W3C home > Mailing lists > Public > public-owl-wg@w3.org > December 2007

Re: A stab at operationalizing Evan Wallace's suggestion for a new Overview

From: Jim Hendler <hendler@cs.rpi.edu>
Date: Sat, 8 Dec 2007 18:36:37 -0500
Message-Id: <B128948F-10B9-4BBB-8EFC-8C834DC53E85@cs.rpi.edu>
Cc: Rinke Hoekstra <hoekstra@uva.nl>, OWL Working Group WG <public-owl-wg@w3.org>
To: Bijan Parsia <bparsia@cs.man.ac.uk>
Bijan - with due respect, i wish I'd been at the f2f to make it clear  
how foolish this idea seems to me.  The reason is simple, IMO by  
mixing these things you are creating one document that will have too  
much in it - it will force people trying to simply see what is in the  
language to get boggled by a lot of terminology and syntax, it will  
make it too laden with examples for people who want a formal spec,  
etc.   I believe there is tremendous value in providing a structure  
that links the clear part from friendly idea to full spec, but not to  
putting it all together in a way that makes it hard to find the  
friendly idea when looking for the spec and vice versa.   So while I  
look forward to seeing what you produce, I remain extremely  
skeptical.  Toc course,  if we do a bad job with the documents, then  
the OWL 1.0 documents will continue to be used (and the new features  
of 1.1 only used by DL experts) so maybe I should stop objecting :-)

On Dec 8, 2007, at 9:36 AM, Bijan Parsia wrote:

> On Dec 7, 2007, at 4:36 PM, Rinke Hoekstra wrote:
> [snip Rinke's description of one of my key concerns with the  
> existing UFDs]
>>  Nonetheless the difference in the way in which the language is  
>> presented to the reader *is* probably relevant, something we could  
>> overcome by maintaining two orthogonal tables of contents.
>> Is there a particular reason why these should be separate real  
>> (i.e. monolithic, linear) documents, and not closely interlinked  
>> 'perspectives' on the same content?
> This is met by my proposal to enhance the structural specification  
> with:
> 	1) more modular/referencelike organization; slightly richer "less  
> formal" english descriptions, and examples,
> 	2) CSS tricks to allow hiding of information/alternative views/ 
> syntaxes, etc.
> 	3) appropriate indexes/interfaces for reference like navigation
> This is intended to replace the reference and that part of the  
> overview which acts as a terse reference/index. My hope is that  
> we'll have *one* definitive place where the description, canonical  
> example, and the formal (syntactic) definition all live (with a  
> link tothe formal semantics) so you have *one clear path* from  
> "friendly idea" to "full specification" for any bit of OWL, even if  
> you choose to explore things in other ways.
> I am tasked by the task force to produce a proof of concept of this  
> for a section of the struc spec. I'm also tasked to produce a proof  
> of concept draft of what I take as replacing the overview/guide  
> which I call the primer.
> Cheers,
> Bijan.

"If we knew what we were doing, it wouldn't be called research, would  
it?." - Albert Einstein

Prof James Hendler				http://www.cs.rpi.edu/~hendler
Tetherless World Constellation Chair
Computer Science Dept
Rensselaer Polytechnic Institute, Troy NY 12180
Received on Saturday, 8 December 2007 23:37:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:42:01 UTC