Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!rutgers!ucla-cs!zen!postgres!larry
From: larry@postgres.uucp (Larry Rowe)
Newsgroups: comp.databases
Subject: Re:  Degree 3 consistency
Message-ID: <3590@zen.berkeley.edu>
Date: Fri, 4-Sep-87 13:40:04 EDT
Article-I.D.: zen.3590
Posted: Fri Sep  4 13:40:04 1987
Date-Received: Sat, 5-Sep-87 18:34:55 EDT
References: <1215@smokey.UUCP> <18182UH2@PSUVM> <1301@smokey.UUCP> <3391@zen.berkeley.edu> <1381@smokey.UUCP>
Sender: news@zen.berkeley.edu
Reply-To: larry@postgres.UUCP (Larry Rowe)
Organization: Postgres Research Group, UC Berkeley
Lines: 10

In article <1381@smokey.UUCP> garyp@cognos.UUCP (Gary Puckering) writes:
>Does this mean that degree-3 consistency can be obtained in IMS?  If
>so, what method is used to eliminate "phantoms".  Most relational
>systems use index or table locking for this.  What does IMS do?

i'm not sure since i've never used ims.  i have some contacts at
ibm that know more about this, i'll ask them and communicate the
answer.  if i had to hazard a guess, i'd say that ims doesn't handle
phantoms.
	larry