I assume you are already aware of the active community of video game historians and preservationists involved with the retro arcade and console emulation movements (ie MAME/MESS, etc). If not, here's a link to get you started: https://www.progettosnaps.net/dats/
The retro preservation community is focused on documenting and archiving these historically significant creative and artistic interactive digital works for the long-term future.
The preservation community is careful to steer clear of shorter-term legal and commercial issues. Copyright typically expires in less than a hundred years which is brief on art history timescales. Commercial interests in monetizing intellectual property are even more ephemeral. Even for immediate purposes, there are broad exemptions to copyright for non-commercial fair use including educational, scholarly, editorial and creative purposes throughout most of the world.
This is well-trodden territory. Over the years the issues have been discussed at length in the preservation community. By searching terms such as "MAME", "MESS", "emulation", "preservation" and "fair use" you can find official position statements as well as detailed legal analysis and individual commentary (example: https://mamedev.emulab.it/haze/2013/12/28/archive-org-covera...).
Excerpt from the linked 2013 post by a long-time MameDev:
> I feel it important to once again point out that MAME (and MESS) are NOT projects about playing free games and that we do not condone or facilitate large scale piracy.
> What we provide is factual references, and emulation of hardware components. Our aim is to make these as good as possible, and be able to emulate ANY possible piece of software that might run on them; emulating and documenting more software allows us to improve our hardware emulations, improving our hardware emulations allows more software to run, naturally we reference what we have used to make this progress, but at no point do we actually provide it.
> In terms of project goal you’ve only got to look at the MESS part of the codebase (that we now ship with the MAME source) to see there are emulations of random devices like EEPROM programmers, Car Computers, Digial Clocks and Homebrew computers etc. This hopefully shows that the project is about something much more important and with a much wider scope than what some people assume (that they’re simply projects about playing games for free). Even in MAME we emulate things like Firmware update programs, and have skeleton drivers for Coin-operated Jukeboxes (and in the most recent update, an electronic Darts board) none of which would be included if it was merely a project about playing games. FWIW this has always been one of my arguments for fully combining the project binaries by default, it makes this position a LOT clearer because most of the cases that demonstrate this well do come from MESS.
> I should also stress that popular systems in MESS are also often handled differently to other emulators, there are reasons we document the proper content of cartridges, and require real rom dumps for things like NES where possible (rather than simply using .nes files which lose this information) Design choices like this might make things more inconvenient for users (and has zero benefits in terms of playability) but does result in our history being better documented which is more important when it comes to the goals of the project and again emphasizes that this isn’t simply a project about ‘free games’
MAME is likely ok but the ROMs of course not even if you pretend to be a museum.
It's just tolerated by owners of the copyrights:
+ they don't know how to make money with it anyway because nobody wants to pay for it (except with physical arcade machines).
+ it can support the license credibility (e.g. Pac-Man).
+ the copyright owners may not be around anymore.
> "it's legal to distribute pirated copies of games if it's to preserve them"
Perhaps you meant to reply to the poster I was replying to. To be clear, I would never say nor imply "it's legal to..." <anything>, as predicting the legality of some specific scenario is essentially impossible with any degree of confidence.
Even expert lawyers, judges and law school professors are frequently incorrect in such predictions. In the U.S. legal system, what is and isn't "legal" varies wildly based on the particular jurisdiction, context, defendant, plaintiff, defense, prosecution, judge, jury, etc involved.
And even predicting a given ruling (guilty/not guilty) doesn't tell us on what specific counts, how they were charged or what the sentence or damages may be nor any possible settlements or appeals. Intellectual property law in general and copyright in particular is notoriously complex, nuanced and situational. Especially so for civil actions. That's why I was so careful in my post not to state any opinion regarding legality at all. I only noted that exemptions like "fair use" exist and some copying in some situations is permitted. The answer to anything else is both a) "it depends", and b) "it's complicated."
There was some kind of rule that allows The Internet Archive to maintain (but not distribute) data that it has acquired without the party the data came from having a license to copy it to them. I don’t remember the name but as far as I remember it was tested in court.
