Some people here talk of the difficulty of citing sources that may disappear. This can easily be solved by just reproducing the proof (and stating explicitly th

Some people here talk of the difficulty of citing sources that may disappear. This can easily be solved by just reproducing the proof (and stating explicitly that that is what you're doing and providing references to the ephemeral sources as they exist when you write it). Of course you should clean up the proof as much as possible. This only needs to be done once since others can site your reproduction of the proof if they have similar misgivings.