Just my 2 cents.
It seems like someone is reinventing the wheel, One has to wonder who in their LIMITED KNOWLEDGE ever was able to convince municipalities to go to an 800 system with poor wave propagation, knowing that the 800mhz signal does not penetrate foliage well, does not penetrate buildings well, and has a very limited line of site range, has the potential to be interfered with by cellular telephone, other business communications in the 800 band and now most television stations are broadcast in guess what, the 800 band.
Now they want to put everything on 700 mhz band which is suseptable to white noise problems, why not just put public safety on the 500 band, better wave propagation, it can be trunked, analog or digital, p25, it would all work, fewer towers needed.
As for PSP, as a communications design engineer I would have suggested giving them more vhf channels, trunking the vhf system to make efficient use of it and creating a group with crossband switching to enable communications with users on the 800 band for interoperability.
They already knew that the vhf system worked, why change it?
The failsoft mode should have been set up to fail back to the original vhf simplex channels that the barracks use, this way there is always a path of communication even in a major controller failure.
Well, you're not far off, but there are some major issues with building wide-area systems that have capacity requirements. TV still can be on channels 2-13 (54-216 MHz), but most of it has been moved to a UHF "digital home" that's somewhere between 512 to 680 MHz. Land mobile has "T-Band" allotments in 11 markets (based on 70's Census data) and was extended and waivered into various places since. All of those were into areas which already qualified for T-Band channels, but had nothing available. TV doesn't operate on 800 anymore. But TV is a powerful lobby to be reckoned with. The broadcasters have a very powerful (pun intended) lobby with Congress and sometimes make the statement that public safety doesn't make good use of whatever spectrum it does have ("wireless" says the same thing).
I can think of several reasons to build out statewide on 800 rather than VHF. First, their NPSPAC RPCs have made the provision for channels allotted to "statewide" use. 700 MHz had a statewide block automatically, but every state has the same block, so usage has to be coordinated between states. VHF, not so much. You have what you have, and can gain more channels with concurrences and with offering existing systems a "ride" on the new system, but if they don't agree, you can't use the frequency; trunking requires exclusivity (and trunking IS more efficient than conventional). A lesser number of sites because of propagation characteristics was absolutely the motivator for some statewide VHF trunking projects further west. But PA has heavy VHF usage on at least two of her boundaries. NJ and NYS would be heavily affected by prospective PA trunking and mitigation measures (special antennas restricting radiation patterns, site placement looking at interference avoidance rather than coverage) might cost as much as deploying a higher-frequency system. The propagation characteristics of higher frequencies make it a little easier to reuse the frequencies.
We shouldn't confuse digital with trunking or 700/800. The biggest advantage of digital communication is to allow multiple conversations to be squeezed into the space that one analog signal occupied, effectively giving the user two or more channels where they only had one. Or, creating more open channels by using better filtering techniques and allowing closer re-use. Yes, it sucks for scanning, because you need a special scanner - or the format can't be received on a scanner (the more we want to move toward a standard, the further we seem to get).
Ham linked systems are nifty but impractical from a frequency usage standpoint (I am a ham, btw). The same conversation can tie up dozens of channel pairs over thousands of square miles. It's great in concept, but would be so much better IF it were simulcast on one output frequency and used a voting comparator for the input. Linked system with a different channel at each site = meh. Linked system with the same input and output frequency over tens of thousands of square miles = impressive. The Garden State Parkway 6 meter system that Glenn, WB2MAZ, ran (covered from Cape May to Rockland County using only two hop-scotching output frequencies to avoid heterodyne in the overlap areas and one continuously voted input frequency) was impressive.
The other thing that should be looked at is network connectivity. One of the hidden expenses I've noticed in PSP's systems are its microwave backbone. Regardless of whether the system is VHF, UHF, or 800, the system needs to be connected somehow. This kind of thing can carry voice, data, and video, so it's a pretty good investment - and it doesn't care if the radio tied to it is on 852 and digital or on 33.86 and analog. In other words, they needed this system whether it was VHF P25 or 800 OpenSky. Looking at very large systems, this network could be the weakest link, especially if there isn't diversity in how the system is configured (you would HAVE TO spend more to make things like self-healing loops and divergent routing - and that's more money, but much improved resiliency). The advantage is that this part could reach a break-even point where the expense is less than leasing control circuits (disaggregated one line here, one line there with mileage costs) from the telephone company.
As for failing to VHF (or relying on 800 in some places and VHF in others), newer multiband radio equipment has something similar in capability. NY's SWN was going to do a mixed 800/VHF deployment, but it never came to fruition. Look to Missouri for doing something just like this with 800 MHz in the St. Louis and Kansas City areas and VHF throughout the interior - but those VHF channels are still very hard to come by.
As for a lot of talkgroups - think of them as virtual channels that can be had without putting in any new radio infrastructure. Say I manage a radio system for a county. Animal control needs a radio system. Without trunking, I have to buy a repeater, put in receivers, maybe a second repeater because you can't cover everywhere with one site, and "take" two frequencies (yeah, right) - which go dead after 1700 hrs until the next morning. With trunking, they buy subscriber units, I sit down at the computer, turn on their talkgroup, set their priority relative to other uses and they're good - without ever needing more frequencies or putting in more sites. From a systems management perspective, this is a very good thing.
