review of Guide

I finally read Guide.  (It wasn't assigned to me.  This is for extra
credit.)

It's a nice overview of RIF.  I'm a little uncomfortable with the tone
for an official WG publication; it's rather conversational.  On the plus
side, that makes it easier to read.  I wouldn't object to publication
just because of the tone, but maybe we can make it a little more formal
in places.  (I haven't tried to figure out how, sorry.)

There are two points I think need to be changed before it can be
published, though:

   1.  It's not appropriate to name specific vendors (IBM and Oracle) in
       the bit about PRD.  They'll have a chance to join the press
       release when we get to Rec, and people can speak about it as
       individuals, but the Working Group shouldn't be seen to pick
       favorites like that.  Also, it could be read as IBM and/or Oracle
       endorsing RIF, and while I hope/expect they'll do that officially
       during Proposed Recommendation, it's not our place to speak about
       that yet.

   2.  I think Core is downplayed far too much.  The document says there
       are only "two main dialects", says (in section 2) that Core is a
       subset of BLD (not mentioning PRD), and most tellingly, Core's
       "*main purpose* is to enable limitted rule exchange between logic
       rule dialects and production rules."  It's quite possible Core
       will be far more popular than BLD or PRD, and it's important to
       me that we not speak negatively or weakly of it like that.

Minor points:

   - Let's take "main" out of the Abstract and go ahead and enumerate
     all the documents.  (That is, we need to add OWLRL and XML).  Okay,
     this isn't quite minor, but I could live with it like this, I
     guess, for now.

   - I'd add a little to the definition of 'extensibility'.  Perhaps:

         Extensibility here means that it should be possible for
	 motivated people to define a new RIF dialect as a syntactic
	 extension toan existing RIF dialect, with new elements
	 corresponding to desired additional functionality.  These new
	 RIF dialects would be non-standard when defined, but might
	 eventually become standards.

   - I'd take out the "by another group" reference, re future dialects
     based on FLD.  I think there's still some possibility we'll
     recharter RIF in such a way that the "new" group would still
     formally be this group.

   - "Development of the RIF framework turned out to be feasible"
                            ^ FLD?


Thats it.  :-)

    -- Sandro

Received on Friday, 25 September 2009 02:41:18 UTC