You make some good points here. Now let me speak to it as an engineer who is actively working to implement some of these features:

1. Multimorphic tries to design their stuff to run peer-to-peer in some cases, so you can directly link on a LAN. Multimorphic does run a pair of matchmaker servers responsible for finding peers across the Internet, but such a thing could be replicated to a stand-alone application and hosted by the community.

2. I've been working with Scorbit on some of their title adoption, and they plan to be fairly open with their APIs so that you can implement your own web service layers and mitigate risk via the community if necessary.

3. For those who mention RFM/P2K, you have to remember that it was originally designed for a LAN and implemented over a weekend. The stuff we have today was pretty much reverse engineered out of that, and various other layers have been put in place to essentially hack it so that it works over the wide area network.

4. Servers are lower cost than you think. Most of the bandwidth would be consumed by edge content (software update downloads). Gameplay bandwidth consumption on a pinball machine is fairly meager, and most messages can fit into less than 0.5 kilobytes. Software updates would be about $0.05 per user that downloads them given a conservative game/asset update package size of 2 gigs at unscaled pricing.

The bigger concern as we work to roll these features out is actually security. A pinball machine is inherently a dangerous device if left in the wrong hands. Things go boom or catch on fire. A lot of focus is put into reducing the attack surface area, and for decades it's been much easier to just leave the whole connectivity piece out of it if you're even thinking_ about venturing outside the LAN.

But, adoption and positive reception from the community (due to some sort of value being provided) will spark the need/demand to maintain these systems for decades to come, but simply not using something because it might_ not be around in 40 years is a bit short sighted. Hell, a lot of US might not be around in 40 years.

The priority, still is and always has been to make the game fun. If we can leverage connectivity to further make the game fun or lower maintenance costs, then great, you'll see it sooner. Leaderboards aren't fun, but connecting up your brand new Stranger Things or Heist pinball machine and playing co-op with a clan across the world in a way that makes the joint storyline change for everyone's games while you're doing it, that sounds pretty fun.

Just my two cents, respectfully.

-- Jimmy