- From: Xiaoshu Wang <wangxiao@musc.edu>
- Date: Fri, 11 Apr 2008 16:03:08 +0100
- To: Julian Reschke <julian.reschke@gmx.de>
- CC: Ian Davis <lists@iandavis.com>, "Roy T. Fielding" <fielding@gbiv.com>, Jonathan Rees <jar@creativecommons.org>, Phil Archer <parcher@icra.org>, "www-tag@w3.org WG" <www-tag@w3.org>
Julian Reschke wrote: > Xiaoshu Wang wrote: >> In that case, I have explained that there is NO need for LINK >> whatsoever (by the principle of orthogonal specification). What is >> the point if you can describe those relation in "content" but you >> insist to put it in another place. This rational, in fact, breaks >> the "uniformness" that > > For once, it will work for all kinds of contents (independently of the > media type), and furthermore it will work always the same way. > >> ... >> HTTP is about deliver *message*. All HTTP-header's role should serve >> one function - that is how to get message and how to transform the >> *message* in *content*. >> ... > > Well, maybe it should, but that isn't what HTTP in fact does. HTTP has > had both request/response headers *and* entity headers for ages. > > BR, Julian I don't know what you are going because I can take your wording in either ways. My basic view point of the web is that its purpose is for communication. The basic client behavior can be summarized as simply as this. Give me the information about a *resource* in this *way* so I can understand the resource. Hence, the resource is denoted the URI and the *way* is specified via content negotiation (format and language etc.). Please summaries the desired user behavior of HTTP LINK? Xiaoshu
Received on Friday, 11 April 2008 15:03:55 UTC