Donald E. Eastlake writes:
[...]
Almost nothing in the real world uses Distinguished Names. Most DN schemes
are crap based on the futile idea of a universal X.500 dirctory. They are
useless until you map them to something real like URL's which introduces all
the problems of the storage, access to, and maintenance of such a mapping.
A good example of a real world "DN" is the ISBN number given to
printed material. This is an example of a "DN" scheme that for all
practical purposes appears to work. Unfortunately, it isn't easy to
read any semantic content out of an ISBN number.
2) the move over the next few years will be toward URN's. These are
independent of DNS resolvers. They provide location and replication
transparency (among other things).
As I am a newcomer to this group, I haven't the slightest idea about
why people seem to arguing about naming schemes for documents. In my
opinion, a naming scheme for a document is unnecessary. Why? Well,
let's see how we usually go about identifying a document:
A document is usually (in the real world) identfied by:
Author(s),Title,Publisher,Edition. In most cases these attributes are
more than sufficient to _uniquely_ identify a document. If these
attributes are _not_ sufficient (as in the case of scientific
articles), there are a host of other standard attributes used to
identify the document in question.
If the DNs proposed for documents are to replace URLs (which actually
_do_ function as proper DNs for all practical purposes), one is in
effect stating that the _location_ of the document is a vital
attribute of its identity. This would mean that _every_ copy of
Shakespeare's "The Comedy of Errors" would be identified as a unique
document, which of course is not the case.
--
tonnes(_at_)nta(_dot_)no