This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
3.3.5 Stop Word Option [1] '[173] FTRefOrList ::= ...' The name "RefOrList" is fairly generic. Perhaps "FTStopWords"? [2] '[174] FTInclExclStringLiteral ::= ("union" | "except") FTRefOrList' The "StringLiteral" in the name doesn't really make sense. Perhaps "FTStopwordInclExcl" ? [3] para 2 "they are used as they occur in the sequence." There's no antecedent for "the sequence". [4] Note "If applied during indexing asking for stop words to not be used during a query, will have no effect." [4a] After "indexing", insert comma. [4b] Except for the fact that it's a Note, this appears to be a license for a Full-Text processor to behave non-conformantly. Surely if an implementation does some pre-emptive computation assuming certain options, and a query uses conflicting options, then the implementation is obliged to ignore the prepared results and do what it takes to give the correct answer. [5] examples I think it would be instructive if you had an example like ... p ftcontains "supports users" with stop words ("the") As I understand it, the result is false (despite the fact that the text contains "supports the users"), which might not be what people expect. [6] 'contains the "then"' Change to 'contains "the" and "then"' ? [7] 'except ("the then")' Are you sure? Supposedly, no tokenization is performed on "the then". I think you mean 'except ("the", "then")', or else you need to clarify how string literals are handled here.
Fixed as decided in FTTF-156. This just leaves 4b open. See http://lists.w3.org/Archives/Member/member-query-fttf/2007Oct/0022.html (member-only link)
[4b] No change. FTTF believes that the note is accurate. Michael, if you agree we are OK with no change, please close this bug.
(In reply to comment #2) > [4b] No change. FTTF believes that the note is accurate. I don't think that answers my concern. Do you agree and intend that it's a license for a Full-Text processor to behave non-conformantly? Is a processor required to answer correctly or not? Is this the kind of Note that's a not-immediately-obvious consequence of some normative statements? If so, which ones?
On additional consideration, WG agreed to clarify the situation, see comment: http://lists.w3.org/Archives/Member/member-query-fttf/2008Mar/0028.html
At meeting 169, the FTTF approved the proposed wording with minor amendments. Mary has committed the changes to the document. I agree that the new wording resolves [4b], which was the only unresolved part of this issue. Therefore I will close this issue.