RE: XML syntax for SWRL

From: Pat Hayes (
Date: 06/24/04

  • Next message: Peter F. Patel-Schneider: "Re: XML syntax for SWRL"
    >This is no big deal. E.g.:
    ><var>the net income per year</var>     could have been
    ><var name="the net income per year"/>
    >But when designing RuleML, we wanted to minimize the use of XML attributes.
    Can you tell me WHY you wanted to so minimize this use? I ask because 
    there seem to be strongly held opinions about how 'best' to write 
    XML, including using no attributes, using attributes as far as 
    possible, using attributes with URLs to maximize future 
    extendability, using striped syntax, most emphatically not using 
    striped syntax, etc, etc.. None of this strikes me as wholly 
    rational, frankly: it sounds like arguing over what is the proper 
    color to paint a kitchen. But if y'all can give a good objective 
    reason why you decided to adopt this style rule in RuleML, or if you 
    can relate it to a clearly stateable goal or design criterion, I'd 
    very much like to hear about it.
    >Since <> we have uniformly used the PCDATA
    >content, rather than CDATA attribute values, as names of ind(ividual)s,
    >var(iable)s, etc. This has also left us the option of embedded XHTML etc.
    >-- Harold
    >-----Original Message-----
    >[]On Behalf Of Peter F.
    >Sent: June 23, 2004 2:29 PM
    >To: Boley, Harold
    >Subject: Re: XML syntax for SWRL
    >A question about RuleML.
    >Why doesn't ruleml.var take the variable name as an attribute?  This seems
    >to be a very good place to use an attribute.
    IHMC	(850)434 8903 or (650)494 3973   home
    40 South Alcaniz St.	(850)202 4416   office
    Pensacola			(850)202 4440   fax
    FL 32501			(850)291 0667    cell

    This archive was generated by hypermail 2.1.4 : 06/24/04 EST