- From: Dimitris Dimitriadis <dimitris.dimitriadis@improve.se>
- Date: Thu, 29 Mar 2001 13:32:23 +0200
- To: "'Arnold, Curt'" <Curt.Arnold@hyprotech.com>, "'www-dom-ts@w3.org'" <www-dom-ts@w3.org>
Very good point. We do want to model dependencies, without implying difference in relevance. If for example 3 breaks if 2 breaks if 1 breaks and 1 is a "central" feature (defined in core) whereas 3 is style, this by no means implies that 3 is less important than 1. /Dimitris -----Ursprungligt meddelande----- Från: Arnold, Curt [mailto:Curt.Arnold@hyprotech.com] Skickat: den 29 mars 2001 01:56 Till: 'www-dom-ts@w3.org' Ämne: Test dependencies I've been walking through the Hacked NIST conformance suite today making sure that the most significant items get into the Adobe SVG bug database which brought the following thought to mind: it would be useful if the test definitions could express any known dependencies between tests. For example, if test B, C, D, E, F, G, H, etc will always fail if test A fails, having that info in the test description will make it easier for a developer to focus in on the problem. Another thing that we will have to address is conformant variations, the most obvious instance is whether entity references are expanded.
Received on Thursday, 29 March 2001 06:36:53 UTC