I'm on a roll here, so I'm going to keep going.
As I said yesterday, the audience that formerly adhered to tabletop RPGs bled away over the past 30 years when alternatives that did what this or that subset wanted better than TRPGs arose and became commercially viable. With each defection, a preference got revealed.
Those who saw the opportunity in producing a product that focused on that preference benefited from the gamble, which is how and why the gaming industry soon spread across multiple media and a plethora of genres. There is little, if any, recognition by the "professional" TRPG people that each segment bled off reveals a gamer preference that they fail to fully satisfy. For the group purporting to be in the business of selling TRPGs, they are utterly incurious as to why droves don't want their product.
The first big break came with those that recognized the dissatisfaction with the time commitment and the necessity for multiple players matching schedules as if this were a bowling league. This arose when the game, and the medium, escaped containment from the original Twin Cities-Lake Geneva wargaming axis and its norms.
There are two parts to this preference: convenience and dependence. Convenience is when you hear people complain about having to set aside time to play a tabletop RPG. Dependence is when you hear people complain about not being able to play because someone no-shows, or when they complain about not able to achieve their objectives or have fun due to others getting in the way somehow. The alternatives that bleed off the audience for tabletop RPGs, first and foremost, address one or both of these preferences.
The RPG form seen in videogames is typical of this. Single-player RPGs for PC, console, or handheld remove the need to rely on others to play and frees the gamer to engage as he desires. Outside of the raiding scene in MMOs, multi-player RPGs rarely have the same degree of "schedule your fun" that tabletop RPGs require to work at all.
The industry trend is towards player convenience, such that discrete play sessions get shorter over time as each competing game in a niche iterates upon predecessors. (e.g. DOTA2 games can go for an hour apiece, whereas Heroes of the Storm hits about 20 minutes.) Drop-in/drop-out is the goal, with match-making by algorithm removing the need for hand-selecting play groups.
You see this also in boardgames and card games. Rarely do you see popular games with discrete playtime sessions of an hour or more. Similarly, you rarely see popular games with much time spent setting up or breaking down; you also see asymmetric weight of rules knowledge, where only one user actually needs to know what's going on in multi-player games (if that) that are cooperative or asymmetric multi-player (e.g. Descent.
The idea is that shorter playtimes with no long-term commitment better fosters widespread play, and therefore is better for the scene and the business. Given the conditions of videogames, card games, and boardgames vs. TRPGs, they're right. Since the tabletop RPG makers aren't doing their jobs and properly addressing this preference, it's on the hobbyists to do it for them. Fortunately, we have a solution from the original days to draw from in solving this mess.
From the Ars Ludi blog we have the West Marches model. This model eliminates long-term commitments by going drop-in/drop-out explicitly, making each play session self-contained, and has players meet only when a sufficient number wish to do so. Combine this with an Open Table policy, and you minimize Convenience and Dependence issues.
Which means that you now have to deal with other preferences, which I will continue on tomorrow.