Surely then you want them in the public domain sooner to prevent that. Sourcing a 28 year old pc to read some 'ancient' format is workable. Finding a 100+ year old pc to do the same seems less workable.