold cathodes and sputters very fast.
As there are no reasons to not make rapid-start or programmed start ballasts for your T8s, I see the american T8s and the T12 ban in the US, done only for planned obsolescense of the fluorescent lamp, in order to make fluorescent lighting lasting shorter than they used to be, in order to promote LED lighting.

I never thought of it that way, but I agree 100%.

Is it the lamps because of the lower mercury content, or is it the ballasts, or both? 

I see the issue with Philips, GE, and Sylvania.  Most noteably with Sylvania and Philips.  The older gen T-8s from Philips seem to last better and longer then the new generation of T-8s from Philips- which have been very disappointing in their reliability.  GEs early generation T-8s were crap but seem to have gotten much better as time has gone on with their product.  Sylvania has been in the middle and is generally what i am using, though without doubt their T-12s were the best compared to their counterparts from GE and Philips. 

It is both the ballasts, as well as the lamps.
The standard T8 ballasts are instant start, so cold start the lamps, so causes a lot of sputtering.
The T12 had one advantage: Their ignition voltage went so low when the cathodes reached thermionic emission (compare to cold electrodes), than the simple RS circuit ensured the lamp ignited only after the electrodes warm up.
The problem with T8 is, the ignition voltage difference between cold and hot electrodes is too low, so the RS circuit would either not ignite the lamp at all, or ignine it prematurely (with too cold electrodes). So practical ballast would ignite the lamps nearly all the time with cold electrodes anyway, so with simpler ballasts it became easier to just plain cold start it and do not complicate the design by electrode heating. So viola, we have the known cold starting T8 ballast.
So the only ballast concept starting only with hot electrodes what remain is the real programmed start. But that is so complex, it become the weakest point in the system.

And for the lamp design: Itis possibleto make lamps with better tolerance towards the cold starts, but such lamp would be more complex and so expensive. Try to compare, how long you have to work  today to buy one T8 lamp and how long you parents had to work to get single T12 tube 30years ago...
And on top of that come the legal incentive for the extreme low mercury dose lamps, what make the unreliable and short living Alto so popular, regardless what engineers and other technically aware people think about the concept. Philips may well know it is crap, but the government made the rules so, these become the nearly most popular type...

What about european energy saver T8s on electronic ballasts? The american T8 have way lower starting voltage than european T8, because of different buffer gas. The fact that their first magnetic ballast were rapidstart, proves this.
With cheap electronic ballast for 18W T8s (Like the one inside the fixture in my father sleeping room), I think that it is still an instant rapidstart like the one for the T5s, which supply heating to the electrodes despite instant starting. I don't see so massively sputtering on the ends of the tube.

What about european energy saver T8s on electronic ballasts? The american T8 have way lower starting voltage than european T8, because of different buffer gas. The fact that their first magnetic ballast were rapidstart, proves this.
With cheap electronic ballast for 18W T8s (Like the one inside the fixture in my father sleeping room), I think that it is still an instant rapidstart like the one for the T5s, which supply heating to the electrodes despite instant starting. I don't see so massively sputtering on the ends of the tube.

What help here is the electrode heat applied during the cold cathode phase - it speed up the cathode heatingm, so shorten the time the cathode is exposed to the glow discharge mode.
Indeed, this concept prolong the lamp cycle life considerably, but still the effect is not that good as with the T12 on a RS.

The US T8 have indeed lower ignition voltage than the European, but for both the difference between cold vs hot ignition is quite small.

With the RS, the preheating was implemented primarily just to allow to start the lamp with lower voltage, nobody cared about cycle life at all, as the fluorescent were not expected to become used on places with frequent switching.
As the T12 lower their ignition voltage when it's electrodes are heated, the extra winding and connection wires was justified by the saving of ballast cost, weight and efficiency tied to the required lower OCV.
Taking the same objective as it was with a RS ballasts in the 60's (start and operate the lamps as cheaply as possible) to the T8 characteristics, you will conclude, than as heating the electrodes make no significant reduction of the required voltage for start, it is just easier to design the ballast with a nit higher OCV and get rid of the extra wires (and an associated risk, than someone wire it incorrectly).

The European style with the electrode heat during glow phase came only later and it have one significant disadvantage: It is not fool-proof, it just can be too easily wired incorrectly.
The US IS concept with just single wire per lamp end is just way easier to wire, so less likely becomes a victim of wrong wiring...
(I just had this experience today - even something such simple was wired wrong by a "DIY")...

So instant start T-8 Ballasts are a no no, what should I be looking for in a T8 ballast if I know the lights are going to be turned on for a few min or even 10 seconds and then off, in cases where we are only turning on the lights throughout the week just to check on security in that space and then for 10 hours on Sunday it gets used without interruption. 

Then I have the issue where the classrooms in the building all have motion detector switches that are timed to 5 minutes to go off if inactivity.. Do they make rapid start T-8 ballasts? 

So instant start T-8 Ballasts are a no no, what should I be looking for in a T8 ballast if I know the lights are going to be turned on for a few min or even 10 seconds and then off, in cases where we are only turning on the lights throughout the week just to check on security in that space and then for 10 hours on Sunday it gets used without interruption. 
Then I have the issue where the classrooms in the building all have motion detector switches that are timed to 5 minutes to go off if inactivity..

For that are intended the programmed start ballasts: They first preheat the electrodes with low enough voltage across the tube, so the discharge can not form. After the defined preheat time the ballast increase the voltage to a level, where it start reliably (with high margin). As the voltage for ignition is high, it start the lamp in any state. But as this voltage is applied only with delay for the electrode heating, the filaments are already hot, so no damage on them.

But even with an instant start, checking the light once per day for few seconds wear the lamp less, than keeping them ON for each complete night, so per calender they will last the same or longer.

Do they make rapid start T-8 ballasts? 

Even when you find some, they won't be effective in protecting the cathodes. The problem is the too small difference in the ignition voltage between lamp with hot vs cold cathodes. The difference is way smaller than the tolerances caused either by manufacturing spread (mainly among manufacturers), aging, temperature,...
So the lamps would most frequently start with cold cathodes (like on an instant ballast) anyway, so get the electrode damage...

Even for the F40, the RS was far from optimal in the means of preserving the cathodes on startup, the concept was created not for frequent switching, but for low cost ballast (per given efficiency requirement).

Thank you again!  I will look into the programmed start ballasts. 

Ok next question:  What are the ramifications of running T12 lamps in T8 ballast fixtures?  I've seen it done, but wanted your opinions.  I know first hand what happens when you try to run a T8 lamp in a T12 ballasted fixture, it will generally burn the ballast up quickly.  Learned that in my young days  

So I guess you are asking about US F32T8 vs F40T12 right? You haven't published your location...
It is the question about the lamp arc voltage and nominal current difference:
F40T12 is rated for 0.43A and have nominal voltage of ~105V
The F32T8 is rated for around 0.25A, with arc voltage about 140V.

The damage of both lamps and ballasts could happen in both ways:
Running F32T8 lamp on a F40T12 ballast with really constant current characteristics (the HPF RS units) mean the lamp get higher power than the 40W (up to 55W) and the ballast primary can not handle that (it mean nearly twice the power dissipation in the primary).
Moreover the lamp is overpowered, what cause it's life to shorten as well...

The NPF ballasts should be OK, as there the currents are getting lower.

In the reverse, some ballasts may run on higher output current (the NPF, many 120V only fixed frequency electronic,...), what can overheat these too. But given the OCV requirement, this difference won't be as large.
biggest problem would be with the lamp: The F32T8 ballasts do not provide heat power during normal operation, while the F40 would be highly underdriven (~0.3A instead of 0.43A), what mean the cathodes will operate too cold.

once again, thanks for the info! 

Do they make rapid start T-8 ballasts? 

Not anymore, but yes they did... startup is just like the F40T12 version. I have a few of them myself

   What are the ramifications of running T12 lamps in T8 ballast fixtures

They will work, but the lamps will be underdriven (about like those LPF residential-grade shoplights do). Absolutely no harm will come to the ballast (infact it will run cooler than with F32T8's).

I know first hand what happens when you try to run a T8 lamp in a T12 ballasted fixture, it will generally burn the ballast up quickly.

I've done that, it didn't hurt the ballast (but only 1 of 2 F40T12's in a 2-lamper changed to F32T8), the F32 was overdriven, but since I was using it as a light for houseplants, I left it that way...


edit:
oh and just noticed something, so:
Welcome fellow Colorado LG member!

Medved brings up some good points. I have two Advance magnetic ballasts designed for F32T8 lamps. They are rapid start, just like the popular Mark III T12 ballasts that used to be found everywhere. In cool temperatures, they start lamps exactly like the T12 ballasts, with a dim glow and a bit of flicker for a second and then a very quick ramp up to full intensity.

However, at room temperature (68F/20C) or higher, the startup is almost instant. I tried various combinations of lamps and ultimately, once the lamps were broken in, this ended up being always the case. This would be explained but what Medved wrote. So, I can't see that this provides any advantage to lamp life over a standard instant-start electronic ballast. In fact, there was a study done that shows the rapid-start T8 magnetic ballast actually demonstrates shorter lamp life than an instant-start electronic ballast in frequently-switched applications!

The one advantage would be ballast life: the magnetic coil & core is less affected by high temperatures, and contains fewer components than an electronic ballast.

I have also seen rapid-start electronic ballasts - these I think have temperature-controlled (PTC) preheat, and a discharge (dim glow) forms along 1/3 to the whole length the tube at the same time, or slightly after the preheating starts, for usually about a second, and then the tube ignites fully. When the lamp is hot, though, it starts instantly.