> imposes unfair burdens on developers (which include indie devs too)
Any such law should include a carve-out so that indie devs and small startups aren't impacted because just the need maintain compliance paperwork can be a burden. Carve out thresholds can be based on a combination of product revenue and units sold. Similar carve outs should generally be part of a lot of government regulations because startup entrepreneurship is so key to job growth, innovation and ensuring more choice for consumers. The best way to keep huge companies honest is making them keep earning their success by enabling smaller, hungrier new competitors laser focused on better serving customer needs.
That said, I do generally agree with your broader point that how regulations are written and enforced matters a lot. Too many start with good intentions but end up being sidestepped, subverted or triggering unintended consequences. If a "Stop Killing Games" regulation is drafted I think it should be narrowly targeted and conceived with the understanding that both the tech and business models will continue rapidly evolving and the market will quickly adapt to sidestep or subvert whatever new rules are put in place. That will likely mean that, realistically, an effective regulation probably shouldn't be as expansive or all-encompassing as we might be imagining from our armchairs.
I'd be happy if the focus was simply on getting large game companies to clearly commit up front what their commitments are over time by listing how long will each aspect of the game will continue to work by type (ie single-player offline, multiplayer self-hosted, multiplayer cloud, feature updates, security updates). Then company management and investors will know to set aside funds to cover server fees for that time period after the final sale. This isn't new or burdensome. Large companies already have accounting practices to accrue future liabilities on their books. When they sell future enterprise services to other companies there's a contract with financial reserves and revenue recognition. Selling a game to consumers with the expectation of future online delivered services should have a similarly spelled out commitment and appropriate financial reserves.
In reality, this may mean some companies choose minimum commitments that we'd all feel are far too short but as long as the consumers know up front what the commitment actually is, the free market can determine over time what costs consumers are willing to pay for which commitments. I expect some companies will try to minimize their financial commitment by making games which could obviously have offline single-player aspects always require online for everything or be subscription-only and only commit to offer the subscription for 1 month after purchase. Let them try and see how the market reacts. Government regulation isn't some magic wand we can wave to just force companies to "do the right thing" or, more specifically, make the products we want and sell them to us on the terms we'd prefer. Companies will either pass the increased costs on to consumers or not go into that business at all. Realistic regulation should focus first on two things: 1) Ensuring a level-playing field for fierce competition and, 2) clear up front disclosure of what the deal is.