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

Re: POST vs. separate methods

From: Larry Masinter <masinter@parc.xerox.com>
Date: Tue, 29 Oct 1996 14:03:00 PST
To: gjw@wnetc.com
CC: ejw@kleber.ics.uci.edu, w3c-dist-auth@w3.org
Message-Id: <96Oct29.140300pst."2763"@golden.parc.xerox.com>
> The difficulty here is tha the processes involved my be unrelated, even
> unaware of eachothers' existence. If they reside on separate hosts they may
> follow different proxy chains simply because of network topology and no
> switching of proxies involved.

I'm sorry, I was being really terse when I talked about "sequential
transparency". What I meant was that if client A does a "GET x; UPDATE
y; GET x" that the second "GET x" properly reflects the changes made
by "UPDATE y"; however, if client B does "GET x", even at a time AFTER
the "UPDATE y" happened, client B might see the original data.

If what you want is "sequential transparency" then you don't need to
deal with totally unrelated processes, you only have to deal with a
single process (client A) which potentially might switch between one
proxy and and another.


Received on Tuesday, 29 October 1996 18:53:58 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:01:09 UTC