- From: Ian Horrocks <horrocks@cs.man.ac.uk>
- Date: Mon, 15 Sep 2003 13:32:55 +0100
- To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
- Cc: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>, der@hplb.hpl.hp.com, www-webont-wg@w3.org
On September 9, Jeremy Carroll writes: > > > > Peter F. Patel-Schneider wrote: > > > > If a completely comprehensive test suite is being developed then it might > > be a good idea to have tests that demonstrate the presence or absence of > > the various comprehension inferences. I think there are many other tests > > that should be generated before these, however. > > > > > That may be true, but that is not where we are at. > > We do not have a complete test suite, and the coverage of the comprehension > axioms is entirely accidental. > > Because of Dave's imlementation style, these axioms are more pertinent for > him, and hence he probably has a better understanding of them than many > people who have implemented them. I don't see any reason to turn down his > offer of *more* tests which help make more apparant the current use of the > comprehension axioms. I agree with this, but a useful/appropriate test is one that highlights some inference/aspect of the semantics that will be incorrectly handled if certain comprehension axioms are missing. This seems to be the case already for some of the tests - "axiom based" implementations may fail these tests when they lack certain relevant axioms. I fail to see how adding a simpler version of the test that can be passed without said axiom(s) would help anyone. More tests that highlight other possible failings due to incomplete axiomatisations should, of course, be warmly welcomed. Ian > > i.e. the WG was not encoruaging me to generate more tests - I agree with > you that if we were having a systematic effort, these would not be the > first to produce. But given the offer of more tests that a least one > implementor has found useful we should take it. > > > >>>Possible responses to this comment include: > >>>1. Modify some of test cases to this simple-conclusion style. > >>>2. Augment the test cases by duplicates in this style. > >>>3. Ignore it and leave the test cases as is. > >>> > >>>Dave > >>> > > Jim: > > >>My preference would be 2 or 1 in that order - anything that makes it > >>easier for people to test implementations and to help them understnad > >>how to build tools seems like a good idea to me! > >> > Ian: > > I am opposed to 2 and strongly opposed to 1. > > > Jeremy: > support 2 > oppose 1 > > (I believe our testing of comprehension is weak, and would be reluctant to > see any of the current comprehension tests watered down). > > Jeremy > > >
Received on Monday, 15 September 2003 08:34:52 UTC