4 ms·

> That way the community has the opportunity to run their own servers if they want to. That might be fine for very small titles - where the "game server" is a

by strags 5mo ago

> That way the community has the opportunity to run their own servers if they want to. That might be fine for very small titles - where the "game server" is a relatively simple binary that can be run anywhere. Larger titles depend on a huge amount of infrastructure, for authentication, progression, matchmaking, etc... It's not feasible to open-source all of that, especially given that it may well still be in use for more recent titles.

Ukv 5mo ago
> It's not feasible to open-source all of that, especially given that it may well still be in use for more recent titles If they're still running their authentication server (for example), then they wouldn't need to release that service. Patching the game to no longer contact the authentication server would also be acceptable, for services that aren't a core part of the game. It's pretty likely the game already allows this for development/debugging. If they've accepted money from people to buy the game, and don't want to keep the authentication service running, and don't want to patch the game to no longer require the authentication service, and don't want to refund people, and don't want to release the authentication service so others can run it - I think it's fair for a regulation to force one of those.
- knollimar 5mo ago
  So do games just have to have a perpetual endowment to fund any shared component costs? This seems like a logical conclusion. You wouldn't get scalability from reuse (e.g. reusing an auth library). Or what's likely cheaper is budgeting for that patch in the game. You may bemoan "oh they just don't want to release the auth service", but it functionally shuffles the cost math. I'd personally rather the 5% cheaper games than trying to play a multiplayer only game 20 years later wtih 6 people on the server.
  - Ukv 5mo ago
    > So do games just have to have a perpetual endowment to fund any shared component costs? This seems like a logical conclusion. They don't need to keep services running perpetually. strags's objection seemed to be that it could be infeasible to release services like authentication that they're still running, to which I'm saying they don't really need to consider any of this until they stop running it. > You may bemoan "oh they just don't want to release the auth service", but it functionally shuffles the cost math. Releasing or patching it out is largely just fulfilling their side of the deal. If I sell you a lawnmower that depends on some authentication server to start up, then shut down the server the next day (I got your money, why would I keep the cost?), and don't release the server code or a patch to work without it, then would you not say I've scammed you? The resource cost of everyone I've sold to losing access to their lawnmowers would be far greater than what it'd cost me to release a patch, just that the former is not a cost borne by me if the law allows me to ignore it. > I'd personally rather the 5% cheaper games than trying to play a multiplayer only game 20 years later wtih 6 people on the server. Allowing a company to cut people off of their software (large cost) just to save having to push out a patch (small cost) will, on net, result on more expensive products - since on net you're wasting more resources. Particularly when it comes to authentication checks, this doesn't just apply to multiplayer games. Imagine if this applied to other forms of media (already kind of happening with DRM), like if we couldn't read books from over 20 years ago.
    - knollimar 5mo ago
      "If I sell you a lawnmower that depends on some authentication server to start up, then shut down the server the next day (I got your money, why would I keep the cost?), and don't release the server code or a patch to work without it, then would you not say I've scammed you?" A lawnmower isn't a piece of software. It's not licensed. There's an expectation it should continue working. The lawnmower is a single player game. I think it's understood that online games are a license (read multiplayer ones, not the Crew). If I sell you an IoT lawnmower, and you get 20 years out of it, do I owe you a full refund if I shut down my server? imo, any refund should be prorated. I get your externalities argument, but I think multiplayer games should be treated differently if there's not an easy solution. >Allowing a company to cut people off of their software (large cost) just to save having to push out a patch (small cost) will, on net, result on more expensive products - since on net you're wasting more resources. I don't think all the patches are small costs when you factor in licensing, etc. Also keep in mind if you use a library for networking and the API changes, do you have to then roll it on your own? I'm skeptical of the middleware that's made life easier >Allowing a company to cut people off of their software (large cost) just to save having to push out a patch (small cost) will, on net, result on more expensive products - since on net you're wasting more resources. This doesn't logically follow. Mandating you need to put out a patch creates a legal obligation that would sit on your books. Cutting the 12 people off your multiplayer game after 20 years isn't a large cost, and it's not going to make your next game more expensive. It was an externality that made consumers sad, not one making products more expensive. >Particularly when it comes to authentication checks, this doesn't just apply to multiplayer games. Imagine if this applied to other forms of media (already kind of happening with DRM), like if we couldn't read books from over 20 years ago. I think the DRM thing is a separate issue from the mandate for recoding games to be usable post server shutdown. I'd like it to be legislated separately as well. The books are analogous to single player games. I think it's a slippery slope to turn that entitlement onto multiplayer games; if not that, then why not all software that you buy? Everyone should get a full refund when any software EoL's and companies go bankrupt whenever an online product stops being profitable. You don't get a full refund in other sectors when they kill the consumable after a long while. Printer cartridges can stop being made and you don't get a refund for your printer. We didn't give HP the option "make your competitors ink work on your printer, give full refunds, support indefinitely, or open up your ink manufacturing line blueprints".
- dontlaugh 5mo ago
  Plenty of games (especially MMOs) have lots of gameplay logic in the server. In many cases that is intertwined with the rest of the intrastructure, like databases, logging, deployment or even subscription services. Lots of games simply wouldn’t be functional without the publisher’s infrastructure. Of course that is regrettable and could be changed, but it would require a significant change in incentives.
- strags 5mo ago
  Authentication is an interesting example - it sounds like might be the easiest component to remove. But without authentication, you don't have identity. And without identity you have no viable notion of accounts - and without accounts you don't have persistence, entitlements, progression, achievements, or any of the meta aspects that are deeply entwined with modern games. Not to mention how extensively identity ties into Matchmaking - another fairly complex backend service. This legislation might be more persuasive if it were tied to a reasonable time limit, but I don't see anything of that nature in the text. An obligation to support or refund customers that lasts for a fair timespan (ie. preventing rugpulls) is far less onerous than an obligation to release your code to satisfy someone's nostalgia.
  - Ukv 5mo ago
    For many games (and software, IoT devices, etc.), persistence/progression is tied to save files on your own device, and any authentication is more to the publisher's benefit of making sure you have an authorized copy - that's where it should be fine to just patch out the authentication check. Even for games with centralized server-side progression, if you don't want to release the authentication service when you stop running it, then it could still be acceptable to patch out authentication by default and let those running community servers substitute in their own access system (like cracked/leaked unofficial servers already do). > This legislation might be more persuasive if it were tied to a reasonable time limit, but I don't see anything of that nature in the text. An obligation to support or refund customers that lasts for a fair timespan (ie. preventing rugpulls) is far less onerous than an obligation to release your code to satisfy someone's nostalgia. Should we be able to read books and watch movies that were released over, say, 20 years ago - or is that just satisfying someone's nostalgia? Maybe you don't think this matters for games, but with DRM it seems other media isn't far behind.
account42 5mo ago
Excuses. If there is a legal requirement you can watch all those concerns evaporate away.

by strags 5mo ago

Ukv 5mo ago

> It's not feasible to open-source all of that, especially given that it may well still be in use for more recent titles If they're still running their authentication server (for example), then they wouldn't need to release that service. Patching the game to no longer contact the authentication server would also be acceptable, for services that aren't a core part of the game. It's pretty likely the game already allows this for development/debugging. If they've accepted money from people to buy the game, and don't want to keep the authentication service running, and don't want to patch the game to no longer require the authentication service, and don't want to refund people, and don't want to release the authentication service so others can run it - I think it's fair for a regulation to force one of those.

knollimar 5mo ago
So do games just have to have a perpetual endowment to fund any shared component costs? This seems like a logical conclusion. You wouldn't get scalability from reuse (e.g. reusing an auth library). Or what's likely cheaper is budgeting for that patch in the game. You may bemoan "oh they just don't want to release the auth service", but it functionally shuffles the cost math. I'd personally rather the 5% cheaper games than trying to play a multiplayer only game 20 years later wtih 6 people on the server.
- Ukv 5mo ago
  > So do games just have to have a perpetual endowment to fund any shared component costs? This seems like a logical conclusion. They don't need to keep services running perpetually. strags's objection seemed to be that it could be infeasible to release services like authentication that they're still running, to which I'm saying they don't really need to consider any of this until they stop running it. > You may bemoan "oh they just don't want to release the auth service", but it functionally shuffles the cost math. Releasing or patching it out is largely just fulfilling their side of the deal. If I sell you a lawnmower that depends on some authentication server to start up, then shut down the server the next day (I got your money, why would I keep the cost?), and don't release the server code or a patch to work without it, then would you not say I've scammed you? The resource cost of everyone I've sold to losing access to their lawnmowers would be far greater than what it'd cost me to release a patch, just that the former is not a cost borne by me if the law allows me to ignore it. > I'd personally rather the 5% cheaper games than trying to play a multiplayer only game 20 years later wtih 6 people on the server. Allowing a company to cut people off of their software (large cost) just to save having to push out a patch (small cost) will, on net, result on more expensive products - since on net you're wasting more resources. Particularly when it comes to authentication checks, this doesn't just apply to multiplayer games. Imagine if this applied to other forms of media (already kind of happening with DRM), like if we couldn't read books from over 20 years ago.
  - knollimar 5mo ago
    "If I sell you a lawnmower that depends on some authentication server to start up, then shut down the server the next day (I got your money, why would I keep the cost?), and don't release the server code or a patch to work without it, then would you not say I've scammed you?" A lawnmower isn't a piece of software. It's not licensed. There's an expectation it should continue working. The lawnmower is a single player game. I think it's understood that online games are a license (read multiplayer ones, not the Crew). If I sell you an IoT lawnmower, and you get 20 years out of it, do I owe you a full refund if I shut down my server? imo, any refund should be prorated. I get your externalities argument, but I think multiplayer games should be treated differently if there's not an easy solution. >Allowing a company to cut people off of their software (large cost) just to save having to push out a patch (small cost) will, on net, result on more expensive products - since on net you're wasting more resources. I don't think all the patches are small costs when you factor in licensing, etc. Also keep in mind if you use a library for networking and the API changes, do you have to then roll it on your own? I'm skeptical of the middleware that's made life easier >Allowing a company to cut people off of their software (large cost) just to save having to push out a patch (small cost) will, on net, result on more expensive products - since on net you're wasting more resources. This doesn't logically follow. Mandating you need to put out a patch creates a legal obligation that would sit on your books. Cutting the 12 people off your multiplayer game after 20 years isn't a large cost, and it's not going to make your next game more expensive. It was an externality that made consumers sad, not one making products more expensive. >Particularly when it comes to authentication checks, this doesn't just apply to multiplayer games. Imagine if this applied to other forms of media (already kind of happening with DRM), like if we couldn't read books from over 20 years ago. I think the DRM thing is a separate issue from the mandate for recoding games to be usable post server shutdown. I'd like it to be legislated separately as well. The books are analogous to single player games. I think it's a slippery slope to turn that entitlement onto multiplayer games; if not that, then why not all software that you buy? Everyone should get a full refund when any software EoL's and companies go bankrupt whenever an online product stops being profitable. You don't get a full refund in other sectors when they kill the consumable after a long while. Printer cartridges can stop being made and you don't get a refund for your printer. We didn't give HP the option "make your competitors ink work on your printer, give full refunds, support indefinitely, or open up your ink manufacturing line blueprints".
dontlaugh 5mo ago
Plenty of games (especially MMOs) have lots of gameplay logic in the server. In many cases that is intertwined with the rest of the intrastructure, like databases, logging, deployment or even subscription services. Lots of games simply wouldn’t be functional without the publisher’s infrastructure. Of course that is regrettable and could be changed, but it would require a significant change in incentives.
strags 5mo ago
Authentication is an interesting example - it sounds like might be the easiest component to remove. But without authentication, you don't have identity. And without identity you have no viable notion of accounts - and without accounts you don't have persistence, entitlements, progression, achievements, or any of the meta aspects that are deeply entwined with modern games. Not to mention how extensively identity ties into Matchmaking - another fairly complex backend service. This legislation might be more persuasive if it were tied to a reasonable time limit, but I don't see anything of that nature in the text. An obligation to support or refund customers that lasts for a fair timespan (ie. preventing rugpulls) is far less onerous than an obligation to release your code to satisfy someone's nostalgia.
- Ukv 5mo ago
  For many games (and software, IoT devices, etc.), persistence/progression is tied to save files on your own device, and any authentication is more to the publisher's benefit of making sure you have an authorized copy - that's where it should be fine to just patch out the authentication check. Even for games with centralized server-side progression, if you don't want to release the authentication service when you stop running it, then it could still be acceptable to patch out authentication by default and let those running community servers substitute in their own access system (like cracked/leaked unofficial servers already do). > This legislation might be more persuasive if it were tied to a reasonable time limit, but I don't see anything of that nature in the text. An obligation to support or refund customers that lasts for a fair timespan (ie. preventing rugpulls) is far less onerous than an obligation to release your code to satisfy someone's nostalgia. Should we be able to read books and watch movies that were released over, say, 20 years ago - or is that just satisfying someone's nostalgia? Maybe you don't think this matters for games, but with DRM it seems other media isn't far behind.

account42 5mo ago

Excuses. If there is a legal requirement you can watch all those concerns evaporate away.