I won't go into Demigod's launch because it's something of a sore topic around here to this day. Suffice to say, the two situations only seem similar on the surface but behind the scenes the only thing they have in common is that the word Stardock was on both boxes.
With Elemental and future releases, as a practical matter, it's a relatively moot issue because in another year, digital will surpass retail in terms of sales. Our plan going forward is going to be to essentially keep games in beta for much longer amounts of time until not only do we think they're done but that they're basically way beyond a "1.0" version.
This is what we do with our non games already. For instance, we released ObjectDock v2 this week. This is a program that will likely generate in its various forms millions of dollars in revenue. But there's no drama in its launch because, realistically, the program was "done" in "beta" last Spring but we simply kept it out in beta months later until we felt it most advantageous for us to release.
By contrast, on games, we had to pick between August and February because our schedules have to meet the needs of retailers and those were our two dates. And while we thought Elemental was ready for the August date, that decision also meant that the beta closed. If we had had a public beta 5, it would have been obvious that our QA had missed some key things and it would have been no biggie to just push the date back.
Hence, what would happen with Elemental under the new system is that we would have had a beta 5 and it would likely have gotten released in say November with it not becoming available at retail until February still.
For all the grief we have gotten on Elemental, I think some people need to take a step back and realize that not many developers are making new games anymore for the PC and ask themselves why.
Since Elemental's release I've seen people assert that we're either incompetent, greedy, dishonest, or uncaring (or all of the above). I have a hard time respecting anyone who takes such a facile position. The answer is, making a brand new game for the PC in 2010 is a significant engineering challenge. Just something to think about when you're buying <Game franchise> <sequel number>: <sub-sequel title> <sub-sequel number>.