RE: feedback on http://www.w3.org/2005/rules/wiki/Primer

Thanks much, Eric. I used most of your suggestions.
Please have another look. Harold


-----Original Message-----
From: public-rif-comments-request@w3.org [mailto:public-rif-comments-request@w3.org] On Behalf Of Eric Prud'hommeaux
Sent: July 26, 2010 5:52 PM
To: public-rif-comments@w3.org
Subject: feedback on http://www.w3.org/2005/rules/wiki/Primer

First, thanks a million for the RIF Primer. Now some unsolicited
editorial offerings indicating my attentiveness:

Here are some wiki-syntax (i.e. paste-able) patches to address
"Editor's Note: Add references, links.":

http://www.w3.org/2005/rules/wiki/Primer#Overview:__What_is_a_rule.3F_What_is_RIF.3F ¶13:

The [http://www.w3.org/TR/rif-overview/ W3C Rule Interchange Format (RIF)] is a standard that was developed to facilitate ruleset integration and synthesis. It comprises a set of interconnected dialects representing rule languages with various features. This document focuses on [http://www.w3.org/TR/rif-bld BLD], the Basic Logic Dialect, augmented by [http://www.w3.org/TR/rif-dtb DTB], a set of datatypes and built-in functions and predicates that can be used when writing rules.


http://www.w3.org/2005/rules/wiki/Primer#Conjunctions_and_Implications ¶3:
[[
In RIF Convenience Syntax (CS), which is used throughout this
document, as in RIF Presentation Syntax (PS), .
]]
I don't understand "In RIF Convenience Syntax (CS), which is used
throughout this document, as in RIF Presentation Syntax (PS), .".
I assume it's supposed to say that CS is another syntax, as is PS.

As a consumer of the document, I'd rather not learn a syntax which is
useful only in this document. I don't love PS, but 11 RIF docs make
it a reality. Also, visually scanning for the prology ":-" is more
reliable than visually balancing the last ()s in the rule.


Putting "'''"s around variables in rule examples seems to improve
readability.


http://www.w3.org/2005/rules/wiki/Primer#Quantifiers

I played with the following markup and it seemed to improve
scanability (or playing with it made it more familiar to me):
[[
''If IMDB contains the facts that '''Vivien Leigh''' played the role of '''Blanche Dubois''', and that '''Blanche Dubois''' is a character role in '''A Streetcar Named Desire''', then DBpedia contains the fact that '''Vivien Leigh''' acted in '''A Streetcar Named Desire'''.''

This rule could be represented as

 If And(playsrole('''vivienleigh''' '''blanchedubois''')
        roleinfilm('''blanchedubois''' '''streetcar''')) 
    actsinfilm('''vivienleigh''' '''streetcar''')
]]
You can paste it in and see if the preview appeals to you.


http://www.w3.org/2005/rules/wiki/Primer#Reasoning_in_RIF ¶4 (the bullet list):
Is this a useful summary of Modus Ponens?
s/Modus Ponens/simple deduction -- A + if A then B => B (Modus Ponens)/


http://www.w3.org/2005/rules/wiki/Primer#Example_Using_Frames ¶3
s/slots into RIF\n\nthus facilitates/slots into RIF thus facilitates/


http://www.w3.org/2005/rules/wiki/Primer#Distinguishing_Slots_in_Object-Oriented_Languages_and_RIF
s{The slot e1.a is first assigned 1, but that value is then replaced by 2;\nIn contrast, consider the following in RIF:}
 {The singleton slot e1.a is first assigned 1, but that value is then replaced by 2. In contrast, RIF frames can be considered multimaps, supporting multiple assignments to the same attribute:}

I expect that multimap explanation may appeal to someone somewhere
under some conditions we've not yet discovered.


http://www.w3.org/2005/rules/wiki/Primer#Functions.2C_Equality
I'm hungry for details about '='; but haven't formed good questions yet.
-- 
-ericP

Received on Monday, 26 July 2010 23:25:07 UTC