I consider myself a technology enthusiast. Witnessing trends in technologies both in the offline and online world converge and transform the personal computer into a multipurpose and powerful center of peoples' digital ventures is an amazing thing. One of the most exciting areas for me being the fast, quickly changing nature of games in this often called "dying" medium, that the pc is regarded at.
It's interesting to see the different themes involved in such a new, uncertain plane of action such as gaming in the linux and in general f/oss arena clash with each other. Themes like users's rights and liberties, game companies' need for a return in investment, technical viabilities in game support, and IP rights overall. It's great to see a good, mature discussion about linux and pc gaming in general occuring here in the Stardock community site, and with the valuable comments from the devs from Stardock themselves.
Most of the points they mention against linux support in games is uncertain returns of investment, and support problems. Arent these times times for breaking molds, and trying new ways ? Maybe the linux gaming market isnt as small, and it certainly is an untapped market, up for grabs right now. There are many good, quality opensourced/freesoftware games played by linux gamers, so the playerbase is there. In linux gaming community sites, the numbers speak for themselves. In all fairness, the linux gamers often say that honestly the level of support, and patch, bug fix, etc of these games (which include but not limited to: nexuiz, cube2, warzone2100, battle for wesnoth, etc) isnt the best, mostly because their dev teams work on their spare time. Another big complain is that there's a lack of good strategy games for them to play (see an opportunity here?) . And this notion that linux gamers wont pay for anything is wrong, and false. They would actually pay to support a commercially released quality game, specially if it's not the usual fast FPS game.
One big concern is the aspect of linux support being a logistical nightmare, that premise isnt that valid anymore. Linux and f/oss in general, has made strides in the last few years towards unification and standarization. Packet managers such as synaptic, portege, etc, provide a quick, relatively painless way to update and maintain a system. Although usually regarded as unnecesary and in some situations annoying, since they more often than not install many additional dependencies and extra clutter, they constitute some sort of backbone for a system's internal layout, which keeps some kind of order and standard in this systems' very varied and diverse nature.
Regarding the old directX vs. OpenGL debate, many devs are of the opinion that directx (specifically it's visual aspect, direct3d) is not in fact any easier to use than opengl. Most, if not all of its popularity derives from the overwhelming majority of windows install base, and the relatively unified, centralized nature of it's release cycles, support, etc. Last year, the release 3 of Opengl was unleashed, with lackluster reviews, and overall negative attitudes towards it. Many complained that release 3 should have been called 2.2 instead, since it provided no new features, and it had mostly been produced to cater to the workstation/CAD market. Maybe a fork in the direction and policies regarding opengl's uses should be spun off from the Khronos Group, the maintainers of the openGL specification, which could have gaming easiness of programming as the first priority instead.
As Stardock's devs have indicated, they and their affiliated game studios' development on their planned games has already began on DX. But maybe, they could probably consider for their next releases, like their upcoming game Elemental: War of Magic, to start out with opengl instead? That way, they can achieve multiplatform support for their games easily. As stated before, the times are now very difficult, and it's the time to be resourceful with new ways and markets. Low-risk gambles like this could pay off big time.