An idea I had was to make an import service that imports into your own archival node. I was thinking of the Twitter GDPR archives as source but it might work with other platform data as well. That way you can recreate content with original time stamps, and serve it back to the network. Combined with a reverse proxy relay service would work well.