Leaf documents, Mutable documents (was: Web trends)

Tim Berners-Lee <timbl@www3.cern.ch>
Date: Wed, 15 Sep 93 10:56:14 +0200
From: Tim Berners-Lee <timbl@www3.cern.ch>
Message-id: <9309150856.AA01663@www3.cern.ch>
To: sanders@bsdi.com
Subject: Leaf documents, Mutable documents (was: Web trends)
Cc: www-talk@nxoc01.cern.ch
Reply-To: timbl@nxoc01.cern.ch
Status: RO

*****	Should webs have many "leaf" documents?

I se nothing wrong with such a web.  Personally, when WRITING
I feel the urge to put cross-references in to reduce
duplication, and help people find things and look at it in their
own way.  I feel it makes my documents more valuable.
This is my style. And my time taken to make the links.

This does NOT mean that databases with a pure hierarchy
and no cross-links are "bad".  Not at all.
Most documents I image and will be for a long
time non-hypertext and the web is very useful as
a place to put and find them.  They are
not second-class citizens.  (Not necessarily

*****   Should documents be different as a function
	of circumstances?
My view is as follows.  When a document header in HTTP
specifies the URL/N (URI) of a document, then it does it
in one of two ways: as a
which means that it does NOT guarantee that the document
will be the same, nd

which DOES.  When you are quoted a URI then you don't know
which sort you have untill you have retrieved it and got the

This is my answer also to Larry's point about mutable objects.