>> that this is actually Steam Machine 2.0. Valve already tried this a decade ago, and it flopped.
> I find this framing to be beyond maddening [...]
> It was also a thoughtful partnering with hardware vendors
As numerous post-mortems (some of which I quoted in the article) recount, the hardware partners themselves largely consider their experiment back then a flop as well.
> But it was a thoughtful, intelligent long-term commitment to an ecosystem
With respect, I think you're overselling it. It's hard to call a machine that basically didn't play any of the at-the-time hits well "a thoughtful, intelligent" move. If you read some of those linked post-mortems, I think you might agree as well.
> I think it's best understood as a return on investment that begun those many years ago
I think there's nuance here, which is that Valve made lemonade from the lemon that was the flop of the Steam Box. They turned that failed move into an initial investment through diligence and effort. In a sense, that's part of what I'm trying to bring attention to -- Valve didn't just write off the failure and abandon the market, but took signal from it and tried again.
Fair point on the vendors - surely they hoped to make $$ from it. But I think you're underestimating the significance of standing up Proton and the critical experience working through bugs and getting experience with hardware, and gradually growing the inventory of compatible games. Simply put, there's no Steam Deck without the Steam Machine, which says everything about the value of the Steam Machine.
> But I think your underestimating the significance of standing up Proton
I don't think I'm underestimating it at all. Proton and SteamOS were huge, they were extremely well-timed, and they've been a boon for everyone involved (except M$ shareholders, I guess).
However, none of that necessitated whatever the Steam Box release was. It's not like it moved a significant number of units and that's why Valve invested in Proton/SteamOS; Steam Box was long discontinued before the first public release of Proton (2018, IIRC).
> Simply put, there's no Steam Deck without the Steam Machine
Agreed, and I call that out in the article, but that doesn't make its original release not a flop. Hence my lemonade comment -- you don't make lemonade from apples; you have to have a lemon first.
I think there is something to be said for companies putting their money where their mouth is. Getting behind Gaming on Linux with a hardware launch is pretty substantial and Valve have been explicit since 2013 that they think it's the way forward.
They also continue to have first class support for BOTH windows and Linux without forcing it down anyone's throat which is not the strategy for literally every tangent related market. There are no super annoying layers to this strategy Valve could have done a number of things to force users to use Linux such as. - Linux game exclusives and Linux discounts. - Preventing the steam machine from running on a windows machine by dropping certain hardware - Making windows users second class by not releasing the latest updates and features. - Making other hardware incompatible with windows e.g. Valve Index, Steam Controller etc