Valve accidentally compiled and pushed earliest Left 4 Dead prototype called "Terror Strike" into the latest CS 1.6 update and community figured out how to play it.
I can see compiling happen accidentally, since it's probably just a compile flag and someone forgot to disable it. But pushing is really surprising, sure it can be automated, but usually you have a manual process for such things (e.g. my company's prod deployment is 100% automated, except for a manual approval step once everything is ready).
So the only way for this to happen imo is if they pushed something intentionally and has accidentally disabled/enabled a flag at some point prior.
It may very well have been two or more different people stepping on each other's feet in the dark.
I recall binge-ing Source leak summary videos. Everything that Valve uses that engine for is extremely tightly coupled.
Whenever CS:GO or Dota 2 gets an update, data miners get to work and discover a bunch of assets of unrelated source games.
Sounds like your company is doing things the halal way and using modern standards. At Valve, it's just a clusterfuck dev tool GUI on top of a monolithic codebase where no one can possibly know a fraction of what's going on.
“Hey, you know what would be fun? Let’s release really old versions of some of our games - I think fans would get a kick out of seeing them!”
“Ugh, no. Why would we want to spend the money on testing and supporting something that only a small fraction of the player base would even care about?”
“Um, ok. How about if we “accidentally” push it with our next release. We won’t have to do anything to support it - modders willl figure out how to get it going, so we don’t have to do anything, and they get a fun Easter egg. Win win.”
“Accidentally?”
“Yeah. People will backfill some reasoning for how even though we’re a professional software company, we have no idea how source code control systems work. It’ll be fun to see what they come up with.”
As much as everyone lauds valve, and its company structure, they do have problems with actually making things.
There's no real financial push to make things. They'll make the same amount this year if they release a new game or not.
So it's down to if people at the company want to make something. And it seems they choose hardware more often and struggle to get things to the finish line without any real push to make that happen.
Losing people like Chet Faliszek doesn't help much. People who got projects finished.
Tbf an environment where people can work on the things they want seems a pretty good one.
Like sure we aren't getting more games but idk the steam deck has been a wonder creation of theirs, plus feels a bit entitled to expect more sequels or games bc we are just consumers of their creation.
Because due to the "flat" corporate structure, it's based on having enough people be "passionate" about it to even get it started.
You might have half the staff "passionate" about Half Life 3, but when it comes to their individual ideas that make them passionate about it, they're actually all on wildly different pages about what they want to do and what technology they want to pursue. This means you'll technically have people on board, but because they're driven by different passions, its harder to get everything and everyone to "line up" so work can be started on such a big project.
It can leave gamers hungry for quality games, but I'm fine with all the work Valve is doing for Linux and Proton/Wine. I think those are important and worthy things to be spending time on, and I'm fine with not getting games from Valve in the meantime.
Steam being a cash cow, a corporate structure that doesn't force some/most people to work on stuff they don't want to and internal politics that disincentivise people to go off on their own.
Using version numbers to name a game that you're still updating is fucking weird. I would've figured valve would have well and abandoned 1.6 by this point.