Hi Jerry,
I use D-STAR Digital Data on 1.2 GHz for my Winlink RMS Packet CMS connection, 24/7. The Pactor 3 mode that I use for Winlink HF forwarding is proprietary. But nobody else has come up with anything near as good (yet). So a lot of hams use it because it does the job well.
I think that was Rob's point. There is no opportunity to learn from it let alone improve upon it because the protocol is at the very least a guarded trade secret if not patented (FWIW, the case of protocols used for HF email mostly outside ham bands I would think there would be very good business reasons for the manufacturer to keep it so). It closes off the doors of learning from it let alone finding ways to improve upon it, and relegates us to nothing more than a consumers of a product. One might argue that ingraining it in a permanent network, whether it be HF email or a shared satellite channel, does more of the same. Having a consumer base that is happy with what they have is pretty good discouragement for incremental innovation....after all, all these consumers have just spent all this money on their radios and modems, why would they want to change or even support it? In the end you may find that "just because it works" sounds specious to people who believe that Amateur Radio Service has had much more to offer than simply transferring information from point A to point B with "maximum efficiency" or whatever. There's a philosophical pragmatism behind this that some people see as having the potential to trump the balance of purposes that define the Amateur Radio Service. I think that's why Rob pointed to the enumerated purposes in CFR part 97.1. (For those outside the US you probably have a similar wording in your countries' laws since this all pretty much summarizes the ITU definition, so this isn't just a "US thing").
(As a side note for the spectators, if you have strong feelings about this you may want think about the long-term consequences the ARRL's upcoming NPRM as it relates to HF digital communication).
Back to satellites. I have no idea if D star has a issue with Doppler shift on radios that tune in 5Khz increments of whatever, but it's been suggested that it could be handled with finer tuning, narrower channels, or some kind of AFC. So, you ask your box manufacturer to allow finer tuning (or the chip manufacturer to enable some kind of AFC). They do the market research and decide that you don't represent a big enough market to justify the redesign. If as one commenter has has is true, then you the consumer have no (legal) recourse. Even if you wanted to do improve it your hands are tied. The alternative of having the ability to change and adopt the code to suit various constraints could certainly be handled in an open and possibly very messy manner involving failed experiments but in which everyone has the ability to learn something according to his or her ability if he or she wishes to participate. It sounds like there are people doing just that. (It's the same reason I am on this reflector but don't talk much--I think there are many more, but the list admin would know for sure).
Or you can just buy "what works" (works for what?) off the shelf, which is certainly the hope of the commercial interests. But something in my bones (and maybe yours, too) tells me that that kind of pragmatism might not be so good in the long run. It's our future, I guess, and I can't make that decision for anyone, but to dismiss these concerns out of hand in the name of "progress", as it often is, seems rather foolish to me. There's a big picture here that everyone owes it to themselves to take some time to ponder.
I'm glad this is being discussed. Cheers, Andy K0SM/2