edsan points out that a good thing about Mike D's Ruins & Ronin project is that it doesn't allow itself to get hung up on details about feudal Japan. It's a fantastical vision of feudal Japan in the same way that D&D is a fantastical vision of feudal Europe. In the same way that D&D didn't worry itself unduly with accurately representing the life of a peasant in 11th-Century Avignon, so there's no reason for a 'Japanese' fantasy game to accurately represent life in 15th-Century Kanazawa.
I both agree and disagree with this. D&D might not be very realistic in its depiction of medieval society, but Harn is. (And during the 2e era, TSR made several forays into historical realism which I found particularly enjoyable.) Similarly, Ruins & Ronin might throw out socio-cultural detail in the name of pulpy fun, but The Blossoms Are Falling proved that basing an entire game on socio-cultural detail can be just as entertaining. It's a matter of those horses and their courses, again. Self-evidently, so long as players are clear as to the aim of the game, it shouldn't matter. As with the social-contract expectation that players should know whether they're turning up to play Tekumel canon or EPT-for-dummies, potential buyers/downloaders of a game should know what it is they're getting into. (This was always my problem with Legend of the Five Rings. If it was supposed to accurately represent Japanese society it was an embarrassing failure, but if it was supposed to be a fantasy game which just happened to take some of its influences from Japan, then I could forgive it. But which it was was never made clear.)
Anyway, in the comments to edsan's post, vraymond (who also happens to have written about this sort of thing recently) mentions running a game set in Kofun era Japan. I've thought about this too. (I wrote about it briefly here.) During that period, roughly between 250 and 538 AD, Japan was a disunited region of chiefdoms and tribes who built huge mound-like tombs (rather like iron age Britain). However, not much else is known. The societies present in the archipelago were all pre-literate, so almost all of our scant knowledge comes from sources in China and Korea, who were more technologically advanced, and who began to make regular journeys to Japan during the period.
My idea for a Kofun-era game was that it would take place in a sort of quasi-historical version of East Asia where mythological beings, magic and so forth are real. The players would be Chinese or Korean diplomats, missionaries, traders and/or adventurers, who make the journey across the Yellow Sea and find a strange, mist-shrouded, forested set of islands full of barbarians, ghosts, dragons and ogres. These strangers in a strange land would then have to make of it what they could. (Rather like how in traditional EPT games, the players are always foreign newcomers trying to survive in Tsolyanu.) I think the advantage to this sort of game is that it allows a DM to come up with essentially whatever he wants - he can throw in any bastard mixture of Japanese mythological tropes - without having snobs (like me) complaining that "it wouldn't have been like that". And yet at the same time, if he's the setting fetishist type, he can spend weeks or months of his life developing an intricately detailed game world, and it won't matter that they players don't know much about it because, by definition, they're foreigners. There's no potential for conflict between players who "know about" Kofun-era Japan and those who don't, because everybody including the DM is in the same position of more-or-less complete ignorance. And at the same time there's plenty of scope for Japan-inspired fantastical weirdness.
I'm working on my mollusc-world/Yoon-suin setting at the moment, but maybe after that I'll think about making a Kofun-era based something-or-other.
