4702 2007-06-23 09:59:00 +0000 [FT] editorial: 2.3.1 Using Weights Within a Scored FTContainsExpr 2007-10-26 22:25:38 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Full Text 1.0 Last Call drafts All All CLOSED FIXED P2 minor --- 1 jmdyck pcase public-qt-comments oldest_to_newest 15530 0 jmdyck 2007-06-23 09:59:00 +0000 2.3.1 Using Weights Within a Scored FTContainsExpr [1] para 1 "Syntactically weight declarations" After "Syntactically", insert a comma. [2] para 2 "The effect of weights on the result score" s/result/resulting/ maybe [3] "However, weight declarations must follow these rules:" Actually, only point #3 (the restriction to 0 - 1000) is a rule for weight declarations to follow. The other two points are rules for the scoring algorithm to follow. [4] "1. Weights in an FTContainsExpr are significant only in relation to each other" [4a] Is it possible to detect if a scoring algorithm breaks this rule? [4b] When you say "in an FTContainsExpr", does that include nested FTContainsExprs? [4c] What should the scoring algorithm do with cases like this?: let score $s := $b/content ftcontains ("web site" weight 0.5) and $b/content ftcontains ("usability" weight 2) (Note two separate FTContainsExprs.) [5] "1. .... to each other; and" s/; and/./ (for consistency with other points) 16546 1 jim.melton 2007-09-13 23:02:25 +0000 As decided in meeting #152 (the minutes of which are at the member-only URI http://lists.w3.org/Archives/Member/member-query-fttf/2007Sep/0005.html), items [1], [2], and [5] have been resolved. That leaves items [3], [4a], [4b], and [4c] remaining to be resolved. 17232 2 pcase 2007-10-16 12:27:25 +0000 [3] The FTTF agreed. The text was reorganized as follows: The weight MUST have an absolute value between 0.0 and 1000.0 inclusive. The weights assigned are not related to any absolute standard, but typically have a relationship to other weights within the same FTContains expression. The effect of weights on the resulting score is implementation-dependent. However, scoring algorithms MUST conform to these constraints: 1. When no explicit weight is specified, the default weight is 1.0; and 2. Weight declarations in an FTContainsExpr for which no scores are evaluated are ignored. [4a]The FTTF agreed. We changed the sentence to: The weights assigned are not related to any absolute standard, but typically have a relationship to other weights within the same FTContains expression. [4b] No change. [4c] No change. These changes will appear in the next build of the internal Full-Text language after the October 11 build, and in the next public version. They close the last items in this bug. Michael, if you approve of the changes, please mark the bug closed.