Re: AW: [xml-dev] RFC for XML Object Parsing

[Hans-Juergen Rennau <hrennau@yahoo.de>]

Gentlemen,

I am not sure if such historical facts and details are really important in the present context. At any rate, what interests me is the relationship between Brian's initiative and current XML. And what strikes me is the following. The XML model defines the information content of a given document; a document is the content which it is, and any glimpse beyond the document is out of scope. In particular, there is no room for distinguishing between a resource and its representation, - resource and representation are always one. But Brian's approach, so it seems to me, would build into the information content of a document a statement establishing a relationship with a seperate instance of information content (the data referenced by the oid), assigning to one (the data containing the oid) the role of being an update of the other - assigning to both the roles of subsequent states of the resource which assumes those states, but is not identical to them. And this
 is certainly an interesting idea.

Hans-Juergen

--------------------------------------------
Arjun Ray <arjun.ray@verizon.net> schrieb am So, 23.3.2014:

 Betreff: Re: AW: [xml-dev] RFC for XML Object Parsing
 An: "xml-dev@lists.xml.org" <xml-dev@lists.xml.org>
 Datum: Sonntag, 23. März, 2014 22:29 Uhr
 
 [Default] On Sun, 23 Mar
 2014 12:11:47 -0600, Brian Aberle
 <xmlboss@live.com>
 wrote:
 
 | Call it what you
 may, HTTP 1.0 didn't have it.  It was added to HTTP
 | because it was needed.
 
 I'm afraid your historical recall is
 faulty.
 
 The fact of the
 matter is that caching was taken seriously quite
 early.  From an archive of the early years of
 the www-talk mailing
 list (referenced at
 http://lists.w3.org/Archives/Public/www-talk/2013SepOct/0002.html):
 
 
     http://inkdroid.org/tmp/www-talk/0237.html
     http://inkdroid.org/tmp/www-talk/0433.html
     http://inkdroid.org/tmp/www-talk/0453.html
 
 What eventually became
 HTTP/1.0 was initially named HTTP2 - because it
 was the second spec.  The original spec was
 like Gopher, and had no
 provision for header
 fields at all. (When HTTP/1.0 was finally named,
 this precursor was then jokingly dubbed
 HTTP/0.96)  
 
 The initial
 drafts of HTTP2 were by Tim Berners-Lee and Dave Raggett,
 in early 1993.  The Last-Modified header date
 back to then, with hints
 that the HEAD verb
 could be used to determined the modification status
 of a document. This was superseded by Roy
 Fielding's "conditional GET"
 proposal in early 1994
 
     http://inkdroid.org/tmp/www-talk/3465.html
 
 This was implemented in
 servers (and caching servers) before the
 -ahem= market-leading browsers (a recurring
 tale on the web, sigh).
  
 And eventually, the official HTTP/1.0 spec had
 If-Modified-Since:
 
    
 http://www.rfc-base.org/txt/rfc-1945.txt