Everything except making a store people wanted to use? Ethan Evans, who was previously Vice President of Prime Gaming at Amazon, has a short retrospective of trying to take on Steam.
So many of their most beloved features are crazily dev intensive to maintain, and critically they’re not static. Amazon never really updates their consumer interfaces, steam is constantly adding new features and reworking their old ones across all their UX. Its just not economically feasible to pop in and replace them if you’re a publicly traded company, the shareholders would look at the maintenance costs alone and faint
Even some of these features are transparent to users but not to devs, for example to say few:
Multiplayer (big one)
Microtransactions (eww!)
Steam Audio (support for specific Audio stuff, not voice)
Steam cloud (Save games)
DRM
Input
Key management
Playtest
Steam Voice
Valve Anti-Cheat
Stats and achievements
Workshop
and not as important but relevant to some:
Virtual Reality support
In a sense, this is not so different from what AWS is doing: Basically you offer a service with an API so your customers don’t need to create the thing from scratch, but at the same time they become dependent of your exclusive services.
They can’t possibly be dev intensive to maintain, given what we know about how many devs Valve has.
They are VERY expensive and difficult to make, though, particularly if you don’t already own the PC platform. It’s not that every competitor wouldn’t like to match their feature set, it’s that Steam has had two decades of a head start and is a whole software company devoted entirely to this, as opposed to trying to simultaneously… you know, make games and stuff.
Steam Multiplayer, VAC and Steam Wallet integration are all extremely dev intensive examples, I’m not sure conflating team size with dev intensity is a great way to look at it since that’s not generally how software development works. (Unless you’ve got client accounts or deep customization, ofc)
Absolutely no argument about the rest, though. Steam built the entire concept of the market it dominates, and now people are trying to build their own little versions without any of the 20 years of novelty in even just defining the medium that valve has done.
So many of their most beloved features are crazily dev intensive to maintain, and critically they’re not static. Amazon never really updates their consumer interfaces, steam is constantly adding new features and reworking their old ones across all their UX. Its just not economically feasible to pop in and replace them if you’re a publicly traded company, the shareholders would look at the maintenance costs alone and faint
Even some of these features are transparent to users but not to devs, for example to say few:
In a sense, this is not so different from what AWS is doing: Basically you offer a service with an API so your customers don’t need to create the thing from scratch, but at the same time they become dependent of your exclusive services.
They can’t possibly be dev intensive to maintain, given what we know about how many devs Valve has.
They are VERY expensive and difficult to make, though, particularly if you don’t already own the PC platform. It’s not that every competitor wouldn’t like to match their feature set, it’s that Steam has had two decades of a head start and is a whole software company devoted entirely to this, as opposed to trying to simultaneously… you know, make games and stuff.
Steam Multiplayer, VAC and Steam Wallet integration are all extremely dev intensive examples, I’m not sure conflating team size with dev intensity is a great way to look at it since that’s not generally how software development works. (Unless you’ve got client accounts or deep customization, ofc)
Absolutely no argument about the rest, though. Steam built the entire concept of the market it dominates, and now people are trying to build their own little versions without any of the 20 years of novelty in even just defining the medium that valve has done.