And open platforms, if browser vendors decide to break your content. Hard to say at this point whether HTML5 games will end up lasting any longer than Flash games did. Openness is valuable but you also just have to look at whether the platform owner(s) are going to stick around (Google isn't going anywhere, at least) and whether they have a long-term commitment to the products and APIs they put out there. HTML5 is starting to look like a churn nightmare between arbitrary changes to audio policy and major features getting shut off due to threats like Spectre.
Adobe was a very bad platform steward regardless, but 5 years from now it may turn out that Win32 was a better choice in terms of developer investment/maintenance than HTML5 for many developers. iOS has had an incredible upkeep cost for indie game developers who released products on it early on, and we've seen many developers opt to pull games from the store instead of spend time+money updating them.
Hopefully WebAssembly helps fix things here by providing a much cleaner compile target (with well-defined APIs) that has good performance. Moving to HTML5 had a major performance hit for people previously using Flash or Unity's plugin but most of that hit is gone once you use wasm.
Adobe was a very bad platform steward regardless, but 5 years from now it may turn out that Win32 was a better choice in terms of developer investment/maintenance than HTML5 for many developers. iOS has had an incredible upkeep cost for indie game developers who released products on it early on, and we've seen many developers opt to pull games from the store instead of spend time+money updating them.
Hopefully WebAssembly helps fix things here by providing a much cleaner compile target (with well-defined APIs) that has good performance. Moving to HTML5 had a major performance hit for people previously using Flash or Unity's plugin but most of that hit is gone once you use wasm.