On Fri, 17 Nov 1995, Brian Behlendorf wrote:
> On Thu, 16 Nov 1995, Jon Wallis wrote:
> > How about somewhere becoming a "copyright library" for the Web? Web authors
> > would be required to supply their URL to this site, which would then
> > classify and catalogue the page (for fee).
> > <take tongue out of cheek>
> Actually what I think we will see are publishers which offer "permanent"
> archives for documents, which serve two purposes:
Although this is off the mark of the original enquiry, it does raise some
interesting issues very much associated with the WWW that should conern
all academic and governmental institution - ie permancy of records. What
you go on to describe is a library or government archive.
> 1) A long-lived URL for future reference, no matter what happens to the
> admistrative and/or logistical infrastructure surrounding the original
> 2) A legal asset which is validated by its datestamp and submitter
> (either through a public key or some out-of-band verification) and thus
> stands as proof of existance for copyright and contractual reasons.
The Internet provides for mirror sites and repositories already -
something that they are good at when it comes to applications, code
libraries and the like. Not so good when it comes to text MIME types.
In fact, I would go so far as to say that the Web has re-introducted in
this new medium the equivalent of the "chained book". Notice has much
text material is designed and written for a particular site making it
difficult to mirror - even when commercial considerations are not apparent.
The greatest protection of our culture has been libraries and the
inherent ability to reproduce literature and science. To tie material to
a site, to have restrictive copyright laws and not to be able to retrieve
material of the Web diminishes access to knowledge.
> Joe english writes:
> > Here's a thought: how about using the Usenet newsgroup hierarchy as a
> > classification scheme? I.e., "if this Web page were a Usenet article,
> > in which newsgroup(s) would it belong?"
> Are you crazy? Usenet as a model hierarchical classification scheme?
> bwahaha. Ack - sorry. I think a lot more success would had using simple
> keywords. To accomplish what you want, though, it would seem like a
> variant of the <LINK> tag, which is designed for use as a "this document
> is related to this other document"-ish expression. Not a direct link,
> but something user-agents could use in interesting ways....