Home

Comment on Bancilhon and Spyratos' "Update semantics and relational views"

Tue, 03/20/2007 - 13:15 — huliganka
Authors: 
Keller, A.M.
Author: 
Keller, A
Year: 
1987
Venue: 
TODS, Vol. 12, Issue 3 (Sept. 1987)
URL: 
http://portal.acm.org/citation.cfm?id=214296&dl=ACM&coll=portal&CFID=17432431&CFTOKEN=96665970
DOI: 
http://doi.acm.org/10.1145/27629.214296
Citations: 
11
Citations range: 
10 - 49
AttachmentSize
Keller1987CommentonBancilhonand.pdf171.4 KB

Bancilhon and Spyratos [l] show how the choice of a complementary view will
select a view update translator. They claim that “User requirements impose [the
constraint that] . . . (ii) The user must have the means to cancel, if he wishes,
the effect of every update that he is allowed on the view.” (p. 562) If this
requirement is eliminated, there are other reasonable translations that become
acceptable. We present a particular view update translator that is quite reasonable,
but that does not preserve any complement, and for which the requirement
does not hold.