“I’m say­ing that, like a real evic­tion, there should be prac­tices in place. When you open your doors to host­ing user con­tent, you should have rules in action that, unless it’s a com­plete and total fire sale and you have no hope of even stay­ing open that long, then you should be required, yes by law, ass­holes, to make the data avail­able to cus­tomers for an extended period of time.
…
If you tell peo­ple they can upload their con­tent, you should have a clear and dis­tinct way for them to retrieve their con­tent. Peo­ple do it ad-​​hoc as they can, but the abil­i­ties of most peo­ple, the peo­ple with­out an engi­neer­ing degree or years of expe­ri­ence, to get back what they put up is min­i­mal. It’s not that impor­tant. We should make it important.”

“… Their pro­ce­dures took the phys­i­cal object as the unit of repro­duc­tion. Because the Bar­rett copy of Uncle Tom’s Cabin was col­lected in the form that it was, I pho­tographed each news­pa­per num­ber within it in order. When it came time to plan my own dig­i­tal project, which included a fac­sim­ile repro­duc­tion of the Bar­rett object and a tran­scrip­tion of Stowe’s text, I had mul­ti­ple choices. But I had already com­pleted a dig­i­tal repro­duc­tion of the Bar­rett object that included pho­tographs of cov­ers, end papers, and the pages of num­bers that lacked install­ments of Stowe’s text. The the­ory, I would sub­mit, is an effort to deal thought­fully with an insti­tu­tional pro­ce­dure of repro­duc­tion that con­trasted with my own inter­est in the tran­scrip­tion. Had I been in charge of the reproduction–not forced to engage against insti­tu­tional practices–I might well have decided to repro­duce only those pages that include Stowe’s text…”