RE: A Rough Guide to Notation3

>I think the only negative is the lack of a comprehensive spec
>(altho Sean has taken us in the right direction).

Perhaps. But if there were a truly comprehensive spec for n3 and by
extension RDF, would there still be the need to noodle (to coin a TimBLism)
manually with the data?
Either way I agree about the value of Sean's piece.

>A non-XML syntax is easier to learn and use.  Witness the
>popularity of RELAX NG's compact syntax.

I've not used RELAX NG myself, and glancing at the doc you quote -

## Represents a language
element lang {
  ## English
  "en" |
  ## Japanese
  "jp"
}

I can do no more than guess at the meaning. On the other hand, I can be
pretty sure I get the meaning of this -

<element name="lang"
    xmlns:a="http://relaxng.org/ns/compatibility/annotations/1.0"
    xmlns="http://relaxng.org/ns/structure/1.0">
  <a:documentation>Represents a language</a:documentation>
  <choice>
    <value>en</value>
    <a:documentation>English</a:documentation>
    <value>jp</value>
    <a:documentation>Japanese</a:documentation>
  </choice>
</element>

A non-XML syntax is easier to learn and use only if it's at least as
sensibly designed as the corresponding XML syntax *and* the learner is new
to XML. The former isn't saying much, but the latter isn't insignificant
these days.

I'd love to see (or help
>develop) a spec for n3 similar to James
>Clark's RNC tutorial [1].  Does anything like this exist?  Perhaps
>I've missed it.

That would be great - perhaps then I might feel more favourable about n3 ;-)

>n3 continues to feel very much like a "hack".  (Is that because it is?)

I defer to Sean's doc on that point...

Cheers,
Danny.

Received on Friday, 23 August 2002 18:48:00 UTC