Re: Seiwald Q & A -- "GET for EDIT" cookies
Roy T. Fielding (fielding@liege.ICS.UCI.EDU)
Tue, 03 Sep 1996 15:24:04 -0700
> 1. What about Content-Version and Derived-From? (Dan Connolly)
>
> If I read the spec right, Content-Version reflects the contents
> of the document. That is, if the same document is dished up
> twice it is supposed to have the same Content-Version value.
It reflects the contents of the Entity, which includes both the body
document and the entity-header fields (metainformation about the document).
If the server is providing different information per checkout, then it
is in fact changing the entity.
> As I argued before (and will continue arguing until I wear people
> down :-) the identity of the source is not sufficient information
> for a "checkin", because the VC system underneath the version-aware
> web server may wish to find any context associated with a prior
> "checkout".
>
> Now Roy Fielding says that Content-Version is opaque and could
> be used exactly for this purpose, 'cause no one would be the wiser
> if the Content-Version were different for each checkout of the
> same document. This is true, but now the names of these fields
> are losing their meaning, no? If it's checkout context, call it
> "Checkout-Context" (or "Checkout-Cookie").
Checkout does not have meaning on all systems, whereas version has a generic
meaning (at times, too generic). It is intended to be as flexible as possible.
The "Content-" prefix is a requirement of MIME for what HTTP calls
entity-header field names.
.....Roy
BTW, I can set the www-vers-wg@ics.uci.edu list Reply-To at any time --
the reason it is not on by default is that many people find it annoying,
particularly when you have cross-list discussions (like this one).