Hello folks, may as well make my first post something to ruminate. :)
I've been considering this issue myself recently - what manner of financial model could a small company like Eagle Dynamics utilise to give them both an element of financial security and the motivation to continue development of a niche consumer product like the Digital Combat Simulator series. I actually think a subscription-based model may not be a bad idea.
Consider:
Currently ED release each product on a self-contained, individual basis as is most traditional entertainment software. An instalment is released, users are given technical support and post-deployment patches are issued to correct bugs and gameplay issues. At some point, resources must be reallocated away from the current release in order to focus development on the next instalment. Given enough time, sales for the release will dwindle to a point where it is no longer financially viable to continue maintaining previous instalments in the series. Given that users are likely to want to combine different aircraft, missions and other functionality in multiplayer and possibly other scenarios in addition to wanting older instalments updated with newer DCS technology as the core engine is developed, this makes it difficult for ED to balance the wishes of the users against the costs of maintaining older instalments and the costs of developing new instalments.
Were the core DCS engine released as a standalone, core product with aircraft, theatres and other significant features deployed as extension modules for the DCS engine, then ED could do something along the lines of charge a one-off purchase fee for major modules and an annual maintenance subscription for the core DCS engine. Consumers could then purchase as much or as little of the suite as they wish. With a modular architecture and continuous revenue from the core engine, ED may then have the financial security to continue development of the product suite as a whole, needing to update modules only when necessary to fix issues and maintain compatibility with the core engine.
Of course, the possibility of doing so depends very much on the architecture of the software and how much of ED's focus lies on consumer products compared with specialised military and other private projects.
Naturally, these are just my own thoughts on the matter. ED may already have very specific plans for the series. Debate is welcome.