Why not a 128-bit identifier? It is important for a universal
identifier to be as compatible as possible with existing technology and usage
while at the same time being capable of identifying all necessary items. We
are now entering the stage of 64-bit processors. This means that programs for
selecting documents, establishing communications, handling transaction, and
managing effectively the storage and retrieval processes can all be based upon
the 64-bit identifiers. If all information is indexed and broadcast by
satellites, retrieval in-house or from external servers can be accomplished
with one 16 numeric hex identifier, while existing credit card numbers, most
of which have 16 characters to base 10, can still be used and identified in
the system. I can provide my breakdown of suggested numeric identifiers for a
16-bit unicode page, all words in all language for subject index terms, all
proper names and places including all domain names and web pages,
communications devices, universal product codes, and geographic segments. A
separate 64-bit index is reserved for all identified astronomic items.
With respect to copyright, I agree that encryption offers the best
solution. My thought would be that a creator of intellectual property would
file an encrypted copy together with the key for use by the Library of
Congress in the United States and by similar institutions in other countries.
Copyrighted material is public material, the use of which is governed by
legislation. I believe that most owners of such material would like to have it
distributed as widely as possible with an index of the contents provided that
they would derive income from the sale of the key. Appropriate items could be
electronically "watermarked" and given varying levels of protection.