W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > July to September 1996

RE: Seiwald Q & A -- "GET for EDIT" cookies

From: Yaron Goland <yarong@microsoft.com>
Date: Sat, 31 Aug 1996 17:01:34 -0700
Message-ID: <c=US%a=_%p=msft%l=RED-44-MSG-960901000134Z-22526@tide21.microsoft.com>
To: "'w3c-dist-auth@w3.org'" <w3c-dist-auth@w3.org>, "'www-vers-wg@ics.uci.edu'" <www-vers-wg@ics.uci.edu>, "'Christopher Seiwald'" <seiwald@perforce.com>
Some systems may want to use cookies, others may want to just do a redirect 
with a URL extension. However my concern isn't that tokens may be needed 
but rather that we are over specifying. We need to be careful about not 
throwing in the kitchen sink. In this case there are clearly systems which 
may need token support but they can use the already existing cookie 
standards to handle this. As such there is no need to explicitly mention 
said support in the versioning standard. Small is beautiful when it comes 
to standards.

From:  Christopher Seiwald[SMTP:seiwald@perforce.com]
Sent:  Friday, August 30, 1996 5:42 PM
To:  w3c-dist-auth@w3.org; www-vers-wg@ics.uci.edu
Subject:  RE: Seiwald Q & A -- "GET for EDIT" cookies

| From: Yaron Goland <yarong@microsoft.com>
| Sure you do a check out on the document and the system records on the
| Server the existence of the check out. Then you do a check in on the
| document and the server knows who has the document checked out. Either 

How does the server connect your checkin with the original checkout, so
that it can clear the checkout state?  It must use some information - some
token or set of tokens to bridge the two.

Clearly the URL is not sufficient: multiple users can be working on the
same document.

URL+username might be better, but I can easily imagine a case where the
same named user (or the same user) is working on a single document from
two different places.

URL+username+hostname?  Maybe.  I can think of times, tho, when I
myself have been editing the same source file on the same machine
in two different places (and fully expecting the SCM system to keep
track of it).

I don't think we can define what information is sufficient for all
version-aware Web servers to be able to join a checkin with its
corresponding checkout.  At a certain point, you're going to have to
concede that it has to be a token of the server's choice.

Received on Saturday, 31 August 1996 20:01:59 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:01:08 UTC