W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1996

Re: REPOST (was: HTTP working group status & issues)

From: Gregory J. Woodhouse <gjw@wnetc.com>
Date: Fri, 27 Sep 1996 15:06:13 -0700 (PDT)
To: Larry Masinter <masinter@parc.xerox.com>
Cc: mau@beatles.cselt.stet.it, MACRIDES@sci.wfbr.edu, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <Pine.SGI.3.95.960927145452.9221B-100000@shellx.best.com>
Personally, I like the idea of a reliable POST. Instead of introducing a
REPOST method for situations where the original POST may or may not have
succeeded, it would seem better to include the necessary handshaking so that
the client will no for sure if the operation succeeded, and so tht if it
fails, partial or incorrect data will not be recorded by the server. This
is more complex, of course requiring several interactions (post, confirm,
commit, acknowledge), but I believe it is worthwhile in cases where forms are
used to update databases and the like. Of course, a response header which
indicates that a REPOST is safe would be enough for many situations, and
developers should be able to use the unsafe POST when it is appropriate to
do so.

---
Gregory Woodhouse     gjw@wnetc.com
home page:            http://www.wnetc.com/
resource page:        http://www.wnetc.com/resource/
Received on Friday, 27 September 1996 15:43:02 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:32:14 EDT